ACT – RoboSynChallenge handle_basket (fine-tuned)

LeRobot 0.3.3 ACT policy for the RoboSynChallenge simulation task handle_basket (CobotMagic dual-arm robot, three 480x640 RGB cameras, 14-D joint state/action).

  • Initialised from RoboSynChallenge/ACT_sim_mixer_operating weights (normalisation statistics recomputed from the task dataset).
  • Fine-tuned for 14,000 steps on RoboSynChallenge/cobotmagic_Sim_handle_basket (1000 episodes), batch size 16, lr 1e-5, AMP; default ACT architecture (ResNet-18, dim 512, 4 enc / 1 dec layers, VAE, chunk size 50).
  • Local check (random setting, seeds 0 and 1, 10 episodes each): 11/20 successes, vs. 5/20 for the released RoboSynChallenge/ACT_sim_handle_basket under the same protocol.

Evaluate with the RoboSynChallenge ACT adapter (LeRobot 0.3.3 required):

bash policy/act/eval.sh handle_basket random <path-to-this-checkpoint> 0 --headless True
Downloads last month
31
Safetensors
Model size
51.6M params
Tensor type
F32
Β·
Video Preview
loading