Robot Learning Tutorial
This directory contains theoretical and practical guides related to robot learning, covering robot intelligent decision-making methods such as reinforcement learning and imitation learning.
Existing demo: Multiple robots for moving furniture
Table of Contents
Reinforcement Learning
- Basic reinforcement learning algorithms (to be added)
- Deep reinforcement learning applications (to be added)
- Robot reinforcement learning environment setup (to be added)
Imitation Learning
- Behavior cloning (to be added)
- Inverse reinforcement learning (to be added)
- Human demonstration data collection (to be added)
Multimodal Learning
- Visual-tactile integration (to be added)
- Language-guided robot learning (to be added)
Learning Path
- Master the fundamentals of machine learning (such as supervised learning, unsupervised learning)
- Understand the basic principles of reinforcement learning (such as Markov decision processes, policy gradients, etc.)
- Learn deep learning frameworks (such as PyTorch, TensorFlow)
- Practice robot learning algorithms (starting with simple tasks)
- Explore cutting-edge research directions (such as meta-learning, multi-agent learning, etc.)
Recommended Tools and Frameworks
- Python 3.8+
- PyTorch/TensorFlow
- Gym/MuJoCo
- RLlib/Stable Baselines
- Robot simulation environments (such as our simulation series experiments)
Project Examples
- Reinforcement learning for robotic arm grasping
- Imitation learning of robotic arm trajectory
- Multi-agent collaboration task learning
- Robot manipulation guided by language commands
Reference Resources
- Reinforcement Learning: An Introduction by Richard S. Sutton and Andrew G. Barto
- Deep Learning by Ian Goodfellow, Yoshua Bengio, and Aaron Courville
- OpenAI Spinning Up
- Berkeley Robot Learning Lab
- Stanford Vision and Learning Lab