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.
https://github.com/user-attachments/assets/211e8b87-6b81-460e-841f-081979d1ea35
Full Changelog:
What's Changed
* [Fix] Use same indices for root vel setter as base uses for articulations by arth-shukla in https://github.com/haosulab/ManiSkill/pull/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 https://github.com/haosulab/ManiSkill/pull/668
* [BugFix] Fix record episode wrapper storing all parallel env episode seeds instead of just one. by StoneT2000 in https://github.com/haosulab/ManiSkill/pull/671
* [Bug] Fix: (unideal) only do .step() if not doing partial reset by arth-shukla in https://github.com/haosulab/ManiSkill/pull/672
* [BugFix] Fix demo_vis_segmentation script by StoneT2000 in https://github.com/haosulab/ManiSkill/pull/675
* [Feature] Unified states for heterogenous simulated envs by StoneT2000 in https://github.com/haosulab/ManiSkill/pull/676
* [Docs] Add docs on simulation properties like object textures and controller properties that can be dynamically updated during runtime. by StoneT2000 in https://github.com/haosulab/ManiSkill/pull/681
* [Feature] Support normals and albedo in observation data by StoneT2000 in https://github.com/haosulab/ManiSkill/pull/682
* [Feature] Added batched input support for look_at matrix construction by Xander-Hinrichsen in https://github.com/haosulab/ManiSkill/pull/678
* [Feature[ Update interactive_panda.py to allow using different viewer shaders by StoneT2000 in https://github.com/haosulab/ManiSkill/pull/687
* [Feature] Extremely fast PPO by StoneT2000 in https://github.com/haosulab/ManiSkill/pull/669
* [BugFix] Pose creation always creates a new object now and does not duplicate tensors. by StoneT2000 in https://github.com/haosulab/ManiSkill/pull/688
* [BugFix] Fix PD EE Delta Pose controller that would sometimes do in place modification of user given action by StoneT2000 in https://github.com/haosulab/ManiSkill/pull/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 https://github.com/haosulab/ManiSkill/pull/690
* [BugFix] Fix bug where consistency check for state in env state recording assumes actors/articulations exist by StoneT2000 in https://github.com/haosulab/ManiSkill/pull/691
* [Feature] Organized data generation scripts by StoneT2000 in https://github.com/haosulab/ManiSkill/pull/692
* [Feature] SAC RGB(D) Baseline and improved memory efficiency by StoneT2000 in https://github.com/haosulab/ManiSkill/pull/698
* Fix flatten_state_dict with empty dictionary by younik in https://github.com/haosulab/ManiSkill/pull/700
* [BugFix] Fix issue with demo_manual_control.py by StoneT2000 in https://github.com/haosulab/ManiSkill/pull/702
* [Feature] Auto-generated task documentation by StoneT2000 in https://github.com/haosulab/ManiSkill/pull/703
* Added custom task (PullCubeTool-v1) with motion planner + Motion Planner for LiftPegUpright and PullCube by Viswesh-N in https://github.com/haosulab/ManiSkill/pull/641
* [Docs] Update docs on PullCubeTool-v1, and upload new demonstrations to HF for motion planning generated datasets by StoneT2000 in https://github.com/haosulab/ManiSkill/pull/704
* [BugFix] Fix kinematics that sometimes will skip an active joint by StoneT2000 in https://github.com/haosulab/ManiSkill/pull/705
* [BugFix] fix bug when a torch based pose instead of sapien pose is passed to RRT planning code by StoneT2000 in https://github.com/haosulab/ManiSkill/pull/706
* [BugFix] Remove extra warnings and clarify why initial poses should be set by StoneT2000 in https://github.com/haosulab/ManiSkill/pull/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 https://github.com/haosulab/ManiSkill/pull/715
* [BugFix] Fix bug for recording episodes with CPU envs by StoneT2000 in https://github.com/haosulab/ManiSkill/pull/718
* [Docs] Documentation on even more Domain Randomization by StoneT2000 in https://github.com/haosulab/ManiSkill/pull/719
New Contributors
* younik made their first contribution in https://github.com/haosulab/ManiSkill/pull/700
* Viswesh-N made their first contribution in https://github.com/haosulab/ManiSkill/pull/641
**Full Changelog**: https://github.com/haosulab/ManiSkill/compare/v3.0.0b12...v3.0.0b13