Training Coach
Doctrine
A report may cite any of these as doctrine-N.
doctrine-1Contract freeze and fingerprint disciplineThe policy I/O contract (observation layout, scales, history semantics, action pipeline) is frozen and fingerprinted; every exported policy is stamped and verified; contract changes ship as new versioned profiles that leave old artifacts bit-identical, and old policies run forever under their era's pinned profile.
Case. The 215-dim omni contract was frozen with a three-machine digest; the one contract-level extension (lateral feed-forward) went in as a new `omni_ff` profile with the old profile provably untouched, and the contract checker caught two real wiring bugs before any training (`contract-freeze-and-checker`). A silently changed gait-clock default would have fed old policies a 25% slower clock - closed by pinned legacy profiles (`legacy-profile-pinning`). A stale derived USD forked plant mass 2.2% until an automated source-vs-derived instrument gated it (`derived-asset-staleness-check`). A gain profile is part of the closed loop a policy was trained in and belongs in its stamp; the recovery line's anchored authority was left out of its manifest and recorded as the gap not to repeat (`gain-profile-belongs-in-the-stamp`), and a second policy behind a deploy-side switch made the handoff state itself a contract (`recovery-two-policies-and-a-state-machine`, `walk-recovery-fsm-handoff`).
Coach application. On any proposal touching obs/action semantics, defaults, or derived assets: demand the version/profile plan, the fingerprint update, and the checker extension in the same change; flag any old artifact that would run under new defaults.
doctrine-2Attribution by resolved training params - never eval-override knobsCapability differences between lineages are explained only by digging each lineage's *resolved* training configuration and eliminating columns; evaluation-side override knobs (kd-scale, power-scale, cycle-time) act on the plant for *every* policy and may serve as deployment mitigations but never as explanations.
Case. Low-friction robustness across 8 lineages x 3840 cells was traced to kd DR *bandwidth* - every lineage had ground friction pinned to (1.0,1.0), so "trained friction" could not be the axis; the parameter axis and the plant axis were explicitly separated after the first attribution conflated them (`kd-bandwidth-mu-law-attribution`). "Weak turning" on hardware was a power-scale plant effect, not a training gap (`deploy-knob-attribution-before-retraining`); slowing the deploy clock was out-of-distribution, not a feature (`cycle-time-override-is-ood`). The ground truth for what a run trained under is the logged per-run config, not the source tree (`resolved-config-is-source-of-truth`).
Coach application. Whenever asked "why is lineage A better", require the resolved-param table first; kill zero-variance columns; refuse explanations phrased in eval-knob terms; when a knob helps, label it deployment mitigation.
doctrine-3PASS gates become constraints; FAIL gates become objectivesOnce a skill passes its gate, that gate converts into a standing regression constraint (budget <= 2/20 against the parent baseline) for all later training; gates currently failing are the only legitimate objectives of the next rung.
Case. The C ladder ran one frozen 13-cell x 20-seed matrix at every rung with promotion = "new skill PASS and old skills within regression budget"; C1 was stopped and re-rooted precisely because it trained away the root's backward PASS (`fixed-acceptance-matrix-per-rung`, `preregistered-stop-criteria-per-rung`). The C4 product shipped only at 260/260 cells with zero regression.
Coach application. Keep the ledger: every PASS adds a constraint row; propose rungs only against FAIL rows; treat any constraint violation as stop-and-attribute, never "the next rung might win it back".
doctrine-4One variable per ladder rung - counted against what the checkpoint sawA rung changes one variable, where "one" is counted against the checkpoint's actual training state, not against the current config's diff; batching is allowed only when each change owns a disjoint symptom space with a pre-registered ablation order.
Case. Two rungs failed identically because resuming s1e-500 under the evolved config silently added four plant variables the checkpoint had never seen ("单变量纪律不只看「我改了什么」,还要看「checkpoint 见过什么」" - `resume-state-dr-audit`). v8 legally batched four orthogonal fixes with a written ablation order (`orthogonal-batch-with-ablation-order`); v9 spent one run completing a 2x2 factorial so either outcome convicted a factor (`fill-the-missing-factorial-cell`); v10b's three-way ablation wrongfully convicted the clock and had to be retried fairly.
Coach application. Before any resume: diff cfg against the checkpoint's logged training state. Before any batch: require the symptom-ownership map and ablation order in writing.
doctrine-5Pre-register risks, readings, and stop criteria before the ladderBefore a ladder or risky rung, write down the known risks, the interpretation of every plausible outcome, and hit-any-one stop criteria - frozen before training, tightened when priors say results should come fast.
Case. The C ladder opened with three numbered risks including the exact falsification condition for its own root choice; A/B arms carried "预注册读法(事后不改)" tables; a level expected to fail was run anyway for its pre-registered diagnostic value (`preregister-risks-and-fork-readings`). Stop criteria caught C4-redo rungs at +200 instead of full caps (`preregistered-stop-criteria-per-rung`); hardware sessions pre-registered per-config expected signatures and the disagreement rule "不改结论改账" (`preregistered-real-expectations`, `feasibility-accounts-lock-design-point`).
Coach application. Refuse to open a rung without the written risk/reading/ stop block; after results, read conclusions off the pre-registered table and flag any post-hoc reinterpretation.
doctrine-6Plant parameters are measured, never inventedEvery plant number carries measurement provenance: armature = N^2 x rotor inertia from no-load tests, friction split by rig and by API column, torque limits shaped by per-joint gait peaks, latency traced through the real pipeline, masses weighed - and DR bands are additive around the measured nominal, sized to the measured dispersion.
Case. Guessed friction was 2.5x low and guessed damping 5x high (`friction-measured-not-guessed`); armature had been 0 with a 9:1 gearbox (81x reflected inertia, `armature-n2-rotor-inertia`); a uniform torque derating was "the wrong shape" vs measured peaks (`torque-limit-shape-by-measured-peaks`); the delay implementation itself was a wrong plant for a whole lineage (`latency-lerp-reverse-extrapolation`); the run design point was locked by three accounts including the tau_limit/kd speed ceiling (`feasibility-accounts-lock-design-point`); identified friction had to land in the right simulator API columns to act at all (`sim-api-friction-columns`). The recovery and one-leg lines opened with the same kind of accounts before any reward existed - a connected static path and the torque along it for an armless get-up, and the gains single support needs to be holdable at all (`get-up-feasibility-accounts-before-training`, `single-support-gain-authority-probe`).
Coach application. For any plant value in a config review, ask "measured how?"; reject absolute ranges with no nominal; check API column mapping and derived-asset regeneration whenever measured values land.
doctrine-7Sim2sim gate before sim2real - under deployment conditionsEvery checkpoint passes a second, independently built simulator before hardware, and both the gate and the smoke loop run under the measured deployment conditions (real pipeline delay, honest contact parameters, the deployment gain/power profile).
Case. The standing order "先sim2sim 再sim2real" (`sim2sim-gate-before-sim2real`); acceptance flipped to match hardware only under measured condim/torsional friction (`eval-plant-honesty-contact-params`); gates moved permanently to `--delay 2` after the kicking incident (`pipeline-latency-is-plant-not-dr`); and the harness itself must be audited - a frame-convention bug in the cross-sim evaluator invalidated a whole line of verdicts (`body-frame-velocity-api-audit`). The recovery line's second simulator caught a torque penalty paid for by bracing the legs together (`torque-penalty-bought-by-leg-bracing`), and a 1.8x torque disagreement between the two plants stayed binding because its one surviving explanation was never tested (`torque-disagreement-between-simulators-unresolved`).
Coach application. Block any hardware request lacking a second-sim PASS at deployment conditions; when sim2sim and training-side metrics disagree, treat the evaluator as a suspect too.
doctrine-8Observation honesty - the actor's inputs are a hardware contractThe actor observes only signals the real robot produces with realistic noise; privileged truths go to the critic; history windows are estimators and must train under plant variation; rewards on quantities the actor cannot observe buy only average suppression, never closed-loop correction.
Case. Ground-truth velocity/forces went critic-only (`observation-honesty-critic-only`); frame_hist under zero DR memorized the trainer's plant fingerprint - 0/20 transfer (`history-obs-needs-plant-variation`); world-frame yaw rewards could not teach pull-back because heading is unobservable to the actor - correction was routed to the deploy outer loop instead of breaking the contract (`reward-observability-limit`, `deploy-heading-loop-and-align-training`).
Coach application. Audit every actor-obs element for hardware existence; require minimal plant jitter whenever history/recurrence exists; for each reward, ask "can the actor see this error?" and route correction tasks to outer loops.
doctrine-9Reward economics are audited in realized currencyReward design decisions are made on realized per-step magnitudes under the actual policy and command distribution: price the do-nothing optimum before adding a mode, compare achieved values to the computed ignore-floor, calibrate thresholds between measured healthy and sick distributions, and ship every new penalty with a withdrawal clause.
Case. feet_air_time at weight 2.0 realized 0.038 vs tracking 1.2 - drag was rational (`realized-contribution-audit`); ignoring a vy command cost 28-180x less than ignoring vx until a gated tracking term was added (`reward-cost-of-ignoring-audit`, `gate-new-reward-terms-by-command`); achieved-vs-floor separated "never learned" from "priced out" (`ignore-floor-diagnosis`); the foot-distance wall was placed between measured healthy (0.6% tax) and sick (55%) policies (`calibrate-threshold-between-healthy-and-sick`); the landing penalty carried a pre-registered stand-down condition and actually stood down (`calibration-threshold-with-withdrawal-clause`); two clearance terms were inert until zero-points and gate occupancy were checked (`inert-reward-term-audit`). A get-up policy sat because three gated terms paid the seated pose 84% of the return and the one term that could tell sitting from standing was an exp kernel reading 4.6e-5 at the real error (`seated-basin-dead-exp-kernel`); a torque-tail term was weighted by its measured steady value beside a peer term after the estimate proved 12x off (`tail-torque-needs-hinge-on-computed-demand`).
Coach application. Never discuss weights in the abstract: demand the realized-contribution table, the ignore-floor number, and the healthy-pay calibration before any reward edit is approved.
doctrine-10The zero-cost option must be the desired behaviorFor every penalty, name what the zero-cost option is; penalize failure events (slip, saturation excess, contact in flight windows), never the motion or joints that healthy behavior uses; make degenerate strategies fatal via termination where penalties cannot price them out.
Case. Joint-usage penalties for drift taxed a 1.4%-of-momentum channel 2.7/step and collapsed training; the slip penalty costs a non-slipping gait exactly zero (`penalize-the-slip-not-the-joint`). A frozen-at-clamp joint pays zero action-rate forever - only a pre-clip saturation penalty flips the cheat economics (`saturation-cheating-zero-rate-cost`). Ungated phase shaping made standing 42x more expensive than stepping and cooked the hip motors (`moving-gate-42x-stand-tax`); crouch-shuffling lived until a height termination deleted it (`termination-closes-degenerate-basin`). A gated penalty is an exit: the policy parked just outside an uprightness gate, then just under a height gate, to stop paying a stance tax, and only a positive band plus an always-on guard closed both (`penalty-gate-is-an-escape-hatch`); a soft-limit penalty that charged the standing pose itself bought a 4.1 deg lean (`soft-limit-penalty-charges-nominal-pose`); an unpriced foot attitude was spent on edge-standing (`unpriced-foot-attitude-is-a-free-variable`); and the one-leg line listed its cheapest cheats before training and still met one through a zero-gradient band (`enumerate-cheapest-cheats-before-training`, `binary-band-reward-fake-touchdown`).
Coach application. Run the "零代价的选项是什么" audit on every proposed term; convert motion taxes into event-conditional penalties; check the termination set against each known degenerate strategy.
doctrine-11Measurement discipline: independent referees, signs, distributionsA disputed measurement is adjudicated only by an independent algorithm from raw state; directional ability requires sign-antisymmetry under command reversal; bimodal metrics are reported as mode shares (never medians, never 3 seeds); ratios are not comparable when totals change; reward values compare only within one command distribution; single chaotic events never cross machines.
Case. The triple reversal - a good metric was "refuted" by a sibling metric that shared the disease (`independent-referee-for-metric-disputes`, `body-frame-velocity-api-audit`); same-signed +/- responses were bias, not turning (`same-sign-response-is-yaw-bias`); the swing median sat in a bimodal gap (`median-hides-bimodal-distribution`); "v6 is jitterier" died on absolute energies (`ratio-metrics-need-absolute-check`); yaw gain measured 15x wrong in an oscillating frame (`heading-integral-not-body-rate`); a 44% improvement evaporated under same-distribution comparison (`same-distribution-reward-comparison`); drift direction was a limit cycle (`multiseed-sign-test-for-drift`); a cross-machine push cliff was chaos (`single-impulse-recovery-is-chaotic`).
Coach application. Before accepting any surprising number: ask for the independent recomputation, the sign pair, the distribution shape, and the comparison conditions. Retract in writing when a metric falls.
doctrine-12The deployment pipeline is plantIrreducible pipeline properties - action latency, rate limits, power/torque scaling, teleop command mappings - are part of the nominal plant, modeled from day one and reproduced in every gate; deploy-side scalings are crutches that flag unmodeled plant, and they cannot be algebraically folded into training constants.
Case. Right-leg kicking was over-trained-delay x loop gain; power 0.8 was a gain-reduction crutch that retired when the delay was modeled (`pipeline-latency-is-plant-not-dr`); power derating damages non-forward axes first (`power-scale-hurts-nonforward-axes`); training at 0.4 scale as the "twin" of deploying 0.5 x 0.8 collapsed 0/20 (`deploy-scaling-not-training-equivalent`); one shared teleop speed sent an out-of-band lateral command and the robot clipped its own foot (`teleop-command-band-per-axis`); the latency DR range had not even covered the measured pipeline (`latency-dr-covers-measured-pipeline`). A rate limiter added at deployment only clipped a policy that kept commanding (`deploy-rate-limiter-windup`); moved into training and anchored on the last command it became an integrator in the balance loop (`slew-anchor-is-an-integrator`); anchored on the measured angle it bounded torque and kept the bandwidth (`beta-anchored-action-target`). The walking lines' safe setting, power-scale 0.8, cut the ends of the recovery policy's full-range travel and left its spikes alone; a gain inside the trained band did the job (`power-derating-cuts-full-range-contract`).
Coach application. Demand the measured pipeline latency/limits in the plant model and in gate conditions; treat every deploy-side derating as a question ("what is this compensating?"); block per-axis command sources that exceed training bands.
doctrine-13DR budget is finite; its distribution is the measured supportRobustness is a conserved budget: disturbance training on an already-hardened lineage borrows from existing margins; DR ranges span the measured deployment support - no fictitious tails (they buy degenerate gaits), no single constants (they allow thin-margin specialization); harden the plant only after the task distribution is final.
Case. The same push dose helped a narrow lineage and damaged a balanced one - budget conservation (`push-dr-conditional-budget-conservation`); wide latency tails bought drag-glide, constant values shipped 60% thinner tilt margins - the answer is a narrow band on the measured support (`dr-tail-plant-continuation`, `constant-value-dr-overfits-margin`); task-first ordering because hardening a soon-to-change task wastes budget (`task-shaping-before-plant-hardening`); COM randomization used deliberately as a behavior-shaping tool, and rolled back on symptom per its own contract (`com-randomization-forces-leg-spread`, `com-dr-rollback-on-symptom`). DR that is switched on can still be thin: the run policy fell in the frontal plane its gain-and-latency randomization never touched (`thin-dr-judged-by-channel-coverage`), and a friction priority settled under one action contract had to be re-measured under the next (`friction-priority-re-measured-after-plant-change`).
Coach application. Before any DR rung: check the untrained policy against the spec, the lineage's current DR load, and the measured real-world range; after it: audit retained margins, not just the new tolerance.
doctrine-14Gates measure what hardware feels: posture, margins, stripped assistsAcceptance batteries carry posture-class rows (tilt max median, per-joint L/R asymmetry, temperature) beside task rows, graded margin columns beside binary gates, chirality scored per side, at least one condition that removes the environment's free stabilization, and validated predictive scalars promoted into the gate.
Case. Three same-shaped judging errors - survival, displacement, wz-difference - all missed what the operator felt; posture metrics had the predictive power (`task-metrics-vs-posture-metrics`, `stand-gate-posture-not-survival`); binary survival saturated and hid a 60% margin gap (`constant-value-dr-overfits-margin`); v5 passed everything on the ground and failed suspended (`suspension-probe-removes-free-stabilizer`); the hip_roll (l+r) scalar predicted real drift direction and ordering and entered the battery (`hip-roll-sum-predicts-lateral-drift`); averages hide chirality (`chirality-scored-separately`); gait-quality gates are judged at speeds that demand a gait (`low-speed-commands-reward-dragging`). The recovery line added the rest of the kit: where failed episodes end, not only where they started (`end-state-confusion-matrix`); a frozen acceptance distribution with a pinned seed (`frozen-acceptance-distribution-and-pinned-seed`); video of the metric rollout itself (`video-as-acceptance-record`); and the admission that a 10 s episode cannot see a stance that fails after a minute (`episode-length-bounds-what-a-gate-sees`). The one-leg line removed a foot-spacing wall that no gate measured, and the feet met on hardware (`removed-wall-returns-on-hardware`).
Coach application. Review every battery for posture rows, margin columns, per-side scoring, and an assist-stripped condition; when operator feel and gates disagree, suspect the metric class first.
doctrine-15Fork and root selection: recoverability, maturity, frozen rewardsChoose fork roots by which candidate's deficits the coming training can pay back (precision is recoverable; lost plasticity, symmetry, and margins are not); prefer mature checkpoints as roots even when younger ones score better as products; never fine-tune through a reward change - continuation is legal only with the reward frozen and plant/DR widening one rung at a time.
Case. s1e-500 beat higher-precision candidates because its exclusive strengths were unrecoverable (`fork-root-recoverable-shortfall`); the b300 arm proved maturity is capital against adaptation shock (`root-maturity-vs-product-quality`); the B-arm scatter/half-recover/collapse signature falsified reward-change fine-tuning and drew the legal boundary for S2 continuation (`fine-tune-reward-change-falsified`).
Coach application. For root debates, build the exclusive-strengths table and ask "which side can be trained back?"; require dual-arm evidence for maturity claims; classify any proposed continuation as reward-frozen or not before approving.
doctrine-16Curricula: verified engagement, lineage counters, disease-phase gatingAutomatic curricula must prove they engage (a saturated ratchet is constant DR wearing a curriculum's name); every ramp counts lineage-cumulative progress, not per-process steps; penalties aimed at late-stage pathologies ramp in after exploration noise decays; difficulty rises on measured per-stratum success, never on schedule.
Case. The s1f ratchet capped at iter 248 and never engaged (`auto-curriculum-engagement-check`); the saturation ramp re-fired at +600 after every resume and no shipped product ever saw the penalty (`curriculum-counter-lineage-steps`); the same penalty worked once gated to the disease phase and became an untouchable mechanism (`gate-penalties-to-the-disease-phase`); record-high aggregate reward hid a fully-failing delay stratum (`aggregate-metrics-mask-subgroup-failure`); bucket share is not a gradient lever (`bucket-share-is-not-a-gradient-lever`). An assist curriculum keyed to a pooled success share was withdrawn on the strength of the categories that already worked (`curriculum-criterion-conditioned-on-lagging-category`); a pace set by per-step income moved only when that income was time-gated (`per-step-income-drives-speed-time-gate`), and the same gate had to be retired in a lineage without the disease (`time-gate-vs-wide-stance-retire-the-fix`).
Coach application. Ask every curriculum three questions: does it engage (show the internal state)? what does it count (process or lineage)? when is it present (against the pathology's phase)? Check where shipped checkpoints sit relative to every ramp.
doctrine-17Probe before training: feasibility first, hypotheses in tablesAfter two failed training attempts at a skill, stop training: demonstrate the behavior open-loop, enumerate hypotheses in a written table audited against actual configs cheapest-first, race one probe per side of the sim2real boundary for hardware-only pathologies, and use suspended tests to acquit or convict actuators before blaming authority.
Case. "在黑暗里试钥匙" - four sidewalk rungs failed until an open-loop probe separated exploration/waveform/authority in one experiment (`open-loop-probe-before-reward-tuning`); the foot-drag mystery fell to a seven-hypothesis config audit (`hypothesis-table-code-audit`); the period-doubling was resolved by racing a reward-side and a plant-side evidence line - and both paid off, one per sub-case (`period-doubling-evidence-race`); the suspended test acquitted the roll actuator in one measurement (`suspended-test-isolates-actuator-authority`). A read-only configuration probe told a wall from a slope in the recovery line's seated basin (`configuration-probe-wall-not-slope`), and the fix it pointed to - where the feet are - took prone from 0/159 to 158/159 (`prone-dead-end-is-foot-placement`); a knob that did not move its variable was recorded as no test of the idea (`dof-vel-penalty-is-not-a-pacing-knob`).
Coach application. When a skill resists training, prescribe the probe before any further reward edits; require verified target trajectories before imitation terms; keep a falsified-fixes list so closed roads stay closed (`amplitude-cut-falsified-yaw-fix`).
doctrine-18External advice is recomputed locally; values transfer as ratiosEvery external suggestion is classified adopt / already-have / modify / trap by recomputing its claim on the local reward table and probe data; numeric values transfer only as dimensionless ratios (to tracking weight, leg length, sqrt(gL), control rate); citations are verified to exist.
Case. "Start vy very small" would have destroyed sidewalk learning on this reward table - the gradient scales quadratically (`external-advice-audit-against-own-arithmetic`); swing-height targets and weights transferred correctly only through leg-length and tracking-ratio scaling (`transfer-ratios-not-absolutes`); the "6-step delay" was refused for lacking a control rate (`latency-dr-covers-measured-pipeline`); a borrowed reference's structure was FK-verified and its amplitude re-derived from the division of labor (`reference-structure-fk-amplitude-division`); retrieval agents fabricated verbatim arXiv quotes - only source-verifiable material was used; and one dismissed suggestion later proved right for a different mechanism, and was credited (`cycle-average-tracking-for-gait-quantities`). An advisor's staged state machine turned out to exist in none of the three papers it cited, and reading them changed the plan (`advisor-paraphrase-vs-paper`).
Coach application. Intercept every "paper X does Y" with the local recomputation; convert absolutes to ratios before comparison; verify quotes; revisit dismissed advice when new mechanisms appear.
doctrine-19Hardware sessions are scripted experiments, not tuning sessionsReal-robot time executes a pre-registered matrix: risk-ordered (baseline first, fragile last with a spotter), stage-gated (suspended smoke before ground), A/B sessions bracketed by a repeated reference run, operators briefed on measured zero-command and untrained-axis behavior, chirality-aware disturbance protocols, no field tuning - the only legal field changes are scripted, single-variable, and self-reversing.
Case. The S2 acceptance sheet (`risk-ordered-real-deployment`, `battery-bracketed-real-ab`, `know-zero-command-behavior`, `push-test-chirality-protocol`, `no-field-tuning-protocol`); the RAM-only torque experiment with automatic power-cycle rollback (`reversible-single-variable-field-experiments`); and the sim-veto rule - even sim's condemnations get one safeguarded hardware check when they judge the purpose-built configuration (`sim-veto-needs-real-confirmation`). The recovery line's first real run went ahead with its preconditions unmet and was stopped as dangerous (`first-real-get-up-violent-stage-one-policy`); after it: a staged hang, mat and floor protocol (`staged-hang-mat-floor-for-get-up`), a fixed power-cycle pre-flight and two-machine discipline (`power-cycle-preflight`, `two-machine-config-discipline`), a fall guard replaced rather than switched off (`fall-guard-becomes-a-state`), and logs that are part of the run (`hardware-log-is-the-attribution-input`).
Coach application. Turn every hardware request into a runbook with order, gates, brackets, briefing, and anomaly plays; refuse improvised parameter changes on the floor.
doctrine-20Close questions in writing; restart when the debt is structuralAudited questions get frozen verdicts with citable wording and an explicit reopening bar; hardware verdicts are dated by deployment-stack and calibration state and expire when those change; and when successive rungs shuffle symptoms without net progress, freeze the lineage as regression baselines, pay the structural debts, and retrain minimal - carrying laws and instruments, not weights.
Case. The chirality and COM questions were closed with frozen wording and "no reopening without new hard evidence" (`frozen-verdicts-semantic-boundaries`); v5/v6's condemnations expired with the deploy stack (`stale-verdicts-under-old-stack`); a 2-degree calibration fix moved the whole runnable envelope (`zero-offset-calibration-shifts-envelope`); plant upgrades are era boundaries with paired re-baselining (`plant-swap-invariants-vs-shifts`); and the 2026-08-05 reset froze v5-v11, fixed the latency FIFO / manifest / sampling / reward-table debts, and restarted - producing the lineage that reached hardware SOTA (`freeze-lineage-fix-structure-restart`, `minimal-reward-table-with-provenance`). The recovery line's real-robot verdicts ended up in three places that disagree, one of them an undated note in a command file (`write-hardware-verdicts-back`).
Coach application. Maintain the closed-questions ledger and quote it when symptoms recur; stamp verdicts with stack/calibration versions; when a team is three rungs into symptom-shuffling, raise the restart question explicitly with the freeze-fix-restart pattern.
doctrine-21Name the quantity in the space it lives inA goal, reward term or acceptance criterion about the feet, the base or the contact state is computed from the quantity itself - world poses, forces, per-category outcomes - never through a joint-angle, single-signal or pooled stand-in that assumes everything else sits at nominal; and every detector is validated on a behaviour known not to contain the event before it becomes a gate.
Case. The recovery line was caught three times: |ankle roll| as "flat feet" sold stance width and the real robot slid into the splits, a hip-roll criterion was confounded by 50 deg of yaw, and the joint table said 0.271 m where the feet were 0.159 m apart; task-space terms produced the first flat, wide stance (`joint-space-proxy-for-task-space-quantity`). Flight detection lied in both directions across two lines - foot height flagged 40% false flight on a walking gait, contact force alone flagged slip chatter as hops (`contact-detector-single-signal-lies`). A pooled height average described a robot that did not exist - six in ten standing, four in ten sitting (`zero-partial-credit-is-not-an-iteration-problem`) - and the walking line had learned the same lesson on yaw rate (`heading-integral-not-body-rate`).
Coach application. For every reward term and gate row, ask what physical quantity it stands for and whether it is measured directly; flag joint-space or single-signal stand-ins for task-space goals, ask for a detector validated on a negative control, and split pooled metrics by category before reading them.
doctrine-22Continuation needs a live gradient; a release is chosen by a scanContinue a converged policy only on a change that creates a live gradient, on a short budget, with every checkpoint scanned on the transfer axis; choose a release by running the full battery over a band of checkpoints and stop on signals, never by taking the last one; and when edits to the terminal phase cannot move a behaviour, roll back and retrain with the constraint present from the start, keeping the order in which the lineage acquired its mechanisms as explicit curriculum phases.
Case. A continuation with no new gradient drifted MuJoCo transfer from 100/98% to 80/28% while every Isaac gate stayed perfect, and a live-gradient continuation at the same depth kept it (`converged-continuation-is-poison`). One-leg checkpoints 100 iterations apart failed 1 and 38 of 40 cells, and late ones degraded (`checkpoint-choice-is-a-full-gate-scan`). Four in-lineage stance fixes failed because the stance was the end of the get-up path, and from scratch it grew right (`stance-decided-by-get-up-path`); fixes stacked on degraded states were rolled back by the user (`stop-stacking-roll-back-and-audit`); and the lineage's final recipe, trained from scratch in one run, sat at 0% because the order of its curriculum was part of the product (`curriculum-history-is-part-of-the-product`). The omni line's short adaptation budgets and mature roots are the same law seen from the other side (`continuation-budget-not-from-zero`, `root-maturity-vs-product-quality`).
Coach application. Before approving a continuation, ask for the new gradient, the budget and the transfer axis in the scan; before approving a release, ask for the scan; after three rungs without progress on the target, propose rolling back to the last good checkpoint and a from-scratch phase plan instead of a fourth patch.
Experience cards
142 cards matching “stance-decided-by-get-up-path”.
Symmetrizing the config made the gait MORE asymmetric - the asymmetry lived in the policy weights
asymmetry-in-weights-not-configLocalize a persistent asymmetry by intervening at the config layer first: if the symptom survives (or worsens), it is in the weights - fix it with symmetry-constrained training, not with trims or offsets.
Symptom
walk_v1 on hardware: straight-line command curved 149 deg in 15 s (9.9 deg/s) with 3.06 m lateral runout; turn gain +31% one way vs +129% the other (75% difference); knee asymmetry 4.4 deg in sim, 9.6 deg on the robot.
Context
The obvious suspect was the asymmetric default pose in the config. The decisive test: symmetrize standing_pose and run the SAME policy in sim - the asymmetry got LARGER (hip_pitch 6.8 -> 9.3 deg). Root cause therefore not in the config but baked into the policy weights: PPO without a symmetry constraint commonly converges one-sided, because splitting the work 50/50 and loading one side yield the same return, and the gradient falls randomly into one of the equivalent optima.
Change
Fix redirected from config trimming to retraining with mirror data augmentation (walk_v2 spec) - a weights-level fix for a weights-level disease.
Outcome
With augmentation (and the symmetric-default precondition), stand_v1 reached 0.0 deg asymmetry on all six joint pairs (from 4.4-7.7 deg), height fluctuation 7 mm -> 1 mm, mean |action| down 33%.
Mechanism
Reward-equivalent solution families (who carries the load) leave the symmetric solution unpreferred; SGD picks an arbitrary member and entrenches it. Config changes move the coordinate frame around the entrenched asymmetric function - they cannot move the function. The counterfactual test (change config, watch symptom) localizes the layer the disease lives in.
Applies when
- a robot veers or loads one side despite a symmetric-looking config
- deciding between config trims and retraining for an asymmetry
- mirrored-turn gains differ by tens of percent
“根因不在配置里:把 standing_pose 对称化后在 sim 里跑同一策略,不对称反而变大(hip_pitch 6.8°→9.3°)—— 说明不对称烙在策略权重里。这是无对称约束的 PPO 的常见收敛结果(左右各担一半与一边多担的回报相同,梯度会随机落进其中一个)。”
train/RETRAIN_v2.md § 1. 为什么是对称增强(证据) A 2-degree joint-zero calibration fix moved the whole runnable envelope - re-test old "cannot run" verdicts after recalibration
zero-offset-calibration-shifts-envelopeDate every hardware verdict with the calibration state; after any zero/mount recalibration, re-test previously condemned policy-power combinations and previously "unexplainable" posture offsets before attributing either to training or model.
Symptom
s1d was on record as "only runs at power 0.7" (kicked wildly at 0.8); after a calibration pass, the same policy ran 12 s at 0.8 with no kicking at all.
Context
The calibration had fixed a 2.08 deg zero offset on r_hip_roll - exactly the constant error source on the dominant joint of the kicking oscillation loop ("恰是乱踢振荡环主导关节的常值误差源"). The three-generation post-calibration hardware sweep also closed a second case: the robot's mysterious "backward lean" disappeared after calibration, and the sim-real posture difference collapsed from opposite-sign 5+ deg to same-sign ~2 deg ("后仰案实质了结") - the lean had been a sensing/zero artifact, not a mass-model error. Booked consequence: if the s1d recovery re-verifies, "真机可跑档整体 上移" - every policy's runnable power envelope shifts up, and downstream lineages' hardware expectations get revised.
Change
Joint-zero and mount calibration promoted from setup chore to a variable that dates hardware verdicts: verdicts about which power/scale levels a policy can run are conditioned on the calibration state they were measured under.
Outcome
One policy rehabilitated at a higher power level; one standing sim-real posture discrepancy closed without touching model or training; a pending re-verification booked rather than asserted.
Mechanism
A constant joint-zero error acts as a persistent disturbance injected at the feedback loop's most-loaded joint; near an oscillation threshold, removing a 2-degree bias is the difference between a stable and an unstable loop. Since the error is additive and machine-side, it shifts every policy's stability envelope simultaneously - which is why verdicts must carry their calibration date.
Conflicts
The s1d rehabilitation awaited one confirming re-run at the time of writing ("待复核一跑坐实") - the offset-as-cause reading is the head suspect, not a closed verdict.
Applies when
- a policy oscillates at a power level others tolerate
- sim and real disagree on a constant posture offset
- deciding whether to re-test old hardware verdicts after maintenance/calibration
“发现①:s1d@0.8 能跑了(旧账「只有 0.7 能跑」)——12s 无乱踢。头号嫌疑 = 标定修正:r_hip_roll offset 修 2.08°,恰是乱踢振荡环主导关节的常值误差源。… 发现②:「后仰」标定后消失 … sim-real 姿态差从反号 5°+ 收敛到同号 2°,后仰案实质了结。”
train/README.md § 真机 @0.8 三代横评(2026-08-07 标定后) Cutting swing amplitude 40% raised yaw-momentum demand 53% - the falsified fix is recorded so nobody walks that road again
amplitude-cut-falsified-yaw-fixTest gait fixes against the quantity the ground must actually supply (torque/force rates vs friction ceilings), not against kinematic proxies; record falsified fixes with their mechanism so the search space shrinks permanently.
Symptom
Support-foot yaw slip stayed at ~223 deg (vs 225 deg) after walk_v6 cut joint swing amplitudes by ~40% (hip_pitch 20.6->10.4 deg, knee 31.3->18.8 deg) - the change built on the theory "smaller swing = less yaw momentum to dump into the ground".
Context
Direct measurement inverted the theory: yaw-momentum amplitude ROSE 16% (+/-0.1000 -> +/-0.1159) and its rate of change rose 53% (1.86 -> 2.84 N*m demanded from the ground), pinned exactly at the foot's supply ceiling (2.8-3.3 N*m at mu 0.6-0.7) - so slip could not drop. The extra demand lives in higher harmonics: v6's crisper foot placement (better clearance 34 mm, lower landing force) shortens the momentum- exchange window, concentrating the same exchange into less time. The verdict was written as a closed road: "下一轮不要再走这个方向". An honest residue was also booked: WHY amplitude down but momentum up 16% remained unresolved, with the named next step (per-rigid-body decomposition of H_z, since per-joint RMS sensitivity ignores phase correlations).
Change
The "reduce amplitude to reduce yaw momentum" lever was removed from the planning space; future yaw-slip work redirected toward the supply side (friction) and momentum-rate mechanics.
Outcome
Slip unchanged (225 -> 223 deg); the falsification and its mechanism became a permanent constraint on the fix search space.
Mechanism
Ground yaw torque demand scales with the rate of change of angular momentum, not its amplitude; kinematic amplitude cuts that also sharpen contact timing can raise dH/dt while lowering range. When demand exceeds the friction-limited supply ceiling, slip is set by the ceiling, so demand-side changes below the ceiling do nothing visible.
Applies when
- attacking foot slip or yaw drift via gait shape changes
- a fix targets an amplitude while the constraint is a rate
- documenting a failed intervention after a version comparison
“walk_v6 | ±0.1159 (+16%) | 2.50 Hz | 2.84 N·m (+53%) … 脚的供给上限 2.8~3.3 N·m(μ 0.6~0.7)—— v6 正好顶在天花板上, 所以滑移一点没降。… 结论: "减小摆动幅度以降低偏航动量"这条被 v6 证伪 —— 砍 39% 幅度, 需求反升 53%。下一轮不要再走这个方向。”
train/WALK_DIAGNOSIS.md § ② 未生效的机理: 减小摆动幅度反而让偏航需求上升 Record where every failed episode ends - an end-state confusion matrix showed all failures finishing seated and overturned a "cannot roll over" diagnosis that per-category success rates hide by construction
end-state-confusion-matrixFor any multi-category acceptance, report where each failed episode ends, not only which category it started in; it costs a few lines and no extra simulation, and it separates "cannot reach the goal" from "reaches the wrong basin".
Symptom
Prone scored 0% for three generations; the working diagnosis was "prone lacks the roll-over skill", and R0.3 spent a run adding prone-to-side roll-arc start states. It bought nothing: the 45-deg roll band itself only moved from 24.2% to 26.6% after 3,000 iterations.
Context
Acceptance reported success per starting category. A final-state table (lying prone / on the side / supine / seated / standing for every failed episode) was added to accept_recovery.py at R0.3.
Change
The confusion matrix became a permanent part of the acceptance output, and the prone diagnosis was rewritten from it.
Outcome
The prone, side and supine columns were all zero - every failure ended seated - and prone had righted its torso in 159/159 episodes (tilt under 30 deg in 100%). The missing ability was standing up from one specific seated configuration, not rolling over, which redirected the next rungs to foot placement and to a configuration probe.
Mechanism
Per-category success rates collapse "reached the wrong basin" and "never reached anything" into the same zero; the end state separates them.
Applies when
- a category sits at 0% and the diagnosis rests on its label
- recovery, manipulation or navigation tasks with distinct terminal states
- an intervention aimed at the presumed cause shows no effect
“**① 末态混淆矩阵 —— 固化(已在 `accept_recovery.py`)。** 它给出的 "趴/侧躺/仰躺三列全 0、所有失败都终于坐姿"是本线最改变决策的一个事实, 而**逐类成功率按构造看不见它**。 … 成本十来行、零额外仿真。 … **② prone 病因更正(旧诊断作废)。** 旧:"缺翻身"。新:**prone 159/159 全部 把躯干翻正**”
git:Lucen-recovery@origin/recovery:train/RECOVERY_V0_SPEC.md § §15 R0.3 判决 + 三件事的判断 Adapting a lineage to one plant increment needs hundreds of iterations, not thousands - long runs only buy specialization
continuation-budget-not-from-zeroBudget continuation rungs by increment class (hundreds of iterations for plant pins and smooth shifts, ~1000-1500 only for behavior-demanding changes like push), enforce a hard cap with frequent evaluation, and treat remaining budget as a reason to stop, not to continue.
Symptom
The default "6000 iterations per rung" (a from-zero-scale budget) was about to be applied to continuation rungs whose only change is one plant/DR increment - overspending compute and, worse, giving each rung thousands of iterations to specialize away retained skills.
Context
The 2026-08-07 budget table replaced the default with "最低适应窗口 + 每 100 iter 验收 + hard cap" scaled to the increment's difficulty: fixed-latency levels 300-500 (cap 500-800; the base has already seen in-band values, this only pins the plant); PD full-band 700 (cap 1000; kp+/-20%/kd+/-30% clearly widens the actuator family); COM +/-20 mm 500 (cap 800; a smooth dynamics shift); friction DR 700 (cap 1000; contact and actuator friction change the gait/contact solution together); push 1000 (cap 1500; a non-static plant change requiring recovery behavior - hardest). Rationale: "续训适应一个 plant 增量不需要从零量级的预算,跑长了只是给特化时间". The C ladder reused the scheme (per-rung caps 500-2000 by increment type), and the deep-training hazard got its own name when long runs sold quality ("深适应卖质量" - deep adaptation sells quality).
Change
Per-rung iteration budgets set by increment class with hard caps and 100-iter watch loops; checkpoint selection inside the window by the smoke curve, never "run to cap because budget remains".
Outcome
S2/C rungs completed in 300-1500 iterations each; the recurring late-run degradations (collapse valleys at 1500+, vx+0.30 decay) fell outside most rungs' caps instead of inside their runs.
Mechanism
A continuation rung's learning problem is local robustification around an existing optimum - low sample complexity; iterations past adaptation are spent sharpening onto the current distribution, which is exactly how retained skills and margins erode. Budgets sized to the increment bound both compute and the specialization damage window.
Applies when
- planning iteration budgets for a robustification or command ladder
- a continuation run keeps improving its training metric late
- retained skills decay in the back half of long continuation runs
“「最低适应窗口 + 每 100 iter 验收(watch_ckpt --every 100)+ hard cap」—— 续训适应一个 plant 增量不需要从零量级的预算,跑长了只是给特化时间 … ⑥ push | 1000 | 1500 | 非静态 plant 变化,要学 recovery 行为,最难”
train/OMNI_V0_SPEC.md § 4. 每级 iter 预算(2026-08-07 用户定) Order hardware runs by sim risk, gate each stage on the last, and put the fragile cell last with a spotter
risk-ordered-real-deploymentScript hardware sessions as a risk ladder: baseline first, sim-riskiest last with a spotter, suspended smoke before ground, each stage gated on the previous, environment (floor mu) recorded as a selection input - and stop at the stage that misbehaves.
Symptom
Five policy-x-gain combinations had to go on hardware in one session, with sim survival ranging from 20/20 down to 17/20 (and zero-command survival down to 2/20) - an unordered session risks breaking the robot on an avoidable run.
Context
The execution sheet fixed the order as sim-risk low to high, control baseline first (current SOTA establishes the floor reference), the fragile cell (fric-2400@kd1.0) last with a person spotting throughout. Stage gating: suspended smoke (feet off ground, 10 s each, all five pass before anything touches down) -> suspended with IMU and forward command (gait forms in the air) -> grounded runs -> speed raise only for combos that survived the previous stage -> zero-command tests only with a spotter, ordered by sim zero-cmd survival, with the 2/20 cell skipped by default. Preconditions include recording the floor material and estimating mu (if mu <~0.6, sim says pick the kd1.2 gain as main), port/CAN self-check, calibration frozen. Any stage failing stops the session at that stage: "任一段出问题就停在那一段, 不要跳到下一段".
Change
Session structured as a risk ladder with per-stage gates instead of a flat checklist; per-combo sim survival numbers written into the run table as the ordering key.
Outcome
The session design localized any failure to the cheapest stage that could reveal it, kept the robot safe for the informative fragile run, and made the control baseline available before any comparison run.
Mechanism
Hardware sessions consume a shared budget (robot integrity, battery, floor time); ordering by predicted risk means information is bought cheapest-first, and stage gates convert an expensive failure into a cheap earlier one. Baselines run first because every later reading is relative to them.
Applies when
- taking multiple policies/configs to hardware in one session
- a candidate is known-fragile in sim but must be measured
- writing a deployment runbook for a new robot
“跑序 = sim 风险从低到高, 最险的放最后 (依据 = 存活门/零指令存活) … ⑤ 是 sim 里最脆的一格 … 放最后跑, 全程留人扶, 起步即给 cmd, 零指令不做。… 任一段出问题就停在那一段, 不要跳到下一段。”
train/REAL_RUN_S2.md § 上机名单 / 全部命令 The hip_roll (l+r) asymmetry scalar predicted real-robot lateral drift - promote validated sim scalars into the gate
hip-roll-sum-predicts-lateral-driftHunt for cheap sim scalars that predict real-robot behaviors, validate them on direction AND ordering across multiple policies, then promote them into the acceptance battery; treat later violations as debt to justify in writing, not noise to ignore.
Symptom
A persistent hip_roll left/right asymmetry row in the sim2sim symmetry table had been dismissed as "calibration or mechanical asymmetry" noise; meanwhile real deployments drifted sideways by policy-dependent amounts.
Context
Forward-kinematics analysis reframed the scalar: both hip_rolls move the feet in +y for positive angle, so a same-signed (l+r) sum IS a lateral translation mode - the scalar is a direct lateral-drift bias estimate. Checked against real deployments: s1e (l+r = -0.0178, smallest magnitude) was the steadiest with least drift; 700 (+0.0253) drifted mildly left; A800 (+0.0267) drifted clearly left with the largest tilt 12.9 deg. Direction correct 3/3, ordering correct 3/3 (the log's heading calls it "四枚四中", four-for-four).
Change
The scalar was promoted into the acceptance battery as a posture-class criterion alongside tilt-max median: "hip_roll 左右不对称 |l+r| 不得比父代大" - doubling as a heat proxy (error ~ torque ~ heating).
Outcome
Used at every later gate; when the C4 product exceeded it by +0.005 rad (~+0.3 deg vs parent), the criterion was not silently waived - it was booked as explicit debt with a mechanism argument (the increment is task-required, far smaller than the sidewalk amplitude +/-2.2 deg) plus a related account (stand saturation 32.4% -> 37.2%).
Mechanism
A policy's static joint-angle bias in a translation-producing mode integrates into real-world drift; sim can measure that bias precisely and cheaply. A sim scalar earns gate status exactly when its predictions are validated against hardware in both direction and ordering - and a validated gate may only be exceeded with a written mechanism-level justification, never silently.
Conflicts
The log's heading says "四枚四中" (4/4) but the evidence table lists three policies and the text says "方向 3/3、排序 3/3"; the fourth instance is not shown in this file.
Applies when
- a real robot drifts or leans in a policy-dependent way
- deciding which sim measurements deserve gate status
- a validated gate criterion is marginally exceeded by a new product
“s1e | −0.0178（绝对值最小）| 微右、最不飘 | 三者中最稳、飘最小 ✓ … A800 | +0.0267 | 左、最飘 | 明显左飘、倾角最大 12.9° ✓ 方向 3/3、排序 3/3。 → 正式纳入验收表（与「倾角 max 中位」并列为姿态类判据）。”
train/C_LADDER_RUN.md § 3e. 顺带：hip_roll 左右不对称 (l+r) 就是横移偏置 —— 四枚四中 Late training leaned on sampling noise as a stability crutch - deterministic play collapsed while training metrics stayed green
noise-crutch-deterministic-collapseEvaluate the deterministic policy in an external harness on a fixed cadence during training (not just at the end), select checkpoints on that curve, and treat a collapsing noise_std with rising training reward as a warning that noise is load-bearing.
Symptom
omni_s1's final checkpoint (model_5999) fell at 4 s even in Isaac's OWN deterministic play, while checkpoints from iter 1700-4000 were fine - and no training metric flagged anything. Policy noise_std had collapsed to 0.045 by iter ~990 (final 0.033).
Context
Diagnosis: the policy had learned to use its exploration noise as a dither/stabilizer - "策略把采样噪声当稳定拐杆,训练指标看不见" (the training metrics cannot see it, because training always runs with noise on). Countermeasures: entropy_coef 0.005 -> 0.01 to slow the std collapse, and - the structural fix - an in-training smoke loop (watch_ckpt.py): every 500 iters, export ONNX directly, run 3-seed MuJoCo evaluation, log CSV/TensorBoard curves plus three-view videos. The doctrine line was written in bold: "训练指标全绿不再是发育健康的 证据,冒烟曲线才是" - green training metrics are no longer evidence of healthy development; the smoke curve is. The follow-up run s1b showed the drift metric follow a U-shape (73 -> 8.6 at iter 3500 -> 76), making checkpoint selection BY the smoke curve (early stop at 3500) the shipping mechanism, with terminal re-degradation booked as known and unresolved.
Change
entropy floor raised; watch_ckpt smoke loop instituted as standing infrastructure; checkpoint selection moved from "last iteration" to "best point on the deterministic smoke curve".
Outcome
s1b shipped from iter 3500 (the U-bottom) instead of a degraded terminus; every later lineage (s1c/s1e, the C ladder's --every 100 loops) inherited the watcher as the standard guardrail.
Mechanism
PPO evaluates and improves the stochastic policy; if noise itself stabilizes the gait (dither smoothing a marginal limit cycle), the deterministic mean policy is a different, worse controller that training never measures. External deterministic evaluation on an independent simulator is the only readout of what will actually be deployed.
Applies when
- final checkpoints underperform mid-training ones
- noise_std collapses early while training reward climbs
- deciding which checkpoint to export and ship
“训练后期确定性脆化——noise_std iter~990 收到 0.045(终 0.033),model_5999 连 Isaac 确定性 play 都 4 s 摔(1700~4000 正常):策略把采样噪声当稳定拐杖,训练指标看不见。对策:entropy_coef 0.005→0.01 + train/watch_ckpt.py 训练中冒烟曲线 … 训练指标全绿不再是发育健康的证据,冒烟曲线才是。”
train/OMNI_V0_SPEC.md § 3. S1.1 修订记录 ② Export every CAD part in the whole-machine frame so URDF rotations are zero and inertia is exact
urdf-shared-origin-exportGenerate the model so that correctness is structural: shared-origin STL export, zero rotations, subtraction-only origins, and an explicit 1e-9 g*mm^2 -> kg*m^2 conversion - never hand-rotate inertia tensors.
Symptom
Hand-assembled URDFs accumulate per-link rotation/origin errors and unit-conversion mistakes in inertia tensors - silent plant corruption that no later calibration can cleanly fix.
Context
Documented CAD -> URDF -> USD procedure from a successful Isaac Lab deployment, kept as the recipe if Lucen regenerates its model.
Change
(1) In CAD, align the whole robot to Z-up, X-forward (Isaac Lab convention) and ground the assembly; (2) export each STL with other parts hidden but the machine's shared origin kept, so all parts share one origin, every URDF rotation is 0, and inertia matrices equal CAD values directly; (3) units: Fusion 360 gives g*mm^2, URDF wants kg*m^2 - multiply by 1e-9; (4) link origin = negative of the joint position; link COM = CAD COM minus joint position; joint origin = difference of the two joint positions; (5) after URDF -> USD import, open the USD separately and set it instanceable before saving.
Outcome
A URDF whose rotations are all zero and whose inertia tensors are CAD-exact, eliminating an entire class of hand-transcription plant errors.
Mechanism
Keeping one shared origin turns every frame transform into a pure translation computable by subtraction, and leaves inertia tensors in the frame CAD already computed them in - no rotation of inertia tensors, the most error-prone manual step, is ever needed.
Applies when
- building or regenerating URDF/MJCF from CAD
- inertia or frame bugs suspected in the plant model
- importing URDF into Isaac Lab / USD
“导出 STL 时隐藏其他零件但导出整机——这样所有零件共享同一原点，URDF 里所有 rotation 全是 0，惯量矩阵直接等于 CAD 值 / 单位：Fusion 360 给 g·mm²，URDF 要 kg·m²，乘 1e-9 / link origin = 该关节坐标取负 … URDF → USD 导入后必须单独打开 USD 设成 instanceable 再存”
Experience.md § URDF 制作流程 (lines 87-92) The median of a bimodal metric lands in the empty gap - check the distribution, and never judge swing on 3 seeds
median-hides-bimodal-distributionBefore quoting a median or mean, look at the distribution; report suspected-bimodal metrics as mode share plus per-mode ranges, use small-seed smoke runs only to screen trends, and size the seed count for decisions by the share resolution you need (here: 20).
Symptom
Years of "high swing variance" and undecidable 3-seed swing readings turned out to be one fact: the metric was bimodal all along - "历代 swing 高方差与 3-seed 不可判由此定性:一直在测双稳态系统" - and every median reported from it (e.g. 12.1 mm) described a value no seed ever produced.
Context
Concrete instances: s2e_pd-1400's 20 seeds split 2.6-4.9 mm vs 19.3-24.0 mm with zero seeds between; s1e-500 read 3.9 mm on seeds 0-2 but 23.1 mm median over 20 seeds; a 3-seed reading of 19.3 was logged as "double-peak optimism, lesson recurrence #4". The selection re-audit codified the sampling rule: "3-seed 的 swing 读数不可判点, 只能筛带,选点必须 20-seed" - 3 seeds may screen a band, only 20 seeds may pick a point.
Change
Swing (and any suspected multi-modal metric) reported as mode shares plus per-mode ranges instead of a bare median; 3-seed smoke numbers demoted to band-screening; all shipping/selection decisions moved to 20-seed batteries.
Outcome
The "swing debt" bookkeeping was reinterpreted as basin probability (see swing-bistability-damping-switch), and checkpoint selection stopped being whipsawed by which basin the first three seeds happened to fall into.
Mechanism
Central-tendency statistics presuppose unimodality; on a bimodal distribution the median tracks the mode SHARE, not any achievable behavior, and small samples alias the share entirely. Mode-aware reporting (share + per-mode stats) is the only faithful summary, and the needed sample size is set by the share resolution required.
Applies when
- a quality metric shows chronic high variance across seeds
- 3-seed smoke readings contradict 20-seed batteries
- reporting swing height, clearance, or any basin-prone metric
“中位数落在空档里,「swing 债 −11mm」实为「50% 概率掉进拖地吸引子」。历代 swing 高方差与 3-seed 不可判由此定性:一直在测双稳态系统。… swing 跨 seed 双峰 (500 在 seed0~2 只读 3.9mm, 20-seed 中位 23.1) —— 3-seed 的 swing 读数不可判点, 只能筛带, 选点必须 20-seed。”
train/README.md § swing 双稳态定性 / s1e 选点重审 Fine-tuning through a reward-table change was falsified (scatter, half-recover, collapse) - continuation training is legal only with the reward frozen
fine-tune-reward-change-falsifiedNever fine-tune through a reward-table change - retrain from zero; reserve checkpoint continuation for frozen-reward plant/DR widening, reset noise_std when branching, and watch for the scatter/half-recover/collapse signature as the abort trigger.
Symptom
The s1c A/B experiment: arm B fine-tuned from an existing checkpoint under the revised reward (same contract, same network, changed reward + small DR) and failed with a characteristic signature - scatter, half-recover, fall back ("打散→半恢复→摔回"); arm A trained from zero under the same config won decisively (full shaping lifted swing to 21.6 mm within 500 iters; shipped at 5500).
Context
Verdict recorded: "从零 + 强塑形是本机唯一验证过的发育路径" (from-zero plus strong shaping is this machine's only validated development path). The signature became a standing stop criterion in every later rung that touched a reward ("s1c B 臂签名,出现即停"). Crucially the boundary of the law was drawn explicitly when S2 continuation training was proposed: "当年证伪的是「奖励表中途改版的 fine-tune」… S2 续训奖励表全程冻结、只逐级加宽 plant/DR,属另一类" - continuing a checkpoint with the reward FROZEN while widening plant/DR one rung at a time is a different class and was allowed (and then worked, powering the whole S2/C lineage) - with the honest fallback that if frozen-reward continuation ever collapses, that rung retrains from zero and the doctrine gets re-examined with data. Fine-tune arms also need mechanical care: reset the checkpoint's collapsed noise_std (terminal 0.033 "会杀死探索") and account for iteration counters re-zeroing (curriculum gates fire immediately).
Change
Reward changes and lineage continuation permanently separated: reward revisions -> from-zero retrain; plant/DR widening -> frozen-reward continuation with per-rung gates; the B-arm signature promoted to a universal tripwire.
Outcome
No later reward revision was attempted by fine-tune; frozen-reward continuation carried S2 (PD/COM/friction rungs) and the C command ladder successfully from the s1e root.
Mechanism
A trained policy sits in an optimum of its reward's geometry; changing the reward moves the optimum but leaves the policy's exploration noise near-zero and its value function calibrated to the old returns - it disassembles the old solution faster than it can assemble the new one. Widening DR under a frozen reward instead keeps the optimum's identity and asks only for local robustification.
Applies when
- proposing to fine-tune an existing policy under a revised reward
- planning a robustification ladder from a validated checkpoint
- a continued run scatters then partially recovers then collapses
“B 臂 fine-tune 证伪(打散→半恢复→摔回——从零 + 强塑形是本机唯一验证过的发育路径)。… 当年证伪的是「奖励表中途改版的 fine-tune」(B 臂,塑形突变致终盘摔回);S2 续训奖励表全程冻结、只逐级加宽 plant/DR,属另一类;若 s2_lag1 续训本身塌方,回退方案 = 该级从零重训,续训教义再议(拿数据说话)。”
train/OMNI_V0_SPEC.md § 3. S1.3 / 4. 与 s1c fine-tune 证伪的关系 Audit which joints your imitation term constrains - a task that needs deviation is fighting the reference
imitation-term-scope-auditList which joints your imitation/deviation terms actually constrain and check the new skill's required motion against that list; for balance-coupled joints deliver references as feed-forward residuals, not absolute-position targets - and never assume "reference = 0" is neutral.
Symptom
Sidewalk would not learn despite a dedicated tracking reward; meanwhile the gait-shaping imitation term (joint_pos_ref) computed its error norm over ALL 12 joints while its reference covered only the 6 sagittal joints - roll/yaw reference was constantly 0.
Context
Two prior generations had shown the forward gait itself was taught by joint_pos_ref, not discovered by PPO (v6 halved the shaping and swing height collapsed 35 mm -> 4 mm). So the reference is load-bearing - but sidewalk requires hip_roll to deviate from nominal, and the all-joints norm punished exactly that deviation: "一边悬赏一边罚过程" (posting a bounty while punishing the process). A follow-up experiment (C4-redo3, free_roll=True releasing the 4 roll joints from the norm) raised the regularization headroom 6x -> 27x yet sidewalk stayed flat and released hip_roll wandered, killing other skills - net negative, withdrawn. A --roll-absolute probe showed the converse failure: pinning roll to a clock-driven absolute trajectory drove tilt 6.9 -> 13.7 deg. Conclusion recorded: absolute-position imitation cannot teach actions that must be superimposed on state feedback.
Change
The audit reframed the problem: neither punishing roll deviation nor freeing roll nor absolute roll tracking works; the reference for a balance-coupled joint must be delivered as feed-forward under the policy's residual control (see feedforward-for-phase-locked-skills).
Outcome
free_roll rung: joint_pos_ref term rose 0.887 -> 1.104 (release confirmed effective) but vy stayed flat; regularization hypothesis eliminated by experiment.
Mechanism
An imitation error norm defines a cage: joints inside it are pulled to the reference in absolute position, so any skill requiring systematic deviation is taxed per step; but joints carrying active balance cannot follow absolute references either, since their correct position depends on state. The scope and the delivery mechanism of the reference are therefore design decisions per joint, not defaults.
Applies when
- adding a skill that moves joints your reference sets to zero/nominal
- an imitation or deviation penalty coexists with a new tracking reward
- considering releasing joints from a shaping term mid-lineage
“前进步态也不是 PPO 自己发现的，是 joint_pos_ref 教出来的（v6 砍半塑形 → 抬脚 35 mm 塌到 4 mm…）。而 ref_joint_offset 原本只写 6 个矢状面关节，roll/yaw 参考恒 0 —— 侧走既没被教，roll 一偏离 nominal 反被 joint_pos_ref 扣分。一边悬赏一边罚过程。”
train/C_LADDER_RUN.md § 3e. 为什么首战 FAIL / 3i. 解锁笼子 Sort external training advice into adopt / already-have / modify / would-trap by recomputing it on your own config
external-advice-audit-against-own-arithmeticNever apply external tuning advice directly: recompute each claim on your own reward table and probe data, classify it adopt / have / modify / trap, and record why - and verify external citations actually exist.
Symptom
External AI/literature advice for the omni ladder arrived plausible-sounding but was written without knowledge of this robot's actual reward table, contract, and history; following it blindly would have broken single-variable discipline and, in one case, made sidewalk unlearnable.
Context
Before the C ladder, every external suggestion was audited: 3 adopted (ellipsoid command sampling; staged wz bands; command-switch acceptance), 3 already present (unified reward; frame history - the frozen 168-dim 5-frame window; per-100-iter acceptance), 2 modified (stand share kept at 20% to avoid a second variable; back share NOT raised because the probe showed backward works untrained 20/20@67%, so oversampling would only crowd out forward), and 1 flagged as a trap: "start vy very small (0.06-0.15)" - on THIS reward table vy was only an L2 tax, so ignoring a vy=0.06 command costs 0.4% of the vx tracking scale, 28-180x cheaper than ignoring forward, with quadratic shrinkage making small commands weaker still. A separate retrieval-reliability note: two search agents returned fabricated verbatim quotes from arXiv PDFs (2 papers, verified fake and discarded); only HTML/abstract/source-verifiable material was used.
Change
Advice classified only after recomputing each claim with local numbers; the "start small" trap was replaced by adding a gated lateral tracking term (the ladder's only true reward surgery) instead of shrinking the command.
Outcome
The adopted items (ellipsoid modes, staged wz, transition acceptance) entered the ladder; the trap was avoided; one external factual error (calling s1g the mainline start - it was falsified 0/20) was caught. Later, one initially-dismissed item (sigma=0.15 too narrow) turned out right for a different reason than claimed - see cycle-average-tracking-for-gait-quantities.
Mechanism
External advice encodes the advisor's reward table and robot, not yours; the transfer-validity test is whether the claim survives recomputation under your own arithmetic (reward margins, probe baselines, contract freeze). Items that survive become experiments; items that don't become documented traps.
Applies when
- incorporating LLM or literature advice into a training plan
- advice conflicts with locally measured baselines
- an external claim depends on reward-table details the advisor cannot know
“其建议 C4「先很小，vy = ±0.06~0.15」—— 在我们这张奖励表下会让侧走学不起来 … 忽略侧走比忽略前进便宜 28~180 倍，且指令越小激励越弱（平方缩放）—— "先很小"在稳定性上对、在梯度上正好把信号缩没了。”
train/C_LADDER_RUN.md § 1. 外部 AI 训练建议的评估（采纳 / 已有 / 要改 / 会踩坑） A nonzero response with the same sign for + and - commands is bias, not ability
same-sign-response-is-yaw-biasBefore crediting any directional skill, test both command signs: response must flip sign with the command; a same-signed pair is a bias to subtract, not an ability to report.
Symptom
Root-selection probe showed nonzero wz "tracking percentages" on turn commands, tempting the read that candidates could partially turn.
Context
During C-ladder root selection, s1e-500's measured yaw rate was +0.084 rad/s for cmd +0.3 and +0.093 rad/s for cmd -0.3 - same sign both ways. The same check on the C2 baseline gave wz+0.20 -> -0.13 and wz-0.20 -> +0.12 (again same sign), while the alternative root s2e_pd-1400 gave +0.16 / -0.16 - opposite signs, i.e. a genuine 16% command response.
Change
Reading corrected and written into the execution sheet: percentages on directional commands are meaningless unless the +cmd and -cmd responses have opposite signs; all three candidates were re-classified as "cannot turn, cannot sidewalk - C2/C3/C4 learn from zero". Acceptance criteria thereafter required "tracking >=50% AND left/right opposite-signed".
Outcome
Prevented crediting turn/sidewalk ability that did not exist; the antisymmetry clause became a standing part of every turn and sidewalk PASS condition (C2, C4, C4-redo levels all carry "且左右反号").
Mechanism
A constant yaw (or lateral) bias projects onto any command's sign convention and shows up as fake fractional tracking; only sign-antisymmetry under command reversal distinguishes a feedback response to the command from an open-loop offset.
Applies when
- evaluating turn/sidewalk/any signed-command tracking percentages
- a candidate shows partial tracking on an axis it was never trained on
- writing PASS criteria for a new directional skill
“C2/C3 那些非零的 wz 百分比不是转向能力 —— 转向+ 与 转向− 的实测同号（s1e：cmd +0.3 → +0.084，cmd −0.3 → +0.093 rad/s），那是恒定偏航偏置。… 三个候选都不会转、都不会侧走。”
train/C_LADDER_RUN.md § 0. 读数纠正（重要，别引错） Brief the operator on the lineage's measured zero-command and untrained-axis behavior before handing over the joystick
know-zero-command-behaviorBefore any teleop/demo, measure and write down the policy's zero-command behavior and per-axis competence, label untrained axes explicitly as not-bugs, and set the floor/procedure to accommodate the known drift.
Symptom
A teleop session was about to start on a policy that does not stand still at zero command and has never been trained on lateral commands - behaviors an unbriefed operator would report as bugs or emergencies.
Context
Three measured facts were written into the teleop instructions ("都有 实测依据, 不是猜"): (1) A/D (lateral) keys will get essentially no response - probe-measured sidewalk tracking ~3%, an untrained axis: "这正是 C4 要解决的事, 不是 bug"; (2) no keypress = cmd 0, and this lineage does not stand still at zero command - a three-generation lineage property: paces in place, drifts right ~5 cm/s, net rotation -30 deg/20 s; sim survival is 20/20 (it will not fall) but it walks away slowly, so leave floor margin especially on the right; (3) S (backward) WILL respond - probe-measured 20/20 survival, 67% tracking untrained, which is also why this root was chosen for the C ladder. Plus a keybinding dry-run while suspended before touching down.
Change
Operator briefing became part of the deployment artifact: expected response per key, expected idle behavior with magnitudes and directions, and the distinction between untrained (expected, not a bug) and abnormal.
Outcome
The session proceeded with correct interpretations available in advance; the known zero-command wander was handled by floor margin and start-with-command procedure rather than misdiagnosed on the spot.
Mechanism
A learned policy's off-nominal behaviors (idle drift, untrained axes) are lineage properties, stable and measurable in sim beforehand; operator surprise converts known properties into false incident reports and unsafe reactions. A briefing transfers the measured behavior model to the person holding the controller.
Applies when
- handing a learned policy to an operator or demo audience
- the policy idles in a non-stationary way at zero command
- some command axes are untrained in the current lineage
“A/D 基本不会有反应 —— s1e 从未训过非零 vy, 选根探针实测侧走跟踪率 ~3% … 这正是 C4 要解决的事, 不是 bug。… 不按键 = cmd 0, 而 s1e 在零指令下不站定 —— 血统属性, 三代实录: 原地踏步 + 右漂 ~5 cm/s + 净旋 −30°/20s。”
train/REAL_RUN_S2.md § 附: WSAD 遥控 上机前必须知道的三条 Diagnose a behavior failure by enumerating hypotheses and auditing each against the actual config, cheapest first
hypothesis-table-code-auditBefore changing anything, write the full hypothesis list for the symptom and audit each against the resolved config and measured magnitudes, cheapest check first; train only on the survivors.
Symptom
Real robot leaned forward "wanting to walk" but dragged its feet instead of lifting them - a symptom with many plausible causes and no obvious single fix.
Context
Seven hypotheses were listed and each checked against the actual training config files (velocity_env_cfg.py, isaac_values.py), ordered by check cost: missing foot clearance term (CONFIRMED, primary - feet_air_time existed but no swing-height term at all); energy penalties dominating (REJECTED - energy terms total -0.19 vs tracking +1.2, 16%); command range too narrow (CONFIRMED - (0.15,0.35)); nominal pose too crouched / action scale too small (HALF - knee 0.5 rad = 28.6 deg deep, scale fine); mixed PD across motor types (REJECTED - already grouped); missing base-height reward (REJECTED - present at -5.0); height-drop termination (REJECTED - none exists, which itself became finding #4 of the fix list).
Change
The audit produced a ranked fix list (add clearance penalty; widen speed range; reduce nominal crouch) with each rejected hypothesis documented so it would not be re-litigated.
Outcome
Three confirmed causes fixed over v5/v6: swing height went 22-23 mm -> 34 mm, tracking 81% -> 87%; the rejected hypotheses stayed rejected (no wasted rungs on energy weights or PD grouping).
Mechanism
Multi-cause symptoms invite guess-and-train loops; a written hypothesis table forces each candidate to be confirmed or rejected against actual values (not impressions), and cost-ordering the checks means most hypotheses die for the price of reading a config.
Applies when
- a real or sim behavior failure has multiple plausible causes
- the team is about to "try a fix" without an audit
- post-mortems keep re-proposing already-rejected causes
“真机现象:躯干前倾像要走,脚抬不起来(拖着蹭)。按成本从低到高逐条核查 … | 1 | 缺 foot clearance | ✅ 成立,首要 | 有 feet_air_time,无任何摆动足高度项 | | 2 | 能量惩罚压过跟踪 | ❌ 不成立 | 能量类合计 −0.19,跟踪 +1.2,只占 16% |”
train/WALK_DIAGNOSIS.md § walk 拖地问题 — 七条假设的代码核查结果 The real robot's right-leg kicking was over-trained-delay times loop gain - irreducible pipeline latency is plant, model it fully from day one
pipeline-latency-is-plant-not-drMeasure the end-to-end action pipeline delay and build it into the nominal plant and every acceptance gate from day one; treat power/scale deratings that "fix" oscillation as gain-reduction crutches flagging an unmodeled delay, and expect higher-feedback-gain policies to be MORE delay-fragile.
Symptom
On hardware, s1c/s1d at action scale 1.0 always kicked wildly with the right leg (s1c only ran as SOTA at power 0.8; s1d only at 0.7) - while sim showed nothing under default evaluation.
Context
Sim reproduced the incident item by item once the real pipeline delay was injected: s1d@1.0 with --delay 1 fell at 10.2 s, --delay 2 at 5.2 s; s1c@1.0 stressed (r_hip_roll saturation 5 -> 16%; "右脚" = the policy's chirality makes the right leg its high-gain leg); and the combos that worked on hardware (s1c@0.8+delay2, s1d@0.7+delay2) all survived in sim. Mechanism: the real pipeline is ~1-2 ticks (BusWorker next-cycle pickup + CAN round trip) but S1.1 trained only to 1 tick - "超训延迟 × 全环路增益 = 振荡;衰减 = 压环路增益换稳定" (delay beyond training x full loop gain = oscillation; the power derating had been buying stability by compressing loop gain). s1d was MORE fragile than s1c because its yaw 3-layer stack had learned higher feedback gain - higher gain, lower delay tolerance. Three changes: latency DR widened to cover reality; acceptance gates and smoke runs moved permanently to --delay 2 ("门必须在真机条件下预测 真机"); and the doctrine written twice-paid: "不可约的管线属性(延迟、 限速)不是'随机化选项',是 plant 本体,第一天就该全额建模" - S1's nominal-then-robust staging falsified by hardware for the second time. The later s1e hardware run at power 1.0 (no kicking, normal force) closed the loop: "0.8 = 旧代拐杖" - the derating had been a crutch for the under-modeled delay, not a real requirement.
Change
Latency modeled as plant from day one of any lineage (measured 1-2 ticks covered, bridge-layer rate limits likewise modeled by default); every gate and smoke evaluation issued under --delay 2.
Outcome
Kicking reproduced, explained, and eliminated in the s1e generation at full scale and full power; the deploy-side crutches (0.7/0.8) retired for the new lineage.
Mechanism
Feedback oscillation onset is a product of loop gain and phase lag; a policy trained below the real delay learns gains that sit past the real stability margin, and any output derating masks it by scaling gain down. Since pipeline delay is deterministic hardware property - not an uncertainty - it belongs in the nominal plant, and every evaluation must include it or the gate predicts a robot that does not exist.
Applies when
- hardware oscillation/kicking that sim only reproduces with added delay
- a policy only runs on hardware at reduced power/scale
- defining what belongs in the nominal plant vs the DR list
“真实链路延迟 ~1~2 拍 … S1.1 只训到 1 拍——超训延迟 × 全环路增益 = 振荡;衰减 = 压环路增益换稳定。s1d 比 s1c 更脆 = yaw 三层栈学出更高反馈增益,增益越高延迟容忍越低。… 教训入账:S1「先标称后鲁棒」第二次被真机证伪——不可约的管线属性(延迟、限速)不是"随机化选项",是 plant 本体,第一天就该全额建模。”
train/OMNI_V0_SPEC.md § 3. S1.4(真机右脚乱踢事故强制) A stand gate judged by survival passes a robot that wanders a meter - judge posture instead
stand-gate-posture-not-survivalFor every gate, ask what behavior the metric is a proxy for and validate its ordering against real observations; replace metrics whose ordering disagrees with reality, and demote them explicitly rather than silently.
Symptom
Real robot "standing" drifted 0.5-1.4 m across the floor while the sim stand gate scored a clean 20/20 - because the gate's metric was episode survival, which wandering does not violate.
Context
C2's stand condition was originally written as "survival regression <=2/20". A real-robot counter-example on 2026-08-08 forced the re-judgment: wandering robots survive. Cross-checking candidate sim metrics against real-robot feel showed max-tilt median ordering agreed with hands-on ranking, while displacement ordering was actually OPPOSITE to real impressions - so displacement was demoted to a reference quantity, not a gate.
Change
Stand PASS criterion rewritten from survival to posture: "stand tilt max median <= root baseline +1.5 deg"; displacement kept only as reference. Applied to all subsequent rungs (C4 and redo levels inherit it).
Outcome
Later rungs gated stand on tilt (e.g. C4 product: 7.7 deg vs parent 7.1 deg, +0.6 deg PASS); the wandering failure mode became visible to the battery instead of hidden by survival.
Mechanism
A gate metric is a proxy for an intended behavior; survival is a proxy for "did not fall", not "stood still". Metric choice must be validated against ground truth (real-robot feel/measurement), and a proxy whose ordering disagrees with reality on real data is worse than no metric - it steers selection backwards.
Applies when
- writing PASS conditions for stand/idle/hold behaviors
- a gate passes policies that visibly misbehave on hardware
- choosing between candidate metrics for an acceptance battery
“stand 必须用位姿判，不能用存活判（2026-08-08 真机反证改判）：站着乱走 0.5~1.4 m 时存活照样 20/20 —— C2 的 stand 条件原写「存活退化 ≤2/20」，选错了指标。改为: stand 倾角 max 中位 ≤ 根基线 +1.5°（倾角与真机手感排序一致；位移排序与真机相反，降为参考量）”
train/C_LADDER_RUN.md § 3b. PASS 条件 ⚠️ stand 必须用位姿判 Curriculum-gate a penalty to the phase where its disease occurs - early on it only taxes exploration
gate-penalties-to-the-disease-phaseFor penalties aimed at late-stage pathologies (freezing, saturation, degenerate attractors), ramp the weight in only after exploration noise has decayed; anchor the terminal weight to measured healthy-vs-sick raw values, and shift all related tripwires to after the ramp completes.
Symptom
The action_saturation penalty, applied from iteration 0 in v8a, taxed exploration itself: with init_noise_std 1.2 the sampled actions paid ~-2.45/step before any policy had formed - while the disease it targets (clamp freezing) is a LATE pathology (v9 froze at iteration ~2624).
Context
v10 re-introduced the same penalty behind a curriculum gate: weight 0 until iter 1000, ramping linearly to -1.0 by iter 2000 - present only when the disease can occur, absent while exploration noise dominates. The trust argument was evidence, not hope: in v8a the term, while active, had pulled joint_pos_ref from 0.041 up to 0.155 and climbing - proof it can extract a policy from the frozen pit. Weight magnitudes were anchored to measured raw values (healthy v5 0.310 / v6 0.106 vs frozen v7 1.145 / v9 1.22 per step: at -1.0 healthy pays 6-18% of tracking, frozen pays 60%+, standing ~0). v10c then isolated the gated term as THE anti-freeze mechanism by single variable, upgraded to untouchable status in v11: "S 的门控机制(v10c 单变量铁案:任何情况下 不许撤,只许调终值)" - and v11 dared to relax other penalties only because S stood guard.
Change
action_saturation gated 0 -> -1.0 over iters 1000-2000 (later terminal value tuned -1.0 -> -0.5 with the gate mechanism itself frozen); tripwires adjusted to respect the gate's timing (freeze check moved to iter 2500-3000 to give the ramped term its effect window).
Outcome
Freezing stopped recurring while early training kept full exploration; the mechanism graduated from experiment to invariant within two versions.
Mechanism
A penalty's incidence depends on who occupies its support: early in training that is exploration noise (whose suppression starves learning), late it is the converged pathology. Time-gating aligns the penalty's presence with its target's presence, buying the constraint without the exploration tax - and tripwire timing must then be computed from the gate schedule, not from ungated precedents.
Applies when
- a structural penalty punishes exploration in early training
- a late-onset pathology (freeze/saturation) needs a standing guard
- deciding when a curriculum ramp should engage
“v8a 实锤它的病根是"罚在采样动作上"——init_noise_std 1.2 的早期等于罚探索(~−2.45/步);而冻结是晚期病(v9 速率 2624 才死平)… 门控让它只在病发期在场。… v8a 里它在场时 joint_pos_ref 从 0.041 爬到 0.155 且仍在升——有从低谷爬出的实证。”
train/WALK_V10_SPEC.md § 2. S 保险 —— action_saturation 课程门控 Two simulators disagreed 1.8x on one policy's torque demand - four explanations were eliminated with numbers, the surviving suspect was never tested, and neither reading was allowed to cancel the other
torque-disagreement-between-simulators-unresolvedWhen two simulators disagree on a safety-relevant quantity, eliminate the measurement explanations one at a time with numbers, name the survivor as a hypothesis, keep the pessimistic reading binding, and run the direct test (replay one action sequence open-loop through both plants) before the quantity is used to pass a gate for hardware.
Symptom
For R3.0, Isaac read hip_pitch torque demand at 75-80% of the limit (PASS); MuJoCo read the same policy at 148-152% (1.5x over the limit).
Context
Candidates were eliminated one by one: PD gains (identical, RS06 kp 30 / kd 1.5), the settle window (both skip the first 0.5 s), the statistic (worse-of-two-legs vs per-joint - explains ~15%), self-collision (the no-contact subset still read 152%), sampling (500 Hz peak vs 50 Hz sample - ~9%). About 1.8x remained. The prime suspect - Isaac's implicit actuator (PD solved inside the PhysX integration, chosen to mimic kHz firmware PD) against MuJoCo's 500 Hz explicit PD - was named but untested.
Change
The Isaac PASS was recorded as valid for the Isaac plant only; the prescribed decider was an open-loop replay of one action sequence through both plants, compared step by step, needing no training. It was upgraded to a precondition of R3.1.
Outcome
R3.1 ran in parallel with, not after, the replay, and the replay never appears as done. By §40 it was "the oldest open account" on the line, to be fed by real-robot logs - which the spec also never records. Meanwhile the 50 Hz sample under-read the peak by 34% as smoothing narrowed the spikes, so Isaac-side readings grew more optimistic exactly when the comparison mattered more.
Mechanism
Each measurement artifact explained a slice of the gap; what remained is a plant difference, and an untested plant hypothesis cannot license discarding the pessimistic simulator on a safety-relevant quantity.
Conflicts
The spec prescribes the open-loop replay (§23), upgrades it to an R3.1 precondition, then records that R3.1 ran in parallel without it (§24), and lists it as the oldest open account before hardware (§40); nothing through §50 (2026-08-14) records a result. Every later "torque gate PASS" on this line is an Isaac-plant reading plus a MuJoCo check, never a reconciled one.
Applies when
- Isaac and MuJoCo (or sim and hardware) report different torques, contacts or slips
- implicit vs explicit actuator models are in play
- a gate passes on one simulator only
“**同一个策略,一侧判 PASS 一侧判超限 1.5 倍。** … 口径项全部扣掉后仍剩 **~1.8×** 没有解释。 … 但**这条没有验证,不能拿它当结论去抵消 MuJoCo 的读数**。 … (做法:拿同一条 动作序列在两边开环回放,逐拍比 τ —— 不需要重训)”
git:Lucen-recovery@origin/recovery:train/RECOVERY_V0_SPEC.md § §23 MuJoCo 复核门:成功率更稳,但 τ 与 Isaac 差 ~1.8× 没收敛 Every new penalty ships with a pre-registered withdrawal clause - if healthy gait must pay above the cap, the term stands down
calibration-threshold-with-withdrawal-clauseIntroduce every new penalty with: the zero-cost-option audit, a replay-calibrated weight formula (healthy pays a fixed small fraction of tracking), and a pre-registered withdrawal condition - and let the clause fire without argument when the calibration says the term cannot be afforded.
Symptom
Three same-shaped crashes had established a failure archetype: v4's clearance, v8a's landing window (weight off by 58x uncalibrated), and v6a's bare hip_yaw suppression all combined a zero-cost "don't move" option with a fee on any motion - a reverse barrier that pushes policies toward standing still.
Context
The v11 landing-window penalty was therefore introduced under a calibration-threshold protocol: (1) shape chosen with the window tightened (h_gate 0.03 -> 0.02, because 0.03 equaled the clearance target and priced the entire descent); (2) weight from a FORMULA, not judgment: measure the term's raw value on healthy replays (v5/v10b), set w = -(0.10-0.15 x tracking reward) / raw_healthy; (3) withdrawal clause pre-registered: if healthy gait must pay >15% of tracking no matter the tuning, the term is withdrawn to the next version rather than forced in - "不硬上". The companion hip_yaw quieting term ran the same protocol (calibrate on replays, healthy pays <=5%) and was later retired entirely when a structural fix (zero action scale) made its shaping tax unnecessary.
Change
Penalty introduction protocol: shape audit (what is the zero-cost option?), replay-based weight formula, healthy-pay cap with a written stand-down condition - all before training.
Outcome
The landing term was in fact withdrawn under its clause (v12 records "P5 落地窗口罚 已撤 … 维持撤下"), demonstrating the protocol firing as designed instead of the fourth same-type crash.
Mechanism
A penalty's damage mode is mispricing healthy behavior; since the healthy price is measurable in advance on replays, both the weight and the go/no-go decision can be computed rather than discovered by a ruined training run. The withdrawal clause converts "make it work" pressure into a clean deferral.
Applies when
- adding any motion-taxing penalty to a working gait
- a proposed term's weight has no measurement behind it
- a previous same-shaped term crashed training
“权重公式而非拍脑袋:先在 v5/v10b 回放上量 h_gate=0.02 的原始值,w = −(0.10~0.15 × 跟踪奖励) / raw_健康;标定门槛:若健康步态无论如何要付 >15% 跟踪,本项撤下留 v12,不硬上 (v4 clearance/v8a-B/v6a 三次同型翻车的教训:代价为零的"不动"选项 + 一动就收费 = 反向壁垒)。”
train/WALK_V11_SPEC.md § 6. P5 —— 落地窗口罚(三代欠账,标定门槛制) When the training reset distribution changes, freeze the acceptance distribution separately and pin its seed - the same checkpoint measured twice differed by 3.8 and 10.2 points
frozen-acceptance-distribution-and-pinned-seedAcceptance distributions are frozen artifacts, decoupled from whatever the training distribution becomes and evaluated with a pinned seed, and every gate row carries its sample size so that a small-n row is never read as a regression.
Symptom
R0.3 added easier roll-arc start states to the training resets, and the acceptance script shared the training category table; separately, the same checkpoint scored side 3.8 points and mid 10.2 points differently on two runs of the same script.
Context
Acceptance ran 512 parallel envs drawn from the fall categories. Had it kept following the training table, 15% of acceptance samples would have landed on states easier than prone - inflated scores and generations that could not be compared. The two same-checkpoint runs had identical per-item height medians, so the ruler had not changed; the spread was reset resampling noise (side n ~ 169, sigma 2.3%; mid n ~ 35, sigma 8.3%). The spec's "20 seeds" had always meant controlled seeds.
Change
accept_recovery.ACCEPT_CATEGORIES pinned to the four R0-R0.2 categories and decoupled from the training FALL_CATEGORIES; --seed 20260809 pinned, after which two consecutive runs were bit-identical. The mid row (n ~ 35) was labelled the bluntest gate.
Outcome
Every later generation (R0.3 through V3.1) was scored on the frozen distribution and seed, which is what let R0.3's intermediate state be read as "no measurable gain" (62.7 -> 62.1%) and R0.2's mid drop be booked as noise rather than a regression.
Mechanism
An acceptance set that follows the training distribution measures a moving target, and an unpinned reset draw adds sampling noise that small-n rows cannot absorb.
Applies when
- the training reset or command distribution changes between generations
- repeated evaluations of one checkpoint disagree
- a small category drives a pass/fail decision
“**分布冻结**:`accept_recovery.ACCEPT_CATEGORIES` 钉死 §5 四类 … 与训练侧 `FALL_CATEGORIES` **解耦**。 … **side 差 3.8 点、mid 差 10.2 点**(h 中位逐项一致,证明不是尺子变了)—— 纯 reset 重采样噪声 … 钉死后两次连跑逐位相同。”
git:Lucen-recovery@origin/recovery:train/RECOVERY_V0_SPEC.md § §14 验收尺子的两处加固(R0.3 起生效,向后兼容) action_rate weight is the sim2real bandwidth knob - re-tune it whenever a rate limiter is removed
action-rate-weight-vs-bandwidthSet action_rate weight relative to real actuator bandwidth, and re-tune it any time another smoothing/limiting element (filter, slew limiter, gain) changes - reward weights are load-bearing parts of the actuator model.
Symptom
With a low action_rate_l2 weight the policy learns fast actions; the unmodeled part of the actuator response is then excited hardest, and sim2real "直接崩" (collapses outright). With too high a weight, actions become so slow the robot cannot maintain balance.
Context
The reference developer called action_rate_l2 the single most important reward for transfer, with side-by-side video evidence that the high-penalty, slower policy is clearly better on hardware. Lucen context: the team had just removed the SOFT_SPD=1.0 velocity limiter, which had been an implicit actuator-bandwidth constraint - leaving action_rate as the only remaining constraint on action speed.
Change
Decision recorded: after removing SOFT_SPD, re-evaluate the action_rate weight rather than keep the old value, since its effective role changed from "additional smoother" to "sole bandwidth constraint".
Outcome
Logged as a priority follow-up ("重新评估 action_rate 权重 - 拆掉 SOFT_SPD 之后这一项的作用变了"); the failure mode it guards against is training high-frequency actions the real actuators cannot track.
Mechanism
Slower actions stay inside the frequency band where the ideal-PD sim actuator and the real actuator agree; fast actions probe the band where unmodeled delay, inductance, and bandwidth limits dominate, so model error is amplified in exact proportion to action speed. Any removed external rate limit transfers that constraint's entire job onto the action_rate penalty.
Conflicts
The low/high tradeoff evidence is the external developer's report (with video); the Lucen-side entry is a pre-registered risk and decision, not yet an on-robot A/B at the time of writing.
Applies when
- removing or adding an action filter, slew limiter, or low-level speed cap
- real robot shows high-frequency chatter or overheating absent in sim
- tuning smoothness rewards before a hardware deployment
“权重低 → 动作快 → 执行器模型不准的部分被放大，sim2real 直接崩 / 权重高 → 动作慢 → 好迁移，但可能慢到无法维持平衡 … 我们刚拆掉 SOFT_SPD=1.0 的限速器，等于把执行器带宽约束整个移除了。action_rate 惩罚现在是唯一还在约束动作速率的东西，需要重新评估权重”
Experience.md § action_rate_l2 是他认为最关键的 reward (lines 61-70) The restart reward table lists a reason for every term AND a lesson for every exclusion - absent terms are removed, not zero-weighted
minimal-reward-table-with-provenanceMaintain the reward table as an evidence ledger: every term cites the episode that justifies it, every excluded term cites the episode that convicted it (including the development stage it is valid at), and retired terms are deleted from the config, never left at weight zero.
Symptom
Seven walk generations had accumulated an entangled reward table where nobody could say which term earned its place; the restart needed a table that could be audited line by line.
Context
The minimal table v2 was built under three written principles: "一项管 一件事、结构性反抬脚的项一个不留、塑形只留一套相位逻辑;不在表里的一律 不加" (one term per job; zero structurally-anti-lift terms; exactly one phase-shaping logic; nothing outside the table gets added). Every row carries its provenance (e.g. base_height target = standing height cites the v4 crouch lesson; split x/y tracking cites the merged-exp gradient hole; world-frame yaw cites the v1/v3 body-frame lesson). Every EXCLUSION carries its same-type precedent: feet_landing_vel out because it is poison while the gait is unformed (drag pays 0, lifting pays - the v4-clearance / v8a-B "reverse threshold" family) though it was fine in v7 when the gait already existed - term validity depends on developmental stage; feet_air_time out with its measured non-lever evidence (v6 had it at 2.0 and still lifted 4 mm); and replaced terms are REMOVED from the config ("置 None,不是权重 0 挂着") so audits see truth, not dormant weight.
Change
Reward table rebuilt as ~20 rows each with weight + provenance column; exclusion list maintained alongside with the falsifying episode for each; dormant terms deleted rather than zeroed.
Outcome
Later revisions (S1.2/S1.3) modified the table by citing and updating specific rows' evidence rather than re-arguing the whole design; the table doubled as the lineage's reward-lesson index.
Mechanism
A reward table is a set of standing hypotheses; attaching each row's evidence makes revisions targeted and reversible, and recording why a term is absent prevents the cycle of re-adding known poisons. Deleting vs zero-weighting matters because config audits and DR interactions see the term either way - a zero-weight term is dormant complexity waiting to be flipped on wrongly.
Applies when
- designing a reward table for a restart or new task
- someone proposes re-adding a previously removed term
- auditing which reward rows still earn their place
“原则:一项管一件事、结构性反抬脚的项一个不留、塑形只留一套相位逻辑;不在表里的一律不加 … feet_landing_vel(评审 #4):拖地时代价恒 0、抬脚才收费——与 v4-clearance/v8a-B 同属「反抬脚门槛」家族,步态未成形时是毒;v7④ 加它时步态已存在。… 已从 cfg 移除(置 None),不是权重 0 挂着。”
train/OMNI_V0_SPEC.md § 3. 最小奖励表 v2 / 明确不带 The advisor's "runtime five-stage state machine + per-stage reference poses + RL residual" appeared in none of the three papers it cited - reading the originals changed the plan and downgraded two widely repeated industry claims
advisor-paraphrase-vs-paperRead the primary source behind any piece of advice before adopting its architecture; record where the paraphrase and the original differ, downgrade claims the originals do not support to speculation, and adopt what the verified sources actually share.
Symptom
After the violent first real run, an advisor proposed re-architecting recovery as a runtime staged state machine with reference poses and an RL residual, citing HoST, HumanUP and StableMimic.
Context
The three papers were read in full on 2026-08-09 and tabulated (deployment form, what the "stages" really are, hard constraints, references). HoST: one end-to-end policy, height-gated rewards in training, action anchored as q + beta*a with a beta curriculum. HumanUP: two training stages with the same observation/action; the vendor's three-stage state machine is the baseline it beats (41.7% vs 78.3%); Stage II tracks an 8x slowed Stage I trajectory (4x too violent, 10x does not converge). StableMimic: a learned soft gate. On 08-10 more sources were checked the same way: the Agility page does not say Digit's self-righting was learned in simulation (only step recovery is stated as RL), so the claim was downgraded to speculation; HoST's support for the 12-DoF armless Mini Pi exists in its code repository, not in the paper text.
Change
Adopted only what the originals share: a hard action bound or anchored action space, strong smoothing including a second-difference term, a slowed hidden reference, heavy DR and real fallen states. The staged runtime state machine was not adopted.
Outcome
The next re-rooting candidates came straight from the verified material, and the beta-anchored action space (HoST, with the Mini Pi configuration as the nearest real-robot precedent) became V2, the lineage that later stood up on hardware.
Mechanism
A paraphrase compresses a paper into the advisor's own architecture; only the original shows what was actually deployed, what was a baseline, and which numbers came with which ablation.
Applies when
- an advisor, agent or summary proposes an architecture with citations
- an industry claim ("X learned it in sim") is about to justify a design
- several papers are cited for one combined recipe
“⇒ **顾问的核心形态"runtime 五阶段状态机 + 每阶段参考姿态 + RL residual"在三篇引文 里均不存在**,其中 HumanUP 还点名 state machine 是局限。”
git:Lucen-recovery@origin/recovery:train/RECOVERY_V0_SPEC.md § §26 三篇引文精读判决(2026-08-09 全文核对;顾问转述与原文有出入) After installing the measured plant, re-run all generations paired on old and new plant - identity metrics carry verdicts, physics metrics re-baseline
plant-swap-invariants-vs-shiftsTreat every plant upgrade as an era boundary: re-run the retained policy set paired (same seeds/flags) on both plants, carry forward only verdicts whose metrics proved plant-invariant, re-baseline the rest - and mine the systematic shifts as measurements of the old plant's biases.
Symptom
With the plant finally fully measured (weighed masses 9.792 kg, bench-identified armature, in-situ friction), no historical sim number was comparable to new runs - "历史 sim 数字跨纪元不可比" - and it was unknown which historical verdicts still held.
Context
The era-2c cross-test ran all seven walk generations on the complete plant under one harness (14/14 survived), then re-ran the same 14 configurations on the OLD plant retrieved from git, same flags and seeds, with a self-check (one historical record reproduced digit-for-digit). The split was clean. Policy-identity metrics moved essentially zero across the plant swap - dominant frequency (v7's period-doubling 1.20 -> 1.20), knee amplitude (9.9 -> 10.0), foot distance (+/-2 mm), saturation (100 -> 100, 0 -> 0) - so all seven cross-generation verdicts (freeze signature, saturation-line closure, knee-collapse location, slip-penalty accounting, N2 lineage, v6's balance, v11's triad) were re-confirmed on the honest plant. Plant-physics metrics shifted systematically with ordering preserved: slip down 10-25% (measured friction makes ground-twisting costlier), landing vertical velocity down 15-50% ("旧 plant 高估落地 凶度"第二次独立证实), landing force mixed (mass up 2.2% vs friction braking the swing - two effects fighting). Exactly ONE behavior-level change: v8's low-speed period-doubling vanished (1.30 -> 2.50, lift normalizing) - confirming it had been machine-dependent bifurcation-edge behavior that armature+friction push off the knife edge, while v7's period-doubling stood untouched: saturation-freeze-driven, a policy property, not a numerical accident.
Change
Era re-baselining protocol: after any plant upgrade, one paired same-seed sweep of all retained generations on old and new plant; verdicts keyed to identity metrics carry over, thresholds re-read against the new-plant table, and the differences themselves become plant-physics findings.
Outcome
Seven verdicts survived with evidence rather than assumption; two causes of the period-doubling family were separated with plant-side proof; and the sim's landing-violence overestimate was independently confirmed a second time.
Mechanism
A policy's structural properties (frequencies, amplitudes, frozen joints) are functions of its weights and survive plant changes; contact-mediated quantities are joint properties of policy and plant and shift when the plant becomes honest. Pairing seeds across plants isolates the plant's contribution exactly, so the sweep both validates history and measures what the old plant had been lying about.
Applies when
- installing measured masses/armature/friction into the sim
- historical thresholds are cited across a plant change
- a hardware-only behavior might be bifurcation-edge sensitivity
“策略身份指标逐位不动:主频(v7 1.20→1.20)、膝摆 … 这些是策略属性,plant 换代携带无损,历史定论因此全部成立。… 唯一行为级变化:v8@0.15 的倍周期消失(主频 1.30→2.50…)——印证当时"分岔边缘、机器相关"的判定:armature+摩擦把 v8 推离刀锋;v7 的倍周期纹丝不动(1.20→1.20),它是饱和冻结驱动的深层属性,不是数值巧合。”
train/README.md § 纪元 2c 全代同机横测 (2026-08-04): 完全体 plant 上历史结论全部存活 A curriculum ramp keyed to the process step counter re-fires on every resume - and shipped policies never saw the penalty
curriculum-counter-lineage-stepsKey every curriculum/ramp schedule to lineage-cumulative progress, not per-process counters; and audit where your shipped checkpoints sit relative to every ramp - a penalty that no product ever experienced is not part of your training.
Symptom
vx+0.30 died at a fixed relative time in every resumed run: resume at 500 -> slide at 1100, zero at 1300; resume at 700 -> slide at 1300, zero at 1500 - absolute depths offset by exactly the resume offset, relative timetable identical.
Context
ramp_reward_weight (the saturation penalty ramp) read env.common_step_counter, which restarts at 0 for every run including --resume. So start_step=600 meant "600 iters after THIS resume", not "lineage iteration 600". The A/B arm pair was the clean proof: their env.yaml differed only in log_dir, only resume point distinguished them, and the omni CurriculumManager had exactly one active term - nothing else could produce that timetable. Second consequence: every shipped checkpoint (s1e-500 at +500, C2-700 at +200, A800 at +100) was selected before its run's +600, so the saturation penalty weight was 0.000 for every product ever shipped - explaining saturation 33% and raw |action| 1.9 against clip 1.0 (hip_roll in bang-bang), i.e. half the heat budget.
Change
Two independent recommendations recorded: (1) make the ramp count lineage steps (add the checkpoint's iteration offset on resume) or pin terminal weights in downstream rungs instead of ramping; (2) give the saturation penalty its own rung - never mixed into a skill-learning rung (that would be two variables again).
Outcome
Explained the recurring +600 death of the highest-amplitude command and the persistent actuator saturation of all shipped products with one root cause; honest caveat booked (at +800 the weight is only -0.086, small, but vx+0.30 is the command demanding the largest action amplitude, so it is squeezed first).
Mechanism
Resumable training splits "the lineage" from "the process"; any schedule keyed to process-local counters silently re-applies its transient to every descendant run, and any product-selection habit that picks checkpoints early systematically samples the pre-ramp regime - the curriculum exists in the config but never in any shipped policy.
Applies when
- resumed/forked training with any scheduled reward or DR ramp
- a metric dies at a fixed offset after each resume
- shipped policies show behavior a late-schedule penalty should prevent
“ramp_reward_weight 读的是 env.common_step_counter，它每个 run 从 0 开始，--resume 也不例外。… 原始 run / 臂B | 500 | 1100 = +600 | 1300 = +800；臂A | 700 | 1300 = +600 | 1500 = +800 … 所有出品其实从没见过饱和罚。… 这解释了 sat_max_pct 33%、raw |a| 最大 1.9（clip 是 1.0）—— hip_roll 一直在 bang-bang，而罚它的那一项权重恒 0。热账的一半在这里。”
train/C_LADDER_RUN.md § 3g. 系统性问题：saturation_ramp 每次 resume 归零 After seven patch-generations, freeze the lineage as a regression baseline, fix the structural debts, and retrain from zero
freeze-lineage-fix-structure-restartWhen successive rungs keep trading one symptom for another, ask whether the remaining problems are structural (contracts, latency, sampling, reward-table architecture); if so, freeze the lineage as regression baselines, pay the structural debts, and restart minimal - carrying forward laws and instruments, not weights and weights' patches.
Symptom
The v5-v11 walk lineage had accumulated interacting patches (reward terms, gates, clamps, per-joint scales) faster than it converged on the user's goal; v12's spec itself was superseded before training by an external review's verdict that the remaining problems were structural, not parametric.
Context
The 2026-08-05 status banner records the pivot: the walk profile was rolled back wholesale to v10b parameters, the v5-v11 lineage frozen "只作回归对照" (kept only as regression baselines), and four structural debts were named as prerequisites for a from-zero straight-walk baseline: the action-latency FIFO (fixed with its own test), the ONNX manifest contract, discrete command sampling, and a minimal reward table. The v12 spec - fully designed, partially implemented - was suspended: "本规格挂起,不再按此开训".
Change
Strategy switched from "one more patch generation" to freeze-fix-restart: lineage checkpoints retained as comparison anchors, infrastructure hardened first, then a clean retrain with a minimal reward table (this restart produced the s* generation that later became the real-robot SOTA line).
Outcome
A designed-and-ready training generation was deliberately not run - the review's structural findings outranked sunk design cost; the restart line inherited seven generations of laws (calibrations, gate batteries, falsified fixes) without inheriting their entangled reward table.
Mechanism
Patch lineages accumulate coupled terms whose interactions eventually cost more to reason about than a restart costs to train; the knowledge worth keeping is the laws and instruments (measured plant values, calibrated gates, falsified directions), not the entangled weights. A restart on hardened structure converts the lineage's lessons into a clean initial design instead of another delta.
Applies when
- repeated rungs shuffle symptoms without net progress
- an external review flags infrastructure/contract debts
- deciding between another patch generation and a clean retrain
“同日外部评审定调换路线:冻结 v5~v11 血统(只作回归对照),修结构性问题(latency FIFO 已修 tests/test_action_latency.py、ONNX manifest 契约、离散命令采样、最小奖励表)后从零训直行基线。本规格挂起,不再按此开训。”
train/WALK_V12_SPEC.md § ⚠️ 状态 (2026-08-05) When training fails repeatedly, inject the target behavior open-loop - stop tuning rewards for an unverified behavior
open-loop-probe-before-reward-tuningAfter two failed training attempts at a skill, stop training: demonstrate the behavior open-loop on the real plant/sim first, and only resume training once you hold a measured, safe, sign-verified target trajectory.
Symptom
Three sidewalk training rounds failed; hypotheses multiplied (exploration failure / wrong reference waveform / insufficient authority) with no way to pick between them by running more training.
Context
Instead of a fourth reward guess, the team wrote probe_side_ref.py: the candidate reference is injected open-loop on top of a frozen policy's output (bypassing PPO entirely), directly measuring "what happens if the robot literally does this waveform" - separating all three hypotheses in one experiment (5 seeds x 8 s per condition, several waveform families and gains). The probe immediately eliminated the authority hypothesis (full-amplitude execution, 5/5 survival) and localized the problem to the waveform/measurement side. The closing principle was written down after the saga: without a verified target behavior, tuning rewards is "在黑暗里试钥匙" (trying keys in the dark).
Change
Standing method: before opening another training rung on a failing skill, build an open-loop (or task-space) generator of the intended behavior, measure whether the physical system can express it and what it looks like - then train toward a verified, quantified target.
Outcome
The probe chain produced the verified waveform (reversed-sign triangle, half gain), quantified safe amplitude (tilt 8.2 deg at band top, foot distance clear of the wall), exposed the metric bug when probe and training disagreed, and ultimately supplied the feed-forward that made C4 pass in +100 iters.
Mechanism
Training couples exploration, reward design, and feasibility into one opaque outcome; open-loop injection cuts the loop and tests feasibility and waveform alone. A behavior demonstrated open-loop converts the remaining failure into a pure credit-assignment/reward question - and its measured trajectory becomes the reference itself.
Applies when
- repeated training failures on one skill with multiple live hypotheses
- uncertainty whether the platform can physically express the behavior
- a reference trajectory's shape/sign/amplitude is guessed, not measured
“三轮 FAIL 之后不再猜，写 train/probe_side_ref.py 把参考开环注入到策略输出之上（绕过 PPO），直接量「照这个波形做会怎样」，一次分开三个假说：甲 探索 / 乙 波形 / 丙 权限。”
train/C_LADDER_RUN.md § 3f. C4 真因定谳（开环探针） / 3k. 建议的下一步 Every power cycle starts with the same read-only pre-flight - read the buses, check the torque limits against 12/17/11, verify the IMU axes, check the ports after any new USB device - and any reassembly re-measures the joint zeros
power-cycle-preflightStart every powered session with a fixed, read-only pre-flight - bus responses, torque limits equal to the simulated ones, IMU axes, device identities - and re-measure joint zeros after any mechanical reassembly before running a policy.
Symptom
Hardware state drifts between sessions in ways no policy can see: a motor that stops answering after a power cycle, a torque limit that differs from the one simulated, an IMU axis flipped, two USB devices swapping identities, a joint zero moved by reassembly.
Context
The runbook's session order before any policy runs: read every motor on both CAN buses without enabling them (the first command after every power cycle); set_torque --check, all twelve motors must read 12/17/11 N*m, and any difference is written back; imu_reader --verify-axes, where the operator tilts the robot forward and to the right and every check must pass before continuing; check_ports after plugging in any new USB device (the IMU and a CAN adapter once collided on USB identity). After re-mounting motors: read the buses, then re-measure the calibration offsets (three repeats, written back) - "skipping it means running everything on the wrong zero". Hanging checklists repeat the torque-limit check (the deploy script also self-checks at start).
Change
A fixed, read-only pre-flight run in the same order every session.
Outcome
The runbook records one earlier hardware check in the same spirit: all 12 motors' implied kp fell within 18.4-22.0 for a commanded 20, inside the kp randomization range used in training.
Mechanism
A policy transfers only if the plant matches the one it was evaluated on; the pre-flight turns silent hardware drift into a failed check before the robot moves.
Applies when
- the first command after powering a robot on
- after swapping adapters, cables or motors
- a policy that worked last session suddenly behaves differently
“python tools/set_torque.py --check # 12 颗应全对 12/17/11, 有 diff 就 --write … 插任何新 USB 设备后都先跑一次 check_ports.py(IMU 和 CANable 的 USB 身份撞过车) … python tools/calib_stance.py --repeat 3 --write # 重标 offset —— 8/9/10 重新装, 机械零位变了”
RL系统/FOLLOW THIS copy 2.md § WALK / STAND 每次开始前 / 换CAN / 装回后必做两件 Training-log reward values and fixed-command eval values live on different distributions - comparing them once claimed a 44% improvement that was really 6-10%
same-distribution-reward-comparisonQuote reward-term values only with their distribution attached (command range, DR on/off, environment), and compare across runs only when those match; re-measure in a common environment before claiming any improvement percentage.
Symptom
A v6-era analysis concluded slip had dropped 44% by comparing the training log's Episode_Reward against values calibrated in a fixed-command play environment; a same-condition re-measurement showed the true improvement was 6-10%.
Context
The training log's reward is an expectation over the training command distribution (vx 0.15-0.5, yaw +/-0.6, with pushes and domain randomization); play-environment calibrations are taken at a single fixed command with DR off. Subtracting one from the other compares apples to oranges - the warning was written into the v7 pre-flight: "奖励数值只能在同一指令分布下比较 … 拿它和固定 cmd 的 play 环境标定值相减会得出错误结论(v6 那轮已经栽过一次)".
Change
Rule adopted: any before/after reward-term comparison must hold the command distribution, DR state, and evaluation environment fixed; training-log values compare only against training-log values of runs with identical command/DR configs.
Outcome
The phantom 44% improvement was retracted; later term-level accounting (e.g. the C4 ignore-floor work) consistently specified its distribution before quoting numbers.
Mechanism
A reward term's expectation depends on the visited-state distribution as much as on the policy; changing the command distribution or DR moves every term's baseline. Cross-distribution differences therefore measure the distributions, not the policy change.
Applies when
- comparing reward telemetry across training runs or vs play evals
- claiming improvement percentages from training logs
- term-level reward accounting for diagnosis
“奖励数值只能在同一指令分布下比较。训练日志的 Episode_Reward 是在训练指令分布上算的(vx 0.15~0.5 / 偏航 ±0.6 / 带推力与域随机化), 拿它和固定 cmd 的 play 环境标定值相减会得出错误结论(v6 那轮已经栽过一次: 据此以为滑移降了 44%, 同条件对拍只有 6~10%)。”
train/WALK_V7_SPEC.md § 3. 开训自查 ⚠️ The smoke watcher panicked at the wrong operating point and its score goes blind when gates saturate - treat it as a survival sentinel, not a judge
smoke-watcher-operating-pointConfigure every automated evaluator at the lineage's declared deployment operating point, give it graded metrics that cannot saturate, and until then scope its authority to catastrophe-detection - never let a mis-configured watcher stop or rank a rung on its own.
Symptom
Two watcher misfires in one ladder: (1) during the friction rung the watcher (evaluating at kd 1.0) reported panic-level 1/3 survival from iter 3300 - falsified by the official kd 1.2 scan, because the lineage's design operating point was kd 1.2 and the watcher lacked the --kd-scale passthrough; (2) during the PD rung the watcher's early-stop score froze at iter 1050 despite ongoing drift improvements, because with all eight gates passing (constant 0/3 failures) the score has no gradient left - "八门全过恒 0/3 时 score 对漂移改善盲, s1f 课文三现".
Context
Both are the same category: the in-training smoke loop is an instrument with its own configuration (operating point, score design), and its verdicts are only as aligned as that configuration. The booked doctrine: "冒烟只当存活哨兵" - until the watcher evaluates at the deployment operating point with graded metrics, its role is detecting catastrophes, not ranking checkpoints; ranking belongs to the full battery at the design operating point (and the watcher's scoring was separately patched to weight survival 3x so recoveries during hard phases are not early-stopped away).
Change
Watcher debt booked (--kd-scale passthrough); score saturation acknowledged with graded columns planned; selection authority kept with 20-seed batteries at the declared operating point.
Outcome
A false panic did not abort a rung that was actually passing at its design point; a frozen score did not hide real drift gains; the instrument's authority was scoped to what its configuration can actually see.
Mechanism
An evaluator is itself configured (gain profile, delay, metrics); evaluating a policy away from its design operating point measures a counterfactual robot, and bounded scores saturate once binary gates pass, losing all sensitivity. Instruments need the same operating-point discipline as deployments and graded outputs to retain gradient.
Applies when
- an automated smoke loop contradicts the official battery
- early-stop scores freeze while graded metrics still improve
- lineages with non-default deployment gain/delay profiles
“watcher (kd1.0 口径) 3300 起 1/3 恐慌被 kd1.2 正式扫描证伪为考纲外假象 —— 工作点评测口径教训: watch_ckpt 缺 --kd-scale 透传 (待补), 冒烟只当存活哨兵。… watcher score 饱和误停 @1050(八门全过恒 0/3 时 score 对漂移改善盲, s1f 课文三现)”
train/README.md § omni_s2e_fric (watcher 恐慌被证伪) / omni_s2e_pd (500 臂) A constant-value plant rung passed every binary gate with record scores - and shipped 60% thinner posture margins that hardware exposed
constant-value-dr-overfits-marginRandomize deployment-critical axes over a narrow band spanning the measured real support - never a single value, never a fictitious tail - and report graded margin quantities (tilt margin) next to binary gates, because saturated gates rank thin-margin and thick-margin policies identically.
Symptom
s2_lag1 (trained at constant 1-frame latency) posted the strongest sim gate sheet in history (20/20 everywhere) yet was unstable on hardware, while s1e (trained across the full 0-3 frame band) was the every-run-stable SOTA at the same power.
Context
The sim autopsy (new --delay-jitter harness modeling the BusWorker's time-varying phase drift): 18 runs across constant and time-varying delays ALL survived - time variation alone does not kill - but the tilt-margin ordering reproduced hardware exactly: s1e 7.7-9.1 deg (thickest) < s2_lag1 10.7-15.0 < s1c 16.2-18.2. Attribution: constant-value training permits precise specialization to that one value; s1e's band diversity forced cross-value robustness - "恒定 1 帧训练 vs s1e 的 0~3 帧全带——分布多样性逼出跨值鲁棒,恒定值允许精确 特化" - so the constant-rung policy's margins were ~60% thinner, fine in sim's clean world, pushed over the line by real-world disturbances. Tool lesson booked: "存活门二值饱和后掩盖裕度差" - binary survival gates saturate and hide margin differences; graded margin columns (tilt-max) belong in the report. The synthesis with the opposite failure (wide tails cause drag-glide): the proposed resolution was a NARROW uniform band (0.02, 0.04) covering exactly the real 1-2 ticks - diversity inside the measured support, no tail, no single point. s1e's root selection later leaned on the same property: its full-band latency training "预装" the delay rungs and delivered "全工况稳定裕度" that survived power derating.
Change
DR-on-an-axis design refined to a three-way distinction: no wide fictitious tails (drag), no single constant values (thin margins), but a narrow band spanning the measured real support; acceptance reports gained graded margin columns alongside binary gates.
Outcome
The tilt-margin column entered the standard report; the s1e root (band-trained) carried the C ladder while the constant-value branch was archived with its three contributions credited.
Mechanism
Robustness margins are shaped by the diversity of the training distribution, not just its support: a point-mass distribution lets the optimizer trade margin for on-point performance, while a band forces solutions that keep margin across the band - and binary survival metrics cannot see the difference until the margin is spent on hardware.
Conflicts
The narrow-band (0.02,0.04) resolution was a pending recommendation ("裁决建议(待用户)") at the time of writing; the lineage instead moved root to s1e whose full-band training predated the staged ladder - the deterministic-staging card and this card record the two failure modes the final design must avoid simultaneously.
Applies when
- a rung trained at a fixed plant value aces sim but wobbles on hardware
- binary acceptance gates are all saturated across candidates
- choosing between constant, banded, and wide DR on one axis
“18 跑全活,时变性单独不足以击杀;但 tilt_max 裕度排序完整复现真机:s1e 7.7~9.1°(最厚)< s2_lag1 10.7~15.0 … 恒定 1 帧训练 vs s1e 的 0~3 帧全带——分布多样性逼出跨值鲁棒,恒定值允许精确特化 … 存活门二值饱和后掩盖裕度差(s2_lag1 sim 门 20/20 史上最强却真机不稳)”
train/README.md § s2_lag1 真机不稳 × s1e 稳的 sim 对拍(2026-08-07,时变延迟实验) The trainer read a stale USD after the URDF mass update - regenerate derived assets and gate on an automated equality instrument
derived-asset-staleness-checkFor every derived plant artifact (USD from URDF, generated value files), pair the generation step with an automated source-vs-derived equality instrument, prove the instrument can fail, gate training on its PASS, and re-run physical audits after every regeneration.
Symptom
Measured link masses had been committed to URDF/MJCF (total 9.58 -> 9.792 kg, weighed values), but Isaac reads the derived USD asset - which still carried the old masses: a silent 2.2% mass fork between the training plant and the evaluation plant.
Context
The v12 checklist made USD regeneration a hard precondition ("硬性 前置") and, crucially, backed it with an instrument: check_usd_mass.py compares USD vs URDF per-link mass AND inertia trace, validated by showing it FAILs the stale asset naming 9 offending links, then PASSes after re-conversion (13/13 links consistent, total 9.7920). The self-collision filter audit was re-run after the regeneration (three poses, 0.00 N) because a regenerated asset invalidates physical audits done on the old one. This milestone was also where the three plants first aligned: "三边 plant 首次对齐(armature+摩擦+ 实称质量)就在这一代".
Change
convert_urdf re-run on the training machine, regenerated asset committed, check_usd_mass.py PASS required as an acceptance gate for the generation; dependent audits repeated post-regeneration.
Outcome
The 2.2% plant fork was closed before it could distort a generation's acceptance numbers; the staleness class of bug now has a permanent detector instead of a memory.
Mechanism
Source-of-truth edits do not propagate to derived binary assets by themselves; any consumer reading the derivative silently trains or evaluates on the old plant. An automated equality check between source and derivative - proven able to fail - turns an invisible staleness into a red gate, and regeneration invalidates every audit performed on the old artifact.
Applies when
- editing masses/inertia/geometry in URDF or MJCF sources
- a trainer or evaluator consumes converted/derived assets
- plant numbers differ between simulators for no visible reason
“25ba997 把连杆质量更新为实称值(总重 9.58→9.792 kg,URDF/MJCF 已改),但 Isaac 读的是 train/assets/laika_v2.usd —— 仍是旧质量。… 否则 Isaac(9.58)与 MuJoCo(9.79)质量分叉 2.2%,v12 验收数字失真。验收门:python tools/check_usd_mass.py 必须 PASS … 对旧资产实测 FAIL/9 连杆点名,仪器已验证”
train/WALK_V12_SPEC.md § 7. 核查单 (⚠️ 先重转 USD) An auto-curriculum ratchet capped out at iter 248 and never engaged - stage difficulty manually or verify engagement
auto-curriculum-engagement-checkPrefer manually staged difficulty with gated transitions; if you use an automatic curriculum, instrument its internal state and alarm when it stops engaging - a saturated curriculum is constant DR wearing a curriculum's name.
Symptom
A curriculum mechanism intended to grow difficulty adaptively (s1f's ratchet) hit its cap at iteration 248 and never bit again - for 96% of the run its effect was equivalent to constant DR, i.e. the curriculum existed in name only.
Context
When external advice suggested graded wz bands (start ±0.15, then ±0.30), the team agreed with grading but explicitly rejected automatic curriculum, citing the s1f episode. The same logic had already been paid for with push grading: ±0.6 failed twice, ±0.3 was feasible - grading matters, but the grade transitions were made by hand at verified checkpoints.
Change
Ladder policy: difficulty staged manually, one band per rung, each transition gated by the acceptance battery; automatic ratchets not used unless their engagement is monitored and demonstrated.
Outcome
Every C-ladder band change (wz ±0.12-0.25 first, wider later) was an explicit, attributable rung; no silent constant-DR-in-disguise runs recurred.
Mechanism
Adaptive curricula couple their own state machine to noisy training metrics; a ratchet that saturates early stops adapting but keeps its name, so the operator believes difficulty is progressing when it is frozen. Manual staging costs more decisions but each decision is observable and reversible.
Applies when
- choosing between auto-curriculum and staged bands for a new skill
- a curriculum's difficulty parameter plateaus early in training
- post-hoc attribution of what difficulty a lineage actually saw
“C2 的 wz 分级（±0.15 → ±0.30，不要一上来 ±0.6）—— 与我们付过学费的 push 分级同形（±0.6 两轮 FAIL，±0.3 才可行）。但必须手动分级，不做自动课程 —— s1f 的课程化棘轮 iter 248 封顶未咬合，96% 时长等价常量 DR”
train/C_LADDER_RUN.md § 1. 采纳 3 条 (C2 的 wz 分级) A single-signal contact detector lied in both directions - foot height flagged 40% false flight on a walking gait, contact force alone flagged false flight during low-friction slip - so flight became force < 5 N AND sole height > 5 mm
contact-detector-single-signal-liesDefine contact and flight events from two independent signals (force and geometry) in conjunction, validate the detector on a behaviour known not to contain the event before using it as a gate, and match the trainer's threshold when comparing across simulators.
Symptom
The run line needed a flight-fraction gate. The first MuJoCo version, "sole higher than 2 mm", measured 40% flight on a walking policy that never flies. Five weeks later the one-leg gate, using contact force alone, reported support-foot "flight" segments at mu 0.4 for a policy that was not hopping.
Context
During a walking step the toe lifts or the heel strikes with the foot pitched, so the ankle-roll origin rises a few mm while part of the sole still touches - height alone calls that flight. Switching to contact force < 5 N (the same threshold Isaac's contact reward uses) zeroed the false flight on walking. In the one-leg re-test every force-only "flight" segment had a measured sole height of 0.0 mm: the normal force chattered while slip corrections played out on low friction.
Change
Run line: flight = contact force < 5 N ("lift-off must be judged by contact force"). One-leg line (2026-09-16): flight = force < 5 N AND sole height > 5 mm, recorded as the same measurement lesson in the opposite direction; the gate's behavioural meaning was unchanged.
Outcome
With force-based detection the walking policy read 0.0 flight and the run policy's zero flight was confirmed by two plants; with the conjunctive definition the one-leg support-foot gate stopped reporting false hops.
Mechanism
Each signal has its own failure: geometry moves without leaving the ground (foot pitch), and forces drop without leaving the ground (slip chatter); requiring both removes both families of false positives.
Applies when
- writing a flight, lift-off, hop or slip detector for a gate
- a gate reports an event the video does not show
- reusing a detector on a different gait or floor friction
“首版用**足底高度>2mm** 判离地, 在 omni_s1e 走路策略上测出 40% 假腾空 … 改用**接触力 <5N**(与 Isaac feet_contact_number 同源阈值)后 空检归零 (walk 策略 flight_frac 0.0)。**课文: 离地判定必须用接触力, 高度判 会把脚的俯仰当腾空**”
train/README.md § run R1 立项 (2026-08-09): Mac 侧新工具 + 一次空检抓获 An exponential kernel on instantaneous velocity punishes gait oscillation - track the cycle average
cycle-average-tracking-for-gait-quantitiesReward velocity tracking on gait-cycle averages (or filtered values), not instantaneous samples, whenever the desired behavior oscillates at stride frequency; widening the kernel does not fix a variance penalty.
Symptom
Even while the robot genuinely sidewalked (verified after the metric fix), the Isaac-side tracking reward sat on the ignore-floor: true sidewalk scored 0.178 vs 0.189 for ignoring the command - the reward was mildly punishing the desired behavior.
Context
Sidewalking is inherently oscillatory: per-frame vy std was 0.177 while the tracking kernel width was sigma = 0.15, applied to the instantaneous value. A kernel-width scan showed widening sigma 0.15 -> 0.50 still loses (-0.14 -> -0.07): "指数核惩罚的是方差，而侧步天生带方差" (the exponential kernel penalizes variance, and side-stepping inherently carries variance). Modeling with measured parameters: replacing instantaneous vy with the mean over one gait cycle (0.5 s) flips the margin decisively (true sidewalk 1.888 vs ignore 1.281, +0.607), half-cycle is neutral (+0.006), two cycles adds nothing more. Explicitly flagged as extrapolation pending Isaac-side implementation. This also vindicated a previously dismissed external note (sigma too small) - right conclusion, different mechanism than claimed (variance, not gradient).
Change
Proposed fix recorded: change the tracked quantity from instantaneous vy to a one-gait-cycle running average; widening sigma alone rejected by the scan.
Outcome
Diagnosis complete and quantified; the C4 product shipped via feed-forward before the reward change was implemented, so the cycle-average fix remained a verified-by-model, not-yet-trained change.
Mechanism
E[exp(-(v-c)^2/sigma^2)] decreases with Var(v) even when E[v] = c exactly; a gait's phase-locked oscillation guarantees variance at the stride frequency, so instantaneous tracking rewards structurally prefer standing still at the command mean. Averaging over exactly one cycle removes stride-frequency variance while preserving command-following error.
Conflicts
The cycle-average fix itself is model-extrapolated ("⚠️ 这一条是外推，须在 Isaac 侧实装并复量后才能当结论") - the diagnosis is measured, the remedy untested in training at the time of writing.
Applies when
- tracking rewards for lateral/turn/any oscillation-carrying velocity
- a verified behavior scores below the ignore-floor
- choosing sigma for exp-kernel tracking terms
“侧走时 vy 的逐帧摆幅 std = 0.177，而 track_lin_vel_y_exp 核宽 σ = 0.15，且作用在瞬时值上 … 真侧走（均值 66%，振荡 ±0.18）0.178 | 完全无视指令 0.189 … 真侧走的得分比无视指令还低。… σ 从 0.15 放到 0.50，侧走仍然吃亏 … 把跟踪目标从瞬时 vy 换成一个步态周期（0.5 s）的平均 vy：… 1.888 vs 1.281”
train/C_LADDER_RUN.md § 3n. 二/三 Isaac 训练奖励为何一直坐在「无视底分」/ 修法不是放宽 σ Size each joint's action authority to its measured working range - lock a channel to zero only when its job is provably elsewhere
per-joint-action-scale-lockdownSet per-joint action scales from measured target ranges and momentum decompositions: full authority for working channels, working-range authority for balance channels, zero for channels whose contribution is measured negligible - and prefer this structural quieting over perpetual reward penalties, keeping the contract dimensions intact.
Symptom
"Walks crooked" - roll and yaw channels wandered; hip_yaw peak-to-peak reached 22.3 deg in v11 while contributing essentially nothing to locomotion; a uniform action scale of 0.5 gave every joint the same authority regardless of its actual job.
Context
The v12 design replaced the scalar action scale with per-joint scales justified by measurements: pitch-class 0.5 (the gait's entire working channel - untouched); hip_yaw 0.0 - lossless because it carries only 1.4% of yaw momentum (v6 decomposition) and straight-line targets use merely +/-0.006-0.03 ("乱动纯属浪费"); roll 0.2 - NOT zero, because lateral balance and weight transfer are roll's unique job (locking it would degenerate into foot-edge rocking, "比现在更歪"), and 0.2 covers the measured working range +/-0.06-0.19 while "0.5 的另一半全是 '歪歪扭扭'的来源". Costs were accepted consciously: turning demoted to an observation row with a fallback (yaw 0 -> 0.1 in v13). Contract preserved: the 12-dim action interface unchanged, yaw values simply neutralized. The structural lockdown also RETIRED the reward-side hip_yaw_quiet penalty - "A 的 yaw=0 结构性取代,不再付奖励塑形成本".
Change
action_scale_joint pitch 0.5 / roll 0.2 / yaw 0.0 wired through robot.yaml -> policy_io (verified: all-ones action gives hip_yaw target exactly 0) -> Isaac action term, guarded by the contract checker ("它就是抓这种双侧不一致的").
Outcome
Designed and verified on the shared side before the lineage freeze; stands as the pattern for authority sizing: structure replaces reward shaping wherever a channel should simply not act.
Mechanism
Action scale is a per-channel authority budget; uniform budgets give noise channels the same voice as working channels, and reward-side quieting then pays a permanent shaping tax for what a zero scale provides for free. But zeroing is only lossless when decomposition proves the channel's contribution negligible AND no unique function (balance) lives there.
Conflicts
Wired and verified on the config/deploy side but never trained - the 2026-08-05 reset suspended v12 before the Isaac-side run.
Applies when
- some joints wander without contributing to the task
- a quieting penalty (deviation/L1) taxes every step forever
- deciding action-space authority for a new task or robot
“yaw=0 是无损的:实测它只贡献 1.4% 偏航动量、直行目标只 ±0.006~0.03,乱动纯属浪费。… roll 不能为 0:横向平衡/重心换脚是它的独有职责,锁死会退化成脚缘摇摆(比现在更歪)。0.2 的依据:各代实测 roll 目标只用 ±0.06~0.19,0.5 的另一半全是"歪歪扭扭"的来源。”
train/WALK_V12_SPEC.md § 3. A —— 逐关节动作幅度(用户"只动 pitch"的安全版) PPO's Gaussian noise cannot compose phase-locked oscillations - deliver them as feed-forward and let the policy learn the residual
feedforward-for-phase-locked-skillsIf a skill needs a temporally coherent (phase-locked) action component, do not expect step-wise exploration to find it: inject a verified feed-forward and train the policy as a residual stabilizer, keeping the feed-forward inside the deployment contract.
Symptom
Four different reward arrangements (no reference / wrong-sign reference / correct-sign reference / cage released) all failed to elicit sidewalk, while open-loop probes proved the behavior existed and was safe on the same platform with the same policy as base.
Context
Producing lateral velocity requires a phase-locked hip_roll oscillation synchronized to the gait clock. PPO's exploration is per-step, zero-mean, uncorrelated Gaussian noise - it can never compose a sustained phase-locked component, so the behavior is unreachable by exploration regardless of how it is rewarded. The fix changed the delivery channel: target = default + scale*action + lat_ff(cmd_vy, phi). The policy's action becomes a residual on top of the feed-forward, retaining full balance authority (it can even cancel the feed-forward); the feed-forward supplies exactly the component exploration cannot. This mirrors why the sagittal joint_pos_ref worked (it also delivered phase structure), just via a different channel.
Change
Contract-level change, done cleanly: new profile omni_ff (= omni + lat_ff_gain -0.5), existing omni profile bit-identical; feed-forward applied after the action delay stage; missing cmd/phase raises instead of silently dropping; deployment must use the same phi as build_obs (recomputing gives a one-tick phase misalignment).
Outcome
From C2-700, +100 iterations sufficed: product omni_c4_ff800 scored vy +120%/+125% (from +4%/-1%), 260/260 cells at 20/20 survival, zero old-skill regression, left/right gap 5 pp - the entire C4 saga resolved by changing the delivery mechanism, not the reward.
Mechanism
Exploration noise spans only the subspace its correlation structure can express; skills requiring coherent oscillation lie outside the span of i.i.d. per-step noise. Feed-forward moves the required structure into the action pipeline where it needs zero probability mass to appear, reducing the learning problem to stabilizing around a demonstrated behavior - which PPO does well.
Applies when
- a periodic/oscillatory skill trains flat under every reward variant
- open-loop injection of the behavior already works
- considering GRU/curriculum/exploration tricks for a rhythmic skill
“病因不在奖励，在探索形式：产生侧向速度需要相位锁定的 hip_roll 振荡，PPO 的逐步高斯噪声零均值无相关，合不出相位锁定分量。… target = default + scale·a + lat_ff(cmd_vy, φ)。策略动作因此是前馈之上的残差，保留全部平衡权限”
train/C_LADDER_RUN.md § 3j. C4-redo4：唯一变量 = 侧步参考改为前馈注入（契约级） Knee swing collapsed because it directly trades against the slip penalty - price the conflict explicitly and clamp what reward cannot hold
knee-swing-vs-slip-pricingWhen a behavior collapses as another metric improves, look for the term pair trading them and set their price ratio deliberately (with escalation and reverse tripwires pre-registered); where the policy actively spends action budget to undo your target, stop paying more reward and clamp the target space structurally.
Symptom
Knee peak-to-peak swing collapsed across generations - v5 33 deg, v10 26-30, v10b 7-8, v11 6.5-8.6 - and rolling the clock back did not recover it, acquitting the clock; the collapse tracked the gated slip penalty instead: "屈膝与不打滑在当前奖励里直接对抗" - v10b's excellent 93 deg slip was purchased with knee amplitude.
Context
Reward-side flexion fixes had failed three times: raising reference amplitude backfired twice (v9/v11), and v11's deep-squat default was actively fought by the policy - it spent 0.68 of action budget pulling the squat straight ("被策略花 0.68 动作拉直反杀"). v12's design accepted the conflict as real and attacked on two tracks: (1) ECONOMICS - a direct knee_swing_amplitude reward (+0.3, target 0.55 rad, capped at 0.6/step = 55% of tracking), explicitly opposed to the slip penalty by design ("显式对立——这正是设计:v12 就是这场对抗的定价实验"), with an escalation ladder (K +0.3 -> +0.5, then slip -0.5 -> -0.3, one layer at a time) and a reverse tripwire (slip telemetry back at v10 levels -> slip weight to -0.8, accept ~20 deg knee compromise); (2) STRUCTURE - knee target bounds [0.2, 0.9] rad so full straightening is physically impossible (straightest 11.5 deg) and the 0.68 fighting budget is released. A bonus falsifiable prediction was attached: phase-lock strength tracks amplitude (v9_probe 48 deg locked 2.5 Hz; v11 low-amplitude 1.36 Hz unlocked), so if K works, hardware phase-lock should return - one change, two verdicts.
Change
knee_swing_amplitude reward + knee target clamp + pre-registered escalation/reverse levers; the failed reward-side-only approach retired.
Outcome
The lineage was frozen before v12 trained (strategic reset), but the diagnosis stands as the walk line's clearest example of two reward terms trading a behavior between them, with the pricing experiment and structural clamp fully designed and calibrated.
Mechanism
When two terms price opposite aspects of one motion (swing amplitude creates yaw momentum that becomes slip), the optimizer settles wherever the price ratio puts it - patching one side moves the equilibrium, not the conflict. Explicit pricing makes the trade a designed quantity; structural clamps remove the regions where the policy spends budget fighting the designer.
Conflicts
The pricing experiment (K vs slip) was designed and calibrated but never trained - the 2026-08-05 reset suspended v12; the collapse attribution table and the 0.68-action counterattack are measured, the remedy's效果 is untested.
Applies when
- one gait quality degrades in lockstep with another's improvement
- the policy visibly fights a default pose or reference
- repeated reward-side fixes for the same behavior have failed
“膝摆塌在 v10→v10b,头号嫌疑是门控滑移罚(四代实测膝 p2p:v5 33° / v10 26~30° / v10b 7~8° / v11 6.5~8.6°;退时钟没救回 → 非时钟)——"屈膝"与"不打滑"在当前奖励里直接对抗 … 奖励侧修屈膝已三败 … v11 深蹲 default 被策略花 0.68 动作拉直反杀”
train/WALK_V12_SPEC.md § 0. 定位 / 2. K —— 膝摆经济(与滑移罚的对偶)