T-Rex Model
A dinosaur-inspired bipedal simulation model with an oversized head and vestigial forelimbs. Its stage 3 task uses contact from a fixed head geom as a bite proxy; the model has no articulated jaw.
Generated specifications
These values are generated from the environment and compiled MJCF model. “Dynamic mass” is a simulator property, not an anatomical weight estimate.
| Gait | Bipedal |
|---|---|
| Observation dimension | 61 |
| Action dimension / actuators | 15 |
| Generalized coordinates (nq) | 28 |
| Generalized velocities (nv) | 27 |
| Compiled dynamic model mass | 85.7 kg |
| Plant contract revisions | Policy r12 · physics r7 · visual r4 |
| Model | environments/trex/assets/trex.xml |
Current curriculum configuration
The Stable-Baselines3 budgets and early-advancement gates below come from the current TOML files and can differ from older published runs. A stage also ends when its configured budget is exhausted. JAX/MJX uses the same task sequence, but its CLI and notebook gate behavior is not yet equivalent.
| Stage | Objective | SB3 configured budget | SB3 early-advancement gate |
|---|---|---|---|
| 1 — Balance | Learn to stand and balance without falling | 11M | reward ≥ 2,100; ≥ 40 episodes/evaluation; 3 consecutive passes |
| recovery — Recovery | Hold the stance against scheduled external pushes and recover from each | 3M | ≥ 40 episodes/evaluation; 3 consecutive passes |
| 2 — Locomotion | Learn forward walking/running | 8M | reward ≥ 100; episode length ≥ 750; avg. velocity ≥ 1 m/s; ≥ 10 episodes/evaluation; 3 consecutive passes |
| 3 — Bite | Sprint to prey and make contact with the head bite proxy | 8M | reward ≥ 100; avg. velocity ≥ 2 m/s; task success ≥ 50.0%; ≥ 10 episodes/evaluation; 3 consecutive passes |
Backend-specific task-success semantics
- Stable-Baselines3 — Head-contact bite proxy: The head-bite geom contacts the prey geom while the bite reward is enabled; the model has no articulated jaw.
- JAX/MJX — Head-tip proximity bite proxy: The head-tip site comes within 0.35 m of the prey target position while the bite bonus is enabled; physical geom contact is not required and the model has no articulated jaw.
Published run summaries
These are historical experiment records. A result is not evidence for the current model revision unless its provenance is explicitly marked current and verified.
PPO · Stable-Baselines3 · 2026-03-18
Provenance: Historical model · unverified · evaluation episode count not recorded · Stable-Baselines3 (version not recorded)
| Stage | Trained steps | Best eval reward | Avg. forward velocity | Task success (Stable-Baselines3) | Passed |
|---|---|---|---|---|---|
| 1 — balance | 6M | 3008.66 | 0.02 m/s | — | Yes |
| 2 — locomotion | 8M | 1936.01 | 3.47 m/s | — | Yes |
| 3 — bite | 8M | 1294.28 | 1.68 m/s | 96.7% | Yes |
Current Stable-Baselines3 definition for this task label: The head-bite geom contacts the prey geom while the bite reward is enabled; the model has no articulated jaw.
Curriculum
The T-Rex is the first species with a four-stage curriculum, identified by a per-species stage manifest rather than bare numbers:
- Stance (historical stage 1) — settle from a randomized spawn and hold
a stable upright pose; gated on physical stance quality
(
stance_quality/v1), not reward, because a zero-action "statue" is the reward optimum of the undisturbed task. - Recovery (stage 1b, new) — the same task plus scheduled external
pushes (165.5 N for 0.20 s on the current plant, derived from the plant's
own mass and stance geometry), warm-started from the certified stance
checkpoint. Only a disturbance can certify active balance control; the
recovery_quality/v1gate judges per-shove recovery against frozen null baselines and is enabled once its thresholds are calibrated. - Locomotion (historical stage 2) — forward walking/running.
- Behavior (historical stage 3) — the head-contact "bite" proxy task.
Integer stage references keep their historical meaning everywhere (stage 2 remains locomotion); the recovery stage is addressed by its semantic id. The catalog tables and stage videos below cover the numbered curriculum; recovery results join them once the stage's gate is calibrated.
Anatomy
The T-Rex model includes:
- Torso - Forward-leaning body (~30 deg from horizontal) with ribcage and belly
- Neck + Skull - Short neck assembly with an elongated skull shape and brow-ridge geoms
- Head - Fixed contact geom used as a bite proxy; there is no articulated jaw
- Legs - Digitigrade-style hind limbs (hip pitch/roll, knee, ankle; 3 passive toe digits per leg since physics r7)
- Arms - Tiny vestigial forelimbs with 2-fingered hands (passive, not actuated)
- Tail - 5-segment heavy counterbalance to skull (4 actuated segments)
Diagnostic Metrics (Stage 3)
After each training stage the eval command reports:
| Metric | Description |
|---|---|
mean_forward_velocity | Average forward speed (m/s) |
gait_symmetry | Left/right stride symmetry ∈ [0, 1] |
stride_frequency | Step frequency (Hz) |
cost_of_transport | Energy efficiency (lower is better) |
mean_pelvis_height | Upright stability (m) |
mean_heading_alignment | cos θ toward prey ∈ [-1, 1] |
success_rate | Fraction of episodes where the fixed head-contact proxy touches the prey target |
min_prey_distance | Closest approach to prey (m) |
Training Videos
Published videos for T-Rex, where available. Artifact provenance is shown on each stage.
Balance
Learn to stand and balance without falling
PPO · Stable-Baselines3 artifact · historical model · unverified · backend version not recorded. It is not evidence for the current model or stage config.
Recovery
Hold the stance against scheduled external pushes and recover from each
No published video is available for this stage.
Locomotion
Learn forward walking/running
PPO · Stable-Baselines3 artifact · historical model · unverified · backend version not recorded. It is not evidence for the current model or stage config.
Bite
Sprint to prey and make contact with the head bite proxy
PPO · Stable-Baselines3 artifact · historical model · unverified · backend version not recorded. It is not evidence for the current model or stage config.
Usage
cd environments/trex
# View the model
python scripts/view_model.py
# Train stage 1 using its current TOML-configured budget
python scripts/train_sb3.py train --stage 1