Uploaded October 2025 | Updated November 2026, 3 hours ago
Episode 2: Training code explanation, reward shaping, simulation configuration
Recent years, the robotics industry has gained increasing popularity, with quadrupedal robots attracting particular attention. To encourage more universities and educational institutions to participate in this trend, DEEP Robotics has decided to use the Lite3 platform to release a series of instructional videos introducing quadrupedal robot motion control, providing in-depth analysis of algorithms and code implementation. For any questions, feel free to join our Discord server for discussion and exchange. Relevant code links are as follows:
Training Code: github.com/DeepRoboticsLab/rl_training
Deployment Code: github.com/DeepRoboticsLab/Lite3_rl_deploy
We will also release videos on the wheeled-quadruped Lynx M20 and humanoid robots in the future. Follow us on:
YouTube:youtube.com/@deeprobotics8601
Bilibili:space.bilibili.com/22477177
Discord:discord.gg/gdM9mQutC8
Episode 2: Training code explanation, reward shaping, simulation configuration
Recent years, the robotics industry has gained increasing popularity, with quadrupedal robots attracting particular attention. To encourage more universities and educational institutions to participate in this trend, DEEP Robotics has decided to use the Lite3 platform to release a series of instructional videos introducing quadrupedal robot motion control, providing in-depth analysis of algorithms and code implementation. For any questions, feel free to join our Discord server for discussion and exchange. Relevant code links are as follows:
Training Code: github.com/DeepRoboticsLab/rl_training
Deployment Code: github.com/DeepRoboticsLab/Lite3_rl_deploy
We will also release videos on the wheeled-quadruped Lynx M20 and humanoid robots in the future. Follow us on:
YouTube:youtube.com/@deeprobotics8601
Bilibili:space.bilibili.com/22477177
Discord:discord.gg/gdM9mQutC8

![[DR02 Pro][Locomotion] [RL] DR02 humanoid robot locomotion, open-source rl training code explanation
In this video, we break down the RL training code for DeepRobotics DR02 humanoid robot locomotion task.
We cover five main parts: code structure, robot model, training environment setup, training pipeline, and training algorithm.
By the end, youll have a solid understanding of how humanoid locomotion training works — the principles, the workflow, and the practical tricks.
Related links:
Training code repo: https://github.com/DeepRoboticsLab/rl_training
Deployment code repo: https://github.com/DeepRoboticsLab/deep-robotics-sdk2
Quadruped Lite3 video: https://www.youtube.com/playlist?list=PLy9YHJvMnjO0X4tx_NTWugTUMJXUrOgFH
Wheeled-legged M20 video: https://www.youtube.com/watch?v=5VKMahWWYM4 [DR02 Pro][Locomotion] [RL] DR02 humanoid robot locomotion, open-source rl training code explanation](https://i.ytimg.com/vi/xmptqd-LDUs/mqdefault.jpg)



![[Lite3] [Locomotion] [ROS2] ROS2 migration and quadruped modularization based on locomotion of Lite3
This video introduces the ROS2 migration of our Lite3 RL deploy code and how to modulize our lite3 robot into a complex robot system. The links mentioned in the video:
lite3_rl_deploy: https://github.com/DeepRoboticsLab/Lite3_rl_deploy
ros2 foxy: https://docs.ros.org/en/foxy/Installation.html
ros2 humble: https://docs.ros.org/en/humble/Installation.html [Lite3] [Locomotion] [ROS2] ROS2 migration and quadruped modularization based on locomotion of Lite3](https://i.ytimg.com/vi/z9SK1B06MNY/mqdefault.jpg)

