mcMotionChallengeGitHub ↗

MODEL / CHECKPOINT

Westlake / SII

Model

Checkpoint artifacts and an independent MotionChallenge inference adapter.

Organization
Westlake / SII
Type
Dense
Temporal convolutional encoder and dense policy head.
Version
track_g1 · 1ebd341d
author checkpoint or pinned source revision
Robot support
Unitree G1 · 29 DoF
Author source revision
f9263865c581802ad531854b8e547e2403a945f3

Checkpoint artifacts

Author release
BotRunner64/Teleopit ↗
Artifact type
Author-released checkpoint
Original file
track_g1.onnx ↗

The adapter pins this published artifact by SHA256. Artifact identity is separate from inference parity.

Download links deliver the adapter files. The release links and verification notes distinguish author files from conversions; SHA256 identifies the bytes used.

Policy checkpoint
Download ↗
1ebd341d9193e1c49a986450f6043ba1a9473ad46636ce0bcb1c7755c856e0de

Inference interface

The adapter declares its control clock and reference window below. Source and configuration describe the observation and action transforms.

Observation requirements

Observation rules ↗

Model-specific handling

Actual torso orientation is reconstructed by FK. Target velocities are finite differences of REFERENCE positions/orientations, not integration of actual measurements.

Exact checkpoint & evidence ↗
Control interval
20 ms
Reference history
0 ms
Reference preview
0 ms
Output interface
joint-pd-feedforward
Configuration
policy.json

Evaluations