Pith Number

pith:53FO7AAM

pith:2026:53FO7AAM5J25EWKGM64HG34GY6

not attested not anchored not stored refs resolved

Terrain Consistent Reference-Guided RL for Humanoid Navigation Autonomy

Aaron D. Ames, William D. Compton, Zachary Olkin

Modulating reference trajectories to fit terrain geometry inside RL training produces humanoid policies that track SE(2) velocity commands reliably on rough outdoor ground and stairs.

arxiv:2605.15517 v1 · 2026-05-15 · cs.RO · cs.SY · eess.SY

Open paper page JSON Open Graph Bundle Merged state Verified badge What is a Pith Number?

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{53FO7AAM5J25EWKGM64HG34GY6}

Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge

Record completeness

1 Bitcoin timestamp

2 Internet Archive

3 Author claim open · sign in to claim

4 Citations open

5 Replications open

✓ Portable graph bundle live · download bundle · merged state

The bundle contains the canonical record plus signed events. A mirror can host it anywhere and recompute the same current state with the deterministic merge algorithm.

Claims

C1strongest claim

Environmentally-conditioned references significantly improve reference tracking performance compared to environment agnostic references. On hardware, we integrate the policy with an MPC + control barrier function planner and demonstrate long-horizon (>70m) closed-loop autonomous navigation on the Unitree G1 through outdoor environments containing rough terrain and consecutive flights of stairs, with all sensing and computation onboard.

C2weakest assumption

That projecting desired footsteps onto valid footholds and adjusting swing-foot and center-of-mass trajectories to match the terrain inside the RL training loop produces stable, generalizable training signals that transfer to real hardware without introducing artifacts, instability, or poor reference tracking on unseen terrain geometries.

C3one line summary

Terrain-consistent reference modulation during RL training yields SE(2)-controllable humanoid locomotion policies that improve tracking in simulation and enable over 70 m closed-loop autonomous navigation on rough terrain and stairs on the Unitree G1 with onboard computation.

References

37 extracted · 37 resolved · 3 Pith anchors

[1] Dtc: Deep tracking control, 2024

[2] Beamdojo: Learning agile humanoid locomotion on sparse footholds

[3] Benet al.Gallant: V oxel Grid-based Humanoid Locomotion and Local-navigation across 3D Constrained Terrains

[4] Learning humanoid locomotion with perceptive internal model, 2025

[5] Attention-based map encoding for learning generalized legged loco- motion 2025

Receipt and verification

First computed	2026-05-20T00:01:02.743486Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

eecaef800cea75d2594667b8736f86c79ac8c73644b3f8f191df60b64f3de227

Aliases

arxiv: 2605.15517 · arxiv_version: 2605.15517v1 · doi: 10.48550/arxiv.2605.15517 · pith_short_12: 53FO7AAM5J25 · pith_short_16: 53FO7AAM5J25EWKG · pith_short_8: 53FO7AAM

Agent API

Resolver JSON Graph JSON Events JSON Schema Signing key

Verify this Pith Number yourself

curl -sH 'Accept: application/ld+json' https://pith.science/pith/53FO7AAM5J25EWKGM64HG34GY6 \
  | jq -c '.canonical_record' \
  | python3 -c "import sys,json,hashlib; b=json.dumps(json.loads(sys.stdin.read()), sort_keys=True, separators=(',',':'), ensure_ascii=False).encode(); print(hashlib.sha256(b).hexdigest())"
# expect: eecaef800cea75d2594667b8736f86c79ac8c73644b3f8f191df60b64f3de227

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "468d0d55f9777f70fb99bb5f892b82dbacd714ee7c0dd0ed642f35c556ff013c",
    "cross_cats_sorted": [
      "cs.SY",
      "eess.SY"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.RO",
    "submitted_at": "2026-05-15T01:27:50Z",
    "title_canon_sha256": "243d8faf633db281828af59cf50f17c0ee5f6db5101d76db652382d4caeab95b"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.15517",
    "kind": "arxiv",
    "version": 1
  }
}