v3.0.0b13

[StoneT2000]

v3.0.0b13 Release

Notable changes/additions:

• A new PPO based on LeanRL has been provided under examples/baselines/ppo_fast. PPO training is much much faster now (solve PickCube in a minute! Solve PegInsertionSide in about 1 hour!)
• Scripts for RL based data generation are provided in scripts/data_generation/rl.sh now
• Multiple new tasks have been added including a few bimanual humanoid tasks
• Some minor bugs have been fixed and simulation speed has been improved in a number of ways.
• Much more documentation on domain randomization tools and how to use them for your own environments: https://maniskill.readthedocs.io/en/latest/user_guide/tutorials/domain_randomization.html
• RGB(D) based baseline for SAC added.
• Fix bug with the pd_ee_delta_pose controller not working when using PPO code due to in place modification in that controller.
• Upgraded documentation showing more information now about all tasks available by default in ManiSkill. Task documentation now automatically details whether the task has demos, dense/sparse rewards, as well as its episode length
• Many more demonstrations have been added with more coming, demonstrations are now all placed in the same hugging face datasets repo here: https://huggingface.co/datasets/haosulab/ManiSkill_Demonstrations.

UnitreeG1TransportBox-v1_rt.1.mp4

Full Changelog:

What's Changed

• [Fix] Use same indices for root vel setter as base uses for articulations by @arth-shukla in [Fix] Use same indices for root vel setter as base uses for articulations #665

• [BugFix] Fix bug with digital twin rgb overlay due to new to tensor system and display availabe obs modes if chosen one is not correct by @StoneT2000 in [BugFix] Fix bug with digital twin rgb overlay due to new to tensor system and display availabe obs modes if chosen one is not correct #668

• [BugFix] Fix record episode wrapper storing all parallel env episode seeds instead of just one. by @StoneT2000 in [BugFix] Fix record episode wrapper storing all parallel env episode seeds instead of just one. #671

• [Bug] Fix: (unideal) only do .step() if not doing partial reset by @arth-shukla in [Bug] Fix: (unideal) only do .step() if not doing partial reset #672

• [BugFix] Fix demo_vis_segmentation script by @StoneT2000 in [BugFix] Fix demo_vis_segmentation script #675

• [Feature] Unified states for heterogenous simulated envs by @StoneT2000 in [Feature] Unified states for heterogenous simulated envs #676

• [Docs] Add docs on simulation properties like object textures and controller properties that can be dynamically updated during runtime. by @StoneT2000 in [Docs] Add docs on simulation properties like object textures and controller properties that can be dynamically updated during runtime. #681

• [Feature] Support normals and albedo in observation data by @StoneT2000 in [Feature] Support normals and albedo in observation data #682

• [Feature] Added batched input support for look_at matrix construction by @Xander-Hinrichsen in [Feature] Added batched input support for look_at matrix construction #678

• [Feature[ Update interactive_panda.py to allow using different viewer shaders by @StoneT2000 in https://github.com/[Feature[ Update interactive_panda.py to allow using different viewer shaders #687

• [Feature] Extremely fast PPO by @StoneT2000 in [Feature] Extremely fast PPO #669

• [BugFix] Pose creation always creates a new object now and does not duplicate tensors. by @StoneT2000 in [BugFix] Pose creation always creates a new object now and does not duplicate tensors. #688

• [BugFix] Fix PD EE Delta Pose controller that would sometimes do in place modification of user given action by @StoneT2000 in [BugFix] Fix PD EE Delta Pose controller that would sometimes do in place modification of user given action #689

• [BugFix] Update ycb actor builder to include every object in YCB dataset, not just the ones that can be picked up. by @StoneT2000 in [BugFix] Update ycb actor builder to include every object in YCB dataset, not just the ones that can be picked up. #690

• [BugFix] Fix bug where consistency check for state in env state recording assumes actors/articulations exist by @StoneT2000 in [BugFix] Fix bug where consistency check for state in env state recording assumes actors/articulations exist #691

• [Feature] Organized data generation scripts by @StoneT2000 in [Feature] Organized data generation scripts #692

• [Feature] SAC RGB(D) Baseline and improved memory efficiency by @StoneT2000 in [Feature] SAC RGB(D) Baseline and improved memory efficiency #698

• Fix flatten_state_dict with empty dictionary by @younik in Fix flatten_state_dict with empty dictionary #700

• [BugFix] Fix issue with demo_manual_control.py by @StoneT2000 in [BugFix] Fix issue with demo_manual_control.py #702

• [Feature] Auto-generated task documentation by @StoneT2000 in [Feature] Auto-generated task documentation #703

• Added custom task (PullCubeTool-v1) with motion planner + Motion Planner for LiftPegUpright and PullCube by @Viswesh-N in Added custom task (PullCubeTool-v1) with motion planner + Motion Planner for LiftPegUpright and PullCube #641

• [Docs] Update docs on PullCubeTool-v1, and upload new demonstrations to HF for motion planning generated datasets by @StoneT2000 in [Docs] Update docs on PullCubeTool-v1, and upload new demonstrations to HF for motion planning generated datasets #704

• [BugFix] Fix kinematics that sometimes will skip an active joint by @StoneT2000 in [BugFix] Fix kinematics that sometimes will skip an active joint #705

• [BugFix] fix bug when a torch based pose instead of sapien pose is passed to RRT planning code by @StoneT2000 in [BugFix] fix bug when a torch based pose instead of sapien pose is passed to RRT planning code #706

• [BugFix] Remove extra warnings and clarify why initial poses should be set by @StoneT2000 in [BugFix] Remove extra warnings and clarify why initial poses should be set #709

• [Feature] More humanoid/bimanual manipulation tasks, fix G1 robot's frictions, optimize robot collisions for franka and G1. Add G1 with head camera, allow record with info for parallel envs by @StoneT2000 in [Feature] More humanoid/bimanual manipulation tasks, fix G1 robot's frictions, optimize robot collisions for franka and G1. Add G1 with head camera, allow record with info for parallel envs #715

• [BugFix] Fix bug for recording episodes with CPU envs by @StoneT2000 in [BugFix] Fix bug for recording episodes with CPU envs #718

• [Docs] Documentation on even more Domain Randomization by @StoneT2000 in [Docs] Documentation on even more Domain Randomization #719

New Contributors

• @younik made their first contribution in Fix flatten_state_dict with empty dictionary #700
• @Viswesh-N made their first contribution in Added custom task (PullCubeTool-v1) with motion planner + Motion Planner for LiftPegUpright and PullCube #641

Full Changelog: v3.0.0b12...v3.0.0b13