ToddlerBot · published recipe
walk (omnidirectional, 2XC)
Record
- Robot
- ToddlerBot stanford/toddlerbot
- Framework
- MuJoCo MJX + RSL-RL PPO, ONNX Runtime on Jetson · MuJoCo Playground / MJX
- Policy license
- not stated by the package · robot software repos: MIT
- File
- toddlerbot/policies/walk.py
- Note
- Checkpoint toddlerbot_2xc_walk_rsl_20251226_114612.zip in the Google Drive folder linked from the README. It is also the WALK_POLICY used by run_multiple_policy.py.
Not in the tab yet
To replay this policy in the browser we need an ONNX export of a trained checkpoint, the observation and action contract copied from the authors' own sim2sim or deploy script, and a MuJoCo model of the robot that the contract matches. Then it must walk 60 s in our runtime without a fall. Until then the source link is the way in.
Other ToddlerBot policies · 8
- get up from the groundMuJoCo MJX + RSL-RL PPO
- crawl (full / only / rotate)MuJoCo MJX + RSL-RL PPO
- crawl up / down stairsMuJoCo MJX + RSL-RL PPO
- climb up / down box, climb wallMuJoCo MJX + RSL-RL PPO
- multi-skill whole-body locomotion (Locomotion Beyond Feet)RL skill policies + depth skill classifier (FoundationStereo)
- push-up (keyframe replay)open-loop keyframe replay (run_policy.py --policy replay)
- cartwheel (DeepMimic-style)MuJoCo MJX + RSL-RL PPO
- bimanual manipulation (diffusion policy)Diffusion Policy (PyTorch)