Step-by-step tutorial at https://learn-robotics.pathon.ai/
- General Introduction to SO-ARM100 and LeRobot: https://learn-robotics.pathon.ai/blog/soarm100
- Using LeRobot in Simulation Environment: https://learn-robotics.pathon.ai/blog/simulation
- Control the Open-Source Robot Arm with ACT (Action Chunking with Transformers) Using LeRobot: https://learn-robotics.pathon.ai/blog/lerobot_act
- Control the Open-Source Robot Arm with Pi0 Using LeRobot: https://learn-robotics.pathon.ai/blog/lerobot_pi0
| Platform | Type | Description | Repository |
|---|---|---|---|
| SO-100 & SO-101 | Robotic Arms | Standard Open robotic arms | https://github.com/TheRobotStudio/SO-ARM100 |
| LeKiwi | Mobile Manipulator | Low-Cost Mobile Manipulator for so-100/101 arm | https://github.com/SIGRobotics-UIUC/LeKiwi |
| XLeRobot | Mobile Manipulator | Built on top of LeKiwi | https://github.com/Vector-Wangel/XLeRobot |
| Bambot | Mobile Manipulator | Built on top of LeKiwi | https://github.com/timqian/bambot |
| Policy | Full Name | Description | Paper |
|---|---|---|---|
| ACT | Action Chunking with Transformers | Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware | https://arxiv.org/abs/2304.13705 |
| Diffusion Policy | Diffusion Policy | Visuomotor Policy Learning via Action Diffusion | https://arxiv.org/abs/2303.04137 |
| TD-MPC | Temporal Difference Learning for Model Predictive Control | Model Predictive Control with Temporal Difference Learning | https://arxiv.org/abs/2203.04955 |
| FOWM | Finetuning Offline World Models | Finetuning Offline World Models in the Real World | https://arxiv.org/abs/2310.16029 |
| VQ-BeT | Vector-Quantized Behavior Transformer | Behavior Generation with Latent Actions | https://arxiv.org/abs/2403.03181 |
| π0 | Pi-Zero | A Vision-Language-Action Flow Model for General Robot Control | https://www.physicalintelligence.company/download/pi0.pdf |
https://github.com/DelinQu/awesome-vision-language-action-model
There are different ways to teleoperate the follower arm
- in lerobot package, the official way is to use a leader arm to control the follower arm
- there are other ways to control the follower arm, like
- using joycon: https://github.com/box2ai-robotics/joycon-robotics
- using VR like meta quest 3: https://github.com/DipFlip/telegrip
- data conversion: https://github.com/Tavish9/any4lerobot
- data explorer
- data operation
- data set cleaning
- data manipulation
- delete episode, combine dataset
- sim2real: https://github.com/StoneT2000/lerobot-sim2real
- digital twin: https://github.com/LiZhYun/BiLerobot
- isaac lab