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
27 cards tagged sim2sim.
A binary reward band on the swing knee had zero gradient everywhere below it, so the one-leg policy parked in an unloaded "fake touchdown" that Isaac's 5 N threshold scored as success and MuJoCo showed as real pressing - a capped constant-gradient ramp, retrained from scratch, passed 40/40
binary-band-reward-fake-touchdownShape approach-to-target rewards as capped ramps with gradient from the starting posture, never as bands or indicators; and compare contact-based terms across simulators, because a policy riding just under a force threshold looks perfect in one and wrong in the other.
Symptom
At iteration 1,000 of the first one-leg run the swing foot never lifted: the policy stood with the "raised" foot resting lightly on the ground. In Isaac the contact-match term paid 96% of full marks; the same policy in MuJoCo pressed that foot on the ground for 450 frames.
Context
The swing-leg goal was "shank folded fully back" (knee 1.5-1.95 rad), rewarded as a binary band: +0.8 inside [1.5, 1.95], zero elsewhere. From knee 0.05 to 1.5 rad the term was flat. Contact is judged at a 5 N force threshold, so a foot carrying less than 5 N counts as lifted. The walk line had hit the same disease with a binary indicator (v4) and fixed it with a capped ramp (knee_swing_amplitude).
Change
swing_knee_fold changed from the binary band to a ramp clamp(|q|/1.5, 0, 1) - a constant gradient capped near 86 deg - and the policy was retrained from scratch (V0r1). After the first real-robot try showed the fold still too low, its weight went 0.8 -> 2.0 (V0.1).
Outcome
V0r1 model_2300 passed the full acceptance 40/40 (swing knee 1.72 rad, about 98.5 deg) and was stamped as oneleg_v0.onnx; the cross-simulator disagreement is recorded as the thing that caught the cheat.
Mechanism
A reward that is flat until the target is reached gives no gradient to approach it, so the policy settles for the nearest state other terms reward - here, a foot that satisfies the contact threshold without lifting; a second simulator with different contact force resolution exposes such threshold-riding.
Applies when
- rewarding a posture target with an in-band / out-of-band indicator
- a contact threshold decides whether a foot counts as lifted
- trainer-side contact terms are near full marks while the video looks wrong
“初版二值带 [1.5,1.95] 在膝 0.05→1.5 全程零梯度，策略停在"卸力虚点地"（Isaac 5N 阈下 contact_match 96% 满分 / MuJoCo 同策略 450 帧实压——跨仿真器互证抓作弊）；v4 二值指示同型病，按 knee_swing_amplitude 判例改常数梯度封顶 ramp，从零重训 … **oneleg_v0.onnx = V0r1 model_2300, 40/40 PASS**”
git:Lucen V2@origin/oneleg-line:train/ONELEG_V0_SPEC.md § §4 奖励表 swing_knee_fold 行 / §8 核查单 5 mj_objectVelocity returns inertial-principal-axis frame - one API assumption poisoned eval and observations for a whole line
body-frame-velocity-api-auditVerify every frame-sensitive API against a hand-computed truth (rotate raw qvel yourself, or command a known world velocity and check where it lands) before trusting any evaluation or observation built on it - especially when a model's inertial frame is rotated from its body frame.
Symptom
Sidewalk vy read ~0 under every condition; separately, whole-policy performance was mysteriously mediocre in sim2sim while training-side numbers looked fine. Four training rungs were declared FAIL partly on these readings.
Context
base_link's URDF inertial frame is rotated 90 deg about x relative to the body frame (iquat = [0.7071, 0.7071, 0, 0]). mj_objectVelocity(flg_local=1) rotates into ximat - the inertial principal-axis frame - not the body frame, and returns center-of-mass point velocity, not body-origin velocity. Consequences measured: the "vy" column was actually vertical velocity vz (walking at cmd 0.25: old reading +0.0093 vs true -0.0424); the angular velocity fed to the policy in sim2sim was [wx, wz, -wy] - a different quantity than Isaac and the real IMU provide. RMS check over 8 s of walking: y/z axes swapped between v6[:3] and the qvel truth.
Change
Fixed sim2sim and both probes to compute ang_b = qvel[3:6] and lin_b = xmat.T @ qvel[0:3] (identical quantity to Isaac's root_ang_vel_b / root_lin_vel_b), with a standalone reproduction script (frame_bug_repro_0809.py).
Outcome
Re-scoring the "failed" C4 lineage under correct coordinates reversed the verdicts: c4r4 checkpoints showed vy 80-126% tracking (old reading: +/-2%) and vx+0.30 at 91-95% where the old metric said 0/5 - the bad frame both mis-measured vy and, via corrupted policy observations, systematically depressed all measured performance. Final product passed 260/260 cells.
Mechanism
A simulator API's frame convention is part of the observation contract; when the model's inertial frame is rotated relative to the body frame, frame-agnostic use of a "local" velocity silently permutes axes. Feeding a policy an axis-permuted angular velocity is an observation corruption that degrades behavior everywhere, not just on the axis being studied.
Applies when
- building or auditing a cross-simulator evaluation harness
- one measured axis reads near-zero under all conditions
- sim2sim scores are inexplicably worse than training-side metrics
- URDF/MJCF inertial frames are rotated relative to body frames
“base_link 的 iquat = [0.7071, 0.7071, 0, 0] … mj_objectVelocity 用的是这个 … 喂给策略的 base_ang_vel 是 [wx, wz, −wy] —— MuJoCo 侧观测与 Isaac / 真机 IMU 不是同一个量；vy_mean 报的是竖直速度 vz —— 前进 cmd 0.25 时旧读数 +0.0093，真值 −0.0424。”
train/C_LADDER_RUN.md § 3l. ⚠️ mj_objectVelocity 读的是惯性主轴系 / 3m. 一 bug 坐实 Model CAN polling skew - joint observations are 6-9 ms stale by read order
can-timing-skew-modelingIf joints are read sequentially over a shared bus, reproduce the per-group observation staleness in sim (or randomize it over the measured range) - synchronous observations are a privileged fiction.
Symptom
Policies trained with synchronous joint observations degrade on hardware where motors are polled sequentially over CAN - hip data is already 6-9 ms old by the time ankle data arrives.
Context
Menlo's core sim2real finding on a leg platform of almost identical mass to Lucen's. CAN is a sequential bus: one poll cycle reads motors in a fixed order, so the observation vector mixes timestamps. Lucen runs 6:6 motors on a dual CAN split, so the problem transfers one-to-one.
Change
Explicitly model CAN timing skew in sim: give the joint groups different observation delays matching physical read order. Menlo went further - running real firmware in the loop with a motor simulator between MuJoCo and firmware injecting 0.4-2 ms uniformly distributed delay.
Outcome
Reported by the reference team as their core sim2real enabler on a same-scale platform; recorded in Lucen's experience log as directly applicable ("这个问题一模一样").
Mechanism
A policy exploits any cross-joint temporal coherence present in training observations; when hardware breaks that coherence per bus position, the learned feedback acts on inconsistent state estimates, injecting phase error exactly at the control bandwidth.
Conflicts
Second-hand episode: outcome numbers are the reference team's report, not a Lucen-run experiment; Lucen adopted the requirement but the corpus has no Lucen A/B of skew-on vs skew-off.
Applies when
- robot polls actuators sequentially over CAN/RS485 or similar shared bus
- sim2real degradation appears as jitter or oscillation not seen in sim
- designing the observation/delay model before a training run
“电机走 CAN 是顺序轮询的，髋部电机的数据到踝部电机上报时已经陈旧了 6-9 ms，他们直接在仿真里显式建模了 CAN 时序偏斜，按读取顺序给三组关节不同的观测延迟。… 注入 0.4–2 ms 的均匀分布延迟。你们是 6:6 双 CAN 分总线，这个问题一模一样”
Experience.md § 执行器 + 时序建模 (line 6) Freeze the deployment contract, stamp every export, and let an automated checker catch wiring bugs
contract-freeze-and-checkerFreeze and fingerprint the policy I/O contract; ship contract changes as new versioned profiles that leave old artifacts bit-identical; and extend the automated contract checker with every pipeline change, forcing the new path to execute in the check.
Symptom
Contract-level changes (observation layout, action pipeline) are where silent sim/real divergence is born; two real wiring bugs appeared the one time the action pipeline was extended.
Context
The 215-dim observation contract was frozen ("纪元 3，三机 digest" - an era number plus a digest agreed across three machines); proposals that would break it (e.g. a GRU memory) were rejected on contract grounds. Every exported ONNX is stamped and verified with a manifest (onnx_manifest --stamp / --verify), and deployment refuses mismatched combinations. When C4 added the lateral feed-forward, it went in as a NEW profile (omni_ff) leaving the existing omni profile's behavior bit-identical; the checker (check_contract) was extended to force the feed-forward path to actually execute (cmd_vy=0.13) and promptly caught two genuine bugs: (1) re-clamping with soft_joint_pos_limits after the feed-forward (0.23 rad deviation) instead of reusing the parent's clip; (2) indexing processed actions by asset.joint_names instead of the action term's own contract-ordered _joint_names, which landed the feed-forward on the wrong joints (l_hip_yaw / r_ankle_pitch).
Change
Contract discipline as implemented: frozen dims + digest; manifest stamping and refusal; contract changes only via new versioned profiles; checker updated in the same commit as any pipeline change, with inputs chosen so new code paths are exercised.
Outcome
Both wiring bugs caught before any training or deployment ("两个都是 check_contract 当场抓出来的 —— 这次它值回票价"); old deployments provably unaffected by the new profile.
Mechanism
The contract is the only interface the policy and robot share; freezing plus fingerprinting makes divergence detectable, and an executable checker turns "the contract holds" from a belief into a test - but only if its inputs actually drive the new code path.
Applies when
- modifying the action or observation pipeline of a deployed policy
- exporting policies for hardware
- proposals that would change observation dims or history structure
“契约校验抓到的两个真错误（记账，别再犯）：1. 前馈后误用 soft_joint_pos_limits（URDF 限位 ×0.9）重钳 → 0.23 rad 偏差 … 2. 用 asset.joint_names 索引 _processed_actions → 前馈落到 l_hip_yaw/r_ankle_pitch 上 … 两个都是 check_contract 当场抓出来的 —— 这次它值回票价。”
train/C_LADDER_RUN.md § 3j. 契约级改动 / 契约校验抓到的两个真错误 Continuing a converged policy on a change that carried no new gradient drifted its transfer from 100/98% to 80/28% over 3,000 iterations while every Isaac gate stayed perfect - scan every checkpoint on the second simulator's friction axis
converged-continuation-is-poisonBefore continuing a converged policy, check that the change creates a live gradient; if it does not, cap the budget at a few hundred iterations, and in every continuation scan each checkpoint on the second simulator's transfer axis (for example low friction) - trainer-side gates can stay perfect while transfer decays.
Symptom
V2.7-A (swap the flat_feet term for a compensated version, continue from v2_6c) finished with the line's best Isaac score (100%) and a MuJoCo transfer collapse: mu 1.0 98 -> 80%, mu 0.4 98 -> 28%; the stance it was meant to widen had not moved.
Context
The new term's calibration run showed a near-zero tax from the start: the policy already satisfied it, so the reward landscape offered nothing new. A checkpoint scan on MuJoCo mu {1.0, 0.4} located the damage: +100 iterations 100/98% (better than the baseline), then 86/54, 60/38, 80/28 - monotonic decay with training length, while entropy and action noise rose (7.77 -> 8.18, 0.588 -> 0.612): drift, not sharpening.
Change
Rule written in: with no new gradient, a continuation budget is short (at most a few hundred iterations) and the MuJoCo transfer axis enters every checkpoint scan. The next rung (V2.7b, a live stance-width gradient) was budgeted at 1,000 iterations with mu {1.0, 0.4} scans every 100 and a stop-on-signal rule.
Outcome
V2.7b kept transfer at the same depth (mu 1.0 98% / mu 0.4 92% at +1,000, where A had already rotted to 86/54) and at +3,000 (100/96%): a live gradient preserved transfer. V2.8 then broke that pattern (mu 0.4 2%): the gradient must also be compatible with the policy's existing form.
Mechanism
On a converged reward landscape PPO keeps updating without a signal to follow, and the random walk is pulled toward whatever the training plant rewards idiosyncratically - invisible in the trainer's own gates.
Conflicts
The drift mechanism is the spec's reading of one decay series plus one contrasting run; V2.8 is recorded as an exception to "live gradient keeps transfer".
Applies when
- fine-tuning a converged policy with a small reward change
- a continuation run's trainer-side metrics improve while real or cross-sim results worsen
- choosing which checkpoint of a continuation to ship
“**checkpoint 扫定死因**(μ1.0/μ0.4):**29500(+100 iter)= 100/98%** (优于基线!)→ 30400 = 86/54 → 31400 = 60/38 → 32398 = 80/28 —— **迁移随续训长度单调衰减**。 … **教训入库:收敛均衡上的长续训是毒药 —— 无新梯度时 续训预算须短(≲数百 iter),且 MuJoCo 迁移轴必须进 checkpoint 扫描。**”
git:Lucen-recovery@origin/recovery:train/RECOVERY_V0_SPEC.md § §43 结果:V2.7-A 判 FAIL —— 换刀本身无罪,毒在续训预算 Tightening the bridge's rate limiter under an unchanged policy cut torque peaks 30-50% and made other things worse - the policy cannot see the limiter, keeps commanding and winds up; a deploy-side limiter is a safety net, not a cure
deploy-rate-limiter-windupA rate or torque limiter added at deployment lowers peaks but the policy still commands as if unconstrained (saturation, windup, new contacts); use it as a safety net mirrored in evaluation, and put the constraint where the policy can learn around it.
Symptom
After the violent first real-robot get-up, the cheapest candidate fix was to tighten the bridge's slew (rate) limit for the recovery policy without retraining.
Context
Probe on R3.1 in MuJoCo (5 categories x 3 seeds, mu 1.0), monkeypatching the limiter with no repository change: TIGHT = RS06 4.0 / RS02 3.0 / RS00 2.0 rad/s (about 0.08/0.06/0.04 rad per policy step) against the current vel_limit setting.
Change
The probe decided the role of the limiter rather than a deployment.
Outcome
Success 14/15 -> 12/15; get-up median 2.35 -> 3.53 s (max 9.30); torque demand peak median hip_pitch 164% -> 111%, knee 166% -> 86%; action saturation still 100%; leg-leg contact 558 -> 860 frames. The limiter was kept only as a real-robot safety net (mirrored into sim2sim evaluation); the cure moved into training - where the next lesson was that a limiter anchored on the last command is itself an integrator (slew-anchor-is-an-integrator).
Mechanism
A policy that never trained with the limiter keeps issuing the targets it learned; the limiter clips them, the target window runs ahead (windup), and the robot follows a trajectory the policy never evaluated.
Applies when
- a trained policy is too violent on hardware and a quick deploy-side fix is tempting
- adding slew, torque or velocity limits in a bridge or firmware
- evaluation and deployment use different limiter settings
“判读:**链路侧收紧立等可取地把 τ 峰值砍 30~50%,但成功率掉、饱和率仍 100%、 腿-腿接触反升** —— 策略感知不到限速器,目标窗口继续狂奔。⇒ 收紧 slew 只配当 **真机侧安全网**(必须同步进 sim2sim 口径,基础设施现成),**不配当治法; 治法必须进训练**。”
git:Lucen-recovery@origin/recovery:train/RECOVERY_V0_SPEC.md § §26 探针:收紧桥层 slew,r3_1 不重训直接测 Training at scale 0.4 is NOT the twin of deploying 0.5 at power 0.8 - the algebra matches, the learned policy does not
deploy-scaling-not-training-equivalentNever assume deploy-side scalings can be folded into training-time constants ("burning the crutch into training"): the learned optimum depends on the training-time authority, so treat such conversions as full experiments with pre-registered expectations and a sim2sim gate before any hardware.
Symptom
s1g (S1.6) trained from zero at action_scale 0.4 - meant as the "training twin" of the hardware-proven s1c-at-power-0.8 (0.8 x 0.5 = 0.4) - was all green in Isaac (zero falls, reward 117) yet scored 0/3 across all eight checkpoints and 0/20 at 20 seeds in the MuJoCo gate, falling forward at median 1.57 s with a 2.9x speed overshoot.
Context
The pre-registered expectation (survival gate should pass, since the conviction matrix showed s1c@0.8+delay2 all-survive) was cleanly falsified, and the harness was acquitted by controls: --delay 0 fell identically (not a delay fragility), check_contract all green, and s1c through the same harness survived 2/3. The verdict: "「s1c@0.8 = 0.4 训练孪生」的代数等价不成立" - a policy deployed with a derated output still LIVES in the 0.5 internal model it trained under (its value function, its expectations of its own authority), while a policy that starts training with reduced authority learns a different, clip-hugging gait with zero margin for plant differences ("部署端打折的策略活在 0.5 的内模里,训练起点收权限学出的是贴 clip 的 另一套步态,对 plant 差异零余量"). Result: the policy was withdrawn before hardware ("撤回——不上真机"), the lineage root moved back to the 0.5-contract s1c-5500, and this became the C ladder's cited fact-check ("s1g 是 0/20 证伪出局的那一代").
Change
The amplitude-surgery route abandoned; contract kept at scale 0.5; the deploy-side 0.8 crutch later retired on its own merits when the delay-complete s1e generation ran at full power.
Outcome
One training run bought a clean falsification of a plausible algebraic identity; no hardware time was spent on it because the sim2sim gate caught it.
Mechanism
Output scaling commutes with the network arithmetic but not with learning: the training-time scale shapes which gait solutions are reachable and how much clip headroom the optimum keeps. A derated mature policy retains the wide-authority solution executed softly; a from-zero narrow-authority policy finds a different optimum that saturates its smaller envelope - the two are not the same controller in different units.
Applies when
- proposing to move a deployment derating into a training constant
- a scaled-down contract policy hugs the action clip
- Isaac-green / cross-sim-zero results on a re-scaled lineage
“预注册 a) 证伪——Isaac 全绿(零摔/reward 117)但 MuJoCo --delay 2 八档 checkpoint 扫描全数 0/3、iter6500 20-seed 0/20 … 「s1c@0.8 = 0.4 训练孪生」的代数等价不成立: 部署端打折的策略活在 0.5 的内模里,训练起点收权限学出的是贴 clip 的另一套步态,对 plant 差异零余量。”
train/OMNI_V0_SPEC.md § 3. S1.6 判决(2026-08-07 验收) 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) Run acceptance under measured contact parameters - honest condim/torsional-friction flipped a false PASS into a real-matching FAIL
eval-plant-honesty-contact-paramsPin the evaluation plant's contact model to measured values (contact dimension, torsional/rolling friction, mu) before trusting any gate that involves slip, impact, or drift - a gate can only fail a policy for physics its simulator contains.
Symptom
walk_v5 passed the old acceptance battery yet failed on the real robot (footfall force, drift, kicking) - the evaluation plant was flattering the policy.
Context
The battery was re-run under "honest contact parameters" - condim 4 (adding torsional contact) with measured torsional friction 0.035 - and v5 then FAILED exactly the rows corresponding to its real problems: heading 185 deg (limit 30), support-foot yaw slip 284 deg (limit 80), landing force 1.72x (limit 1.5x), suspended tilt 45.9 deg (limit 10). The slip physics depends on torsional friction, which the default contact model (condim 3) does not even simulate - a slip problem is invisible to an evaluator that cannot represent yaw friction at the foot. Term-sizing measurements for the new rewards were likewise taken under the same honest parameters (cmd 0.45, skipping the 5 s start transient).
Change
Acceptance harness pinned to condim 4 / torsion 0.035 (measured); verdicts issued under defaults declared non-citable for these rows.
Outcome
Sim acceptance verdicts began agreeing with hardware ("现在失败, 与真机一致"); the v6 fixes could be developed and validated against an evaluator that could actually see the disease.
Mechanism
An evaluator is a plant model too: contact dimensionality and friction values decide which failure modes exist in the simulation at all. Evaluating under default contact parameters tests the policy in a world where its real failure is physically impossible, producing structurally false PASSes.
Applies when
- sim acceptance passes policies that fail on hardware
- slip/drift/impact gates run under default simulator contact settings
- setting up a cross-simulator evaluation harness
“accept_v2.py 已加三条判据, walk_v5 在诚实的接触参数下(--condim 4 --torsion 0.035)现在失败, 与真机一致:直行 15s 航向累计 <30° | 185° ✗ … 落脚力峰值 <1.5× 体重 | 1.72× ✗”
train/WALK_V6_MINIMAL.md § 5. 验收 Friction DR was demoted after a measurement (94% success at mu 0.4 with no friction randomization) and promoted again when the action contract changed and mu 0.4 fell to 76% - DR priorities belong to a plant and contract, not to a task
friction-priority-re-measured-after-plant-changeRe-measure transfer along the friction axis for every new action contract or plant, not once per task; a DR priority settled under one action parameterization does not carry to the next.
Symptom
Getting up is all scraping and pushing against the ground, and training pinned friction at 1.0, so friction looked like the first thing to randomize.
Context
The MuJoCo gate on R0.5 (5 categories x 10 seeds x 4 friction levels) measured 100/100/98/94% at mu 1.0/0.8/0.6/0.4: degradation showed first as time (prone 3.2 -> 5.3 s), not failure, so friction DR was demoted and the DR budget earmarked for mass/COM. After the switch to the beta-anchored action space, V2.2 read 90/94/90/76%: mu 0.4 was now the weak row.
Change
V2.3 (single variable): friction DR static (1.0, 1.0) -> (0.2, 2.0), dynamic (0.15, 1.6), the HiFAR range keeping the base dynamic/static ratio; restitution untouched. Continued from v2_2.
Outcome
Isaac nominal 99.8% (DR did not hurt the nominal plant); MuJoCo 98/98/96/92% - mu 0.4 76 -> 92%, mu 1.0 back to R3.1's 98% with bounded torque.
Mechanism
How much a policy leans on friction depends on how it moves; the spec records that the sensitivity rose after the action contract changed but does not establish why.
Applies when
- changing the action space, gains or authority of an existing skill
- deciding which DR axis to spend the next rung on
- an earlier sweep justified leaving an axis unrandomized
“**μ 砍到 0.4(训练值的 40%)仍有 94%**,退化先体现在**用时**(prone 3.2→5.3 s) 而不是成败。μ≥0.8 完全无损。→ **§17 曾把"摩擦随机化提到 R4 第一项"当作优先 事项,这条实测把它降级了**”
git:Lucen-recovery@origin/recovery:train/RECOVERY_V0_SPEC.md § §21 MuJoCo 复核门 ② 摩擦依赖 Measure yaw rate by integrating heading, not by averaging body-frame angular velocity - the two differed 15x
heading-integral-not-body-rateFor any secular rate (turn gain, drift), integrate the world-frame angle over the window; never average instantaneous body-frame rates during oscillatory motion - and when code comments warn about a measurement, believe them before re-measuring.
Symptom
Two measurements of the same turn gain disagreed by a factor of ~15: time-averaged body-frame omega_z gave -0.05 while the sim2sim harness's heading-angle integration gave +0.473.
Context
The harness code comment had already documented and predicted the failure: during gait the torso oscillates (body-frame omega_z std up to 0.7); projecting world angular velocity onto a swaying body axis and then averaging biases the estimate systematically - "实测体系均值 −0.04 而实际在以 +0.15 转" (measured body-frame mean -0.04 while actually turning at +0.15). The author's own -0.05 measurement was declared void and the training machine's 1.58/2.45 turn gains confirmed valid.
Change
Measurement doctrine fixed: yaw rate for evaluation = net heading change by integration over the window; instantaneous body-frame rates are unusable for averaged directional statistics during legged gait.
Outcome
Subsequent friction sweeps and turn-gain accounting were all conducted in the heading-integral currency, making cross-simulator comparisons (MuJoCo vs Isaac 1.04/1.02) meaningful.
Mechanism
Averaging a vector quantity expressed in an oscillating frame couples the frame's oscillation into the mean (a rectification bias); the heading integral is computed in the world frame where the gait oscillation integrates to ~zero, leaving the secular component.
Applies when
- measuring turn gain, heading drift, or any secular angular rate
- a body-frame-averaged statistic disagrees with trajectory-level truth
- writing evaluation code for oscillating platforms
“我用体坐标系 ωz 的时间均值测,得 −0.05;sim2sim 用航向角积分,得 +0.473。差 15 倍。… 步态中躯干摇晃(体系 ωz std 可达 0.7),把世界角速度投到摇摆的体轴上再取均值会系统性偏掉 … 结论:偏航率必须用航向积分,体系瞬时角速度取均值不可用。”
train/WALK_DIAGNOSIS.md § ③ 转向增益 —— 我的测法是错的,训练机的 1.58/2.45 成立 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 A frame-history observation under zero DR memorizes the trainer's plant fingerprint - the estimator must see variation to learn estimation
history-obs-needs-plant-variationIf the observation carries history (stacked frames, RNN), keep at least minimal plant variation (gain/latency jitter) on from the first iteration - "nominal first, robust later" is structurally invalid for estimator-bearing contracts.
Symptom
omni_s1 (fresh 215-dim contract with a 5-frame history window, trained with DR fully off): training all green, yet the MuJoCo gate scored 0/20 on all eight doors - falls within 2 s, seven checkpoints, not one transferred.
Context
The history window exists precisely to let the actor implicitly estimate line velocity and actuator dynamics (the actor is denied base_lin_vel by observation honesty). Under a constant plant that implicit estimator has nothing to estimate - it learns the trainer's exact response fingerprint instead, and any other simulator's micro-differences are out-of-distribution: "5 帧窗按设计就是隐式估计器, plant 恒定时它学到 Isaac 精确响应的指纹". The planned "nominal-first-robust-later" staging was declared STRUCTURALLY incompatible with history observations: "估计器要见过变化才学估计, 否则学背诵" (an estimator must see variation to learn estimation, otherwise it learns recitation). Honest confound note kept: this is mixed with "zero DR does not transfer, period" - but both attributions prescribe the same fix, so no control was run.
Change
S1.1: minimum actuator jitter turned on from day one - kp/kd +/-10%, latency 0-1 frame (friction/COM/mass still nominal, no push - those stay for the S2 ladder).
Outcome
Transfer restored: survival 0/20 -> 20/20, speed 19/20, foot distance 20/20 (remaining failures moved to gait quality, a different disease); the staging doctrine was amended - history-carrying contracts never train under a frozen plant.
Mechanism
A recurrent/history channel fits whatever temporal structure minimizes loss; with a deterministic plant the cheapest structure is the plant's own impulse-response signature, yielding features that are simulator-specific rather than physics-general. Plant variation forces the channel to carry state-estimation features that transfer.
Conflicts
Attribution is explicitly confounded with the simpler "zero DR never transfers" reading ("与「零 DR 本身就不迁移」混杂 … 两种归因处方相同, 不做对照") - the source chose not to spend a control run separating them.
Applies when
- adding frame stacking or recurrence to an actor observation
- a nominal-plant policy fails a cross-simulator gate within seconds
- planning DR staging for a new contract
“frame_hist × 零 DR = plant 指纹过拟合——5 帧窗按设计就是隐式估计器, plant 恒定时它学到 Isaac 精确响应的指纹, MuJoCo 的微小差异即 OOD, 2 s 内摔, 七个 checkpoint 无一迁移。「先标称后鲁棒」的分段与历史观测结构性冲突:估计器要见过变化才学估计,否则学背诵。”
train/OMNI_V0_SPEC.md § 3. S1.1 修订记录 ① Never referee a suspect metric with another metric from the same code - they can share the disease
independent-referee-for-metric-disputesTo adjudicate a disputed measurement, compute the quantity by an independent method from raw state; never accept a sibling column from the same pipeline as the tiebreaker.
Symptom
A triple reversal on one question: sidewalk sign diagnosis (correct) was retracted using a second metric from the same script, then the retraction itself had to be retracted when that second metric turned out to be the buggy one - two opposite-direction errors on the same problem in one day, both written into the execution sheet.
Context
The probe's net-displacement metric suggested the sidewalk reference sign was inverted. Worried about yaw-drift pollution of net displacement, the author checked the same table's body-frame vy_mean column (~0.003 everywhere, 20-50x smaller) and retracted the sign diagnosis. But vy_mean came from mj_objectVelocity, which was silently reporting vertical velocity due to a frame bug - the "referee" was the diseased measurement. Re-measured with a truly independent computation (xmat.T @ qvel, world trajectory), the original diagnosis was confirmed: saw -0.5 gave vy +0.058/-0.130 (76%/106%), consistent with the net-displacement values all along (yaw pollution was real but only 10-21 deg, nowhere near reversal-sized).
Change
Lesson written twice, verbatim, as a hard rule: when questioning a measurement, the referee must be an independent algorithm (different code path, different physical derivation), e.g. rotate qvel by the body matrix directly, or inspect the raw world trajectory.
Outcome
With the independent referee in place the frame bug was confirmed, fixed, and the whole C4 line re-scored - revealing sidewalk had been working (see body-frame-velocity-api-audit).
Mechanism
Metrics sharing a code path (or an upstream API) share failure modes; agreement between them is evidence about the code, not the world. Only a measurement with an independent derivation can break the tie, because its errors are uncorrelated with the suspect's.
Applies when
- two metrics of the same quantity disagree
- about to retract a conclusion based on a second readout
- auditing evaluation code after a surprising result
“我用一个坏指标去质疑一个好指标，并把撤回写进了执行单。教训（写死）：质疑一个测量时，不能用同一份代码里的另一个测量当裁判 —— 它们可能同源同病。裁判必须是独立算法（这次的裁判应该一开始就是 xmat.T · qvel[:3]，或直接看世界轨迹）。”
train/C_LADDER_RUN.md § 3m. 二 我今天犯了两个方向相反的错 / 3n. 五 元教训 The action-delay was implemented as lerp - beyond one step it extrapolated BACKWARD, so a whole lineage trained on a fictitious actuator
latency-lerp-reverse-extrapolationUnit-test plant-model code (delays, filters, randomizers) against hand-computed truth across its FULL configured range, not just the nominal case - a delay must be a queue, and any interpolation used outside [0,1] is a silent plant corruption that training will faithfully absorb.
Symptom
Every walk/stand model up to v11 had been trained on a silently wrong plant: the action-latency implementation lerp(cur, prev, lag) is only an interpolation for lag <= 1 - at lag 3 it computes 3*prev - 2*cur, a REVERSE extrapolation. With the configured (0, 0.06) s at 50 Hz (lag in [0,3]), about 2/3 of environments were adapting to actuator dynamics that do not exist.
Context
Listed as evidence item #1 for the full restart: "全部旧模型训在错误 plant 上" and the head suspect for the real robot's wild kicking. The fix replaced it with a true FIFO delay line (commit 7f13793) plus its own regression test (tests/test_action_latency.py) - but every exported ONNX predated the fix, which is part of why the lineage was frozen rather than patched.
Change
Delay implementation rewritten as an honest FIFO with unit tests; the restart baseline trained on the corrected plant from day one.
Outcome
A generation-scale training investment was revealed to have a corrupt plant underneath; the class of bug (plausible-looking math that silently changes meaning outside its valid range) got a permanent test.
Mechanism
lerp(a, b, w) leaves the segment for w > 1; used as a delay it fabricates high-gain inverted dynamics precisely in the largest-delay draws, so the policy learns compensation for an actuator that cannot exist - and DR then trains robustness to the artifact rather than to reality. No training metric can catch this: the sim is self-consistent, just wrong.
Applies when
- implementing or auditing action delay / filtering in a trainer
- a lineage behaves as if compensating dynamics nobody modeled
- deciding whether old checkpoints are salvageable after a plant bug
“动作延迟旧实现 lerp(cur, prev, lag) 在 lag>1 时是反向外插(w=3 → 3·prev−2·cur),配置 [0,0.06]s@50Hz 即 lag∈[0,3],约 2/3 env 在适应不存在的执行器动态。7f13793 已换真 FIFO,但所有 ONNX 均训于修复之前 —— 真机"乱踢"的头号嫌疑。”
train/OMNI_V0_SPEC.md § 0. 为什么从零 (1) 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 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) 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 Write each config's expected hardware signature before the session - and if reality disagrees, change the books, not the conclusion
preregistered-real-expectationsBefore hardware runs, write per-config expected signatures and the disagreement rule (hardware outranks sim; discrepancies get recorded, not reconciled); validate the harness by checking it reproduces at least one known real behavior.
Symptom
Hardware impressions are easily narrated after the fact; without written expectations, any real-robot outcome can be made to "match" the sim story.
Context
The S2 acceptance sheet carried a section titled "sim 侧预注册预期 (事后核对, 不许事后改)" - per-configuration behavioral signatures written before the session: s1e@0.8 the disturbance king (push 159/160, zero chirality, all-mu 20/20) at the cost of speed gates 0/20 and zero-command wander ~0.98 m with -29.5 deg/20 s rotation; fric-3000 "walks accurately but is easier to push over"; fric-2400 neither. Credibility check included: the sim harness had reproduced the already-recorded real behavior (pace in place + right drift + net rotation -30 deg/20 s), which "提高本单全部预期的可信度". The anomaly clause fixed the epistemics in advance: if results systematically disagree with sim, "不改结论改账" - don't massage the conclusion, write the discrepancy into the books, and per the earlier zero-warning lesson, hardware wins.
Change
Every hardware session ships with a pre-registered expectation table (signature per config), a baseline-match credibility check, and a written precedence rule for disagreement.
Outcome
The A/B session became falsifiable: agreement confirms the proxy, disagreement is booked as a proxy-bias finding rather than argued away.
Mechanism
Pre-registration converts qualitative hardware sessions into tests of the sim-to-real mapping itself; a reproduced known behavior calibrates trust in the remaining predictions; and fixing "who wins on disagreement" beforehand prevents authority from drifting to whichever source flatters the plan.
Applies when
- planning any hardware acceptance or A/B session
- the sim harness's credibility in this regime is unestablished
- post-session write-ups tempt narrative fitting
“⚠️ sim 复现了真机已记录的「原地踏步 + 右漂 + 净旋 −30°/20s」—— harness 与真机行为对得上, 提高本单全部预期的可信度。… 结果与 sim 系统性不符 → 不改结论改账: 写进 README 该节, 按 「Isaac 指标三次零预警」的教训, 以真机为准。”
train/REAL_RUN_S2.md § 2. sim 侧预注册预期 (事后核对, 不许事后改) / 4. 异常处置 When the eval proxy has a known systematic bias, gate on within-proxy differences, not absolutes
relative-metrics-survive-proxy-biasWhere the evaluator is known-biased, design gates as within-evaluator contrasts (left vs right, A vs B, pre vs post) that cancel common-mode error; reserve absolute thresholds for quantities whose proxy calibration has been checked.
Symptom
MuJoCo systematically overestimated yaw turn gain (1.58/2.45 where Isaac measured 1.04/1.02), and friction alignment recovered only part of the gap (mu 1.0 -> 0.6 pulled it to 1.43/2.27) - so any absolute turn-gain acceptance threshold in MuJoCo would be judging the proxy's bias, not the policy.
Context
The acceptance criterion was rewritten to use only the left/right difference of the turn gain (standard: <20%): both directions pass through the same biased proxy, so the bias largely cancels in the difference while the policy's chirality - the thing being gated - survives. The absolute-gain row was dropped: "转向只判左右差,不判绝对值". A parallel task was still opened to align MuJoCo contact parameters to Isaac's (mu, restitution), since drift proved highly friction-sensitive (straight-line drift -65 deg at mu 1.0 vs +1 deg at mu 0.4) - bias reduction and bias-robust metrics proceeded together.
Change
Gate metric changed from absolute turn gain to left-right gain difference; proxy-alignment work scheduled separately rather than blocking acceptance.
Outcome
Turn acceptance became meaningful across proxy versions (v5 70% -> v6 19% difference measured the real improvement) while the absolute bias question was pursued without holding the ladder hostage.
Mechanism
A systematic multiplicative or additive proxy bias applies to both arms of a mirrored measurement; differencing (or ratioing) mirrored conditions cancels the common-mode bias to first order, leaving the asymmetry signal. Metrics built this way remain valid while the proxy is imperfect - which it always is somewhere.
Applies when
- a sim proxy disagrees with the trainer or hardware in absolute terms
- writing acceptance thresholds for direction-paired skills
- proxy calibration work would otherwise block a ladder
“转向只判左右差,不判绝对值 —— MuJoCo 的偏航增益系统性高估(实测 1.58/2.45 vs Isaac 1.04/1.02),摩擦只能解释一部分(μ 1.0→0.6 仅拉到 1.43/2.27)。差值是相对量。”
train/WALK_V5_SPEC.md § 6. 验收 Train with self-collisions ON (filtering nested-link ghost pairs) - the reward wall prevents, the physics makes cheating impossible
self-collision-physics-plus-reward-wallNever train a contact-risk behavior with self-collisions disabled; enable them with an audited filter list for nested/overlapping pairs (zero contacts across a pose sweep), record the fps cost, and keep a calibrated distance penalty as the preventive layer on top.
Symptom
walk_v8 logged 107 frames of leg-on-leg contact while still earning 0.751 tracking score - because training-side self-collisions were OFF, leg clipping was literally imperceptible to the policy ("碰腿在训练里 根本感知不到").
Context
Enabling self-collisions naively is its own trap: an Isaac audit had shown PhysX auto-filters adjacent bodies (base-hip clean for free) but nested links generate ghost forces - calf and ankle_roll overlap 65 mm at the zero pose, producing 12x body-weight phantom forces. The v10 recipe: enable self-collisions, explicitly filter only the two nested pairs (l/r calf-ankle_roll), then run a zero-contact audit at three poses (nominal stand, walk crouch, swing-extreme) requiring contact count = 0, adding any residual pair to the filter and re-auditing; a 500-iter sanity run for NaN and an fps-cost record (measured -8.8%). Redundancy with the reward-side foot-distance wall was argued, not assumed: "N2 离得远(奖励侧预防),SC 碰了疼(物理侧兜底)" - the reward keeps distance at range, the physics makes contact hurt - so the v8-style "clip legs and still score" outcome becomes physically impossible.
Change
enabled_self_collisions=True + 2-pair filter + three-pose zero-contact audit (re-verified at 0.00 N after the later mass update) + fps budget recorded.
Outcome
Leg contact entered the training signal; the audit protocol caught the nested-pair ghost-force hazard before it corrupted training; combined with the calibrated distance wall, later versions held contact = 0 on hardware and in sim.
Mechanism
A hazard absent from the training physics cannot be learned about, no matter the reward; but collision meshes that interpenetrate at rest inject large fictitious forces if enabled blindly. Filtered enabling plus a pose-swept zero-contact audit gives true contact physics with no phantom energy - and layering prevention (reward) with consequence (physics) covers both learning and enforcement.
Applies when
- real robot self-contacts while training scored it healthy
- enabling self-collisions on a model with nested collision meshes
- deciding between reward-side and physics-side fixes for clipping
“PhysX 自动过滤相邻体(base↔hip_pitch 免费干净),幽灵力只在 calf↔ankle_roll(零位嵌套 65mm,12 倍体重)。… 与 N2 互补不冗余:N2 离得远(奖励侧预防),SC 碰了疼(物理侧兜底)—— v8 那种 107 帧互碰拿 0.751 跟踪分的事从此物理上不可能。”
train/WALK_V10_SPEC.md § 4. SC —— 训练侧自碰撞(范围已探明,比想象便宜) A sim veto needs real confirmation too - the worst sim cell was scheduled as the most informative hardware run
sim-veto-needs-real-confirmationNever let sim alone both condemn a purpose-built configuration and escape audit: spend one cheap, safeguarded hardware run on the condemned cell, pre-registering what agreement and disagreement would each imply about the proxy.
Symptom
The fric-2400@kd1.0 combination was sim's worst cell across the board (survival 17/20 - the only miss, mu0.4 1/20, push 103/160, zero-cmd 2/20), yet it was the only product specifically trained for the kd1.0 deployment gain - discarding it on sim evidence alone would leave the sim's own validity untested exactly where it mattered.
Context
The team had been burned in the other direction before ("Isaac 指标三次 零预警" - training-side metrics gave zero warning three times), so the symmetric rule was written: sim's rejection also needs hardware confirmation ("sim 判被支配 ≠ 真机被支配 … sim 的否决也要真机确认"). The run was pre-registered with a dual reading: real matches sim -> the S2f ladder closes and the fork root is settled; real clearly better than sim -> the MuJoCo proxy has a systematic bias in the kd1.0/low-margin region, "那比选型本身重要得多" - and every S2f sim acceptance would need re-scoring.
Change
The condemned configuration was kept on the hardware roster (last, spotted, minimal exposure) explicitly as a proxy-validation probe, not as a deployment candidate.
Outcome
Session design captured either result as progress: selection confirmed, or a proxy bias discovered that would re-price the whole ladder's verdicts.
Mechanism
Every sim verdict is a joint statement about the policy AND the proxy; cells where a policy was purpose-trained for the exact condition sim condemns are where proxy error is most likely and most costly. Testing the veto converts a selection decision into a calibration measurement of the evaluator itself.
Applies when
- sim rejects the configuration that targets the actual deployment condition
- the eval proxy's calibration has never been checked in that regime
- deciding which hardware runs are worth their risk
“但它也是唯一为 kd1.0 部署档专门训的产物 —— sim 判被支配 ≠ 真机被支配, 「Isaac 指标三次零预警」的教训反过来同样成立: sim 的否决也要真机确认。… 若真机明显好于 sim → MuJoCo 代理在 kd1.0/低裕度区有系统性偏差, 那比选型本身重要得多。”
train/REAL_RUN_S2.md § 上机名单 note / 2. sim 侧预注册预期 ⑤ Cross-simulator gate (Isaac Lab to MuJoCo) comes before any hardware attempt
sim2sim-gate-before-sim2realGate every policy through a second simulator with an independently built plant before hardware; treat sim2sim failure as a contract or overfitting bug, and sim2real failure after a sim2sim pass as a plant/actuator gap.
Symptom
A policy that only ever ran in its training simulator carries untested dependencies on that simulator's solver, contact model, and defaults; the first place those dependencies surface should not be the real robot.
Context
Standing order of operations for the whole Lucen program, recorded as the opening line of the experience log: train in Isaac Lab, gate in MuJoCo, only then go to hardware. The MuJoCo side is the same plant used for evaluation batteries, so a sim2sim pass also validates the exported policy + contract (obs ordering, scales, defaults) outside the training stack.
Change
Pipeline rule adopted - every checkpoint must pass the MuJoCo evaluation battery (sim2sim) before it is considered for real deployment (sim2real).
Outcome
Used generation after generation as the cheap filter; hardware sessions only ever received policies that had already survived a second simulator.
Mechanism
Two simulators disagree exactly where a policy is overfit to simulator-specific artifacts (contact softness, integrator, default parameters, obs conventions); a cross-sim transfer catches contract bugs and solver overfitting at zero hardware risk, so real-robot failures that remain are attributable to genuine plant/actuator gaps.
Applies when
- planning the path from training to first hardware trial
- exported policy behaves differently outside the training framework
- triaging whether a real-robot failure is contract vs plant
“先sim2sim - 从isaaclab 到mujoco / 再sim2real”
Experience.md § opening lines (1-2) Add a single-point-suspension test to acceptance - the ground is a free stabilizer that hides divergence
suspension-probe-removes-free-stabilizerInclude at least one acceptance condition that strips the environment's free stabilization (suspension, or equivalent) - the sim-passing policy that fails on hardware is often failing a condition the battery never posed.
Symptom
walk_v5 looked healthy in every on-ground sim test yet diverged on the real robot - the acceptance battery had never measured a condition that would have revealed it.
Context
The battery gained a single-point-suspension probe (robot hung, feet free): measure torso tilt while the policy runs without ground contact. v5 scored 45.9 deg mean tilt suspended - wildly unstable - which the file calls "最灵敏的失稳探针(拿掉地面这个免费稳定器)": ground reaction forces passively stabilize a marginal policy, so on-ground metrics saturate long before the policy's internal balance is actually sound. v6 halved it (23.0 deg, target <10 deg) - progress visible on a scale where on-ground numbers showed nothing.
Change
Suspended-tilt added as a standing acceptance row; run under the honest contact parameters battery (accept_v2 with measured condim 4 / torsional friction 0.035), under which v5 correctly FAILS in agreement with the real robot.
Outcome
The sim battery's verdict on v5 flipped from pass to fail, matching hardware; suspended tilt became the discriminating metric between v5 and v6 (45.9 vs 23.0 deg) when ground metrics differed little.
Mechanism
Contact with the ground closes a stabilizing feedback loop the policy gets for free; removing it exposes the policy's own attitude control authority. A metric measured only in the assisted condition cannot rank policies by the unassisted quantity that hardware will actually demand during perturbations and flight phases.
Applies when
- sim acceptance passes but hardware diverges
- designing an acceptance battery for a legged robot
- two candidates tie on ground metrics
“单点吊那条是最灵敏的失稳探针(拿掉地面这个"免费稳定器"), v5 在地上一切正常却在真机发散, 就是因为验收从没测过这个工况。”
train/WALK_V6_MINIMAL.md § 5. 验收 The run policy never left the ground and fell in the second simulator from the frontal plane - its DR (gains and latency only) covered the actuator axis, not the frontal-plane contact and inertia disturbances the doubled stride amplified; "is DR on" is the wrong question
thin-dr-judged-by-channel-coverageJudge a DR recipe by whether its randomized terms cover the channel where the skill can lose stability, not by whether DR is enabled; when a new skill lengthens single support or enlarges motion in one plane, add disturbances in the plane it destabilizes before training.
Symptom
run R1 (6,000 iterations, 78 min): no flight phase ever appeared, and every one of 13 checkpoints failed the eight-gate MuJoCo smoke. In Isaac: zero terminations in 6,000 iterations, 4.2 deg tilt. In MuJoCo at delay 2: 1/6 survived, falls within 1.9-6.2 s at 50.8-58.7 deg, the most saturated joints all roll joints.
Context
The run contract doubled sagittal travel (knee action scale 0.9, knee swing peak 1.14 rad) with a 0.60 s period and 0.40 duty - long single support - while roll/yaw scales were deliberately left at 0.5. DR copied the s1e recipe: kp/kd (0.9, 1.1) and latency on; mass, COM, joint friction and push all off; ground friction pinned at (1.0, 1.0). Flight was read two independent ways: Isaac's per-foot contact reward stayed 0.845-0.857, never above 0.87 - the arithmetic ceiling of a gait with zero flight - and 30 of 36 MuJoCo seeds had flight fraction exactly 0 (the nonzero six were all tumbling falls). Foot lift itself worked (46-59 mm against a 50 mm design point): the walk-era "not enough travel" failure did not recur.
Change
Verdict FAIL, with the pre-registered first knob (exploration noise 1.0 -> 1.2) explicitly rejected as aimed at a different axis. The lesson was generalized and applied at the next line's design review: the one-leg spec made push, body mass, base COM and friction DR mandatory for its permanent single support and banned the thin recipe.
Outcome
The run line did not continue past R1 in the sources. The one-leg V0 with the wider DR passed its friction-variant gate (mu 0.4 and 1.2) inside a 40/40 acceptance.
Mechanism
Randomizing gains and latency covers the actuator's axis; a skill whose failure lives in frontal-plane contact and inertia needs randomization on that channel (push, mass, COM, friction), or the trainer's exact plant becomes the only one the policy can stand on - the omni_s1 transfer trap a second time, this time with DR switched on.
Applies when
- a policy is flawless in the trainer and falls immediately in a second simulator
- reusing a DR recipe from a skill with a different support pattern
- failures concentrate on one axis (roll, yaw) the DR does not touch
“**机理**: 矢状面行程翻倍 (膝摆动峰 1.14 rad) + T 0.60 + duty 0.40 的长单支撑, 把额状面扰动放大了一个量级; 而 roll/yaw 通道按 §3 **刻意没有放大** (仍 0.5), DR 又是 s1e 复刻的薄配方 (mass/COM/关节摩擦/push **四关全关**, 地面摩擦钉死 (1.0, 1.0))。 … 说明**薄 DR 的判据不能只看"有没有开 DR"**, 要看**开的那几项 是否覆盖失稳所在的通道** —— kp/kd 与延迟是执行器轴向的, 对额状面接触/惯性 扰动零覆盖。 … 0.87 正是「零腾空的走路步态」的天花板算术”
git:Lucen V2@origin/run-line:train/README.md § run R1 FAIL (2026-08-09, run 21-30-30_run_r1): 腾空零, 但病根在额状面不在探索 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 A torque-tail penalty was paid for by bracing the legs against each other - the second simulator's leg-contact count caught it, and the first explanation ("the trainer can't see self-collision") was retracted from the run's own config
torque-penalty-bought-by-leg-bracingWhen a penalty lowers a demand metric, look for what the policy traded to get there - keep self-contact frames and foot spacing as standing sim2sim readouts - and check any "the trainer cannot see X" explanation against the run's resolved config before it enters the record.
Symptom
After R3.1's torque_headroom term collapsed the demand tail, MuJoCo success fell 100 -> 98% and leg-leg contact frames at mu 1.0 rose 750 -> 2,190 (worst rollout 177 -> 450). The one failure (prone seed 2) had the legs crossed, one foot on the other leg, trapped at 0.067 m - visible on video.
Context
Across the ten prone seeds, foot spacing and leg-leg contact frames were monotonically anti-correlated, and the failure was the extreme of the series. Pulling the legs toward the midline shortens the hip_roll lever arm and lowers torque demand. At the time the spec explained it as Isaac training without self-collisions ("a free lunch in a simulator without self-collision").
Change
Leg-leg contact frames and foot spacing were tracked in every MuJoCo gate; R3.2's candidates were "train with self-collision on" or "a minimum leg spacing term" - not stacked.
Outcome
The next rung's joint-velocity penalty incidentally erased the dependency (2,190 -> 86 frames). On 08-10 the runs' logged env.yaml showed enabled_self_collisions true in both r3_1 and v2_2 (inherited from walk v10): the tangle was physically learned bracing, visible to both simulators, and the Isaac/MuJoCo contact-count gap was mesh and contact fidelity. The "self-collision debt" narrative was withdrawn for the whole line.
Mechanism
A penalty on demand rewards any configuration that lowers demand; legs pressed together act as a mutual support that fails when contact geometry shifts slightly.
Conflicts
§24 attributes the dependency to self-collisions being disabled in training; §36 retracts that from the runs' env.yaml ("§24's mechanism explanation was wrong") and keeps the older sections unedited as history.
Applies when
- a torque, impact or energy penalty improves its metric and cross-sim success drops
- legs or links approach each other after a regularization change
- an explanation relies on a simulator setting nobody checked in the run config
“prone 十个 seed 逐条看,脚距与腿-腿接触帧数单调反相关, 而唯一失败的那条正是最极端的一条 … 机制上说得通:把腿收到身体中线附近能缩短 `hip_roll` 力臂、降低力矩需求”
git:Lucen-recovery@origin/recovery:train/RECOVERY_V0_SPEC.md § §24 代价:MuJoCo 成功率 100% → 98%,病因是两腿卡住