Uploaded April 2026 | Updated November 2026, 3 hours ago
Built for the toughest terrain 🤖🏁
At the off-road track, Lynx M20 demonstrates what true all-terrain mobility looks like — confidently navigating rugged paths, uneven ground, and unpredictable surfaces under real racing conditions.
#DEEPRobotics #LynxM20 #AllTerrain #Robotics hashtag#EmbodiedAI #Innovation
Built for the toughest terrain 🤖🏁
At the off-road track, Lynx M20 demonstrates what true all-terrain mobility looks like — confidently navigating rugged paths, uneven ground, and unpredictable surfaces under real racing conditions.
#DEEPRobotics #LynxM20 #AllTerrain #Robotics hashtag#EmbodiedAI #Innovation




![Lite3| [Perception] [HardwareExtension] Compute and Sensor Extension,based on reproducing FAST-LIVO2
This video introduces how to mount external compute and sensors on the lite3 robot and reproduce a complex SLAM algorithm called FAST-LIVO2. The links mentioned in the video:
librealsense(including realsense viewer): https://github.com/realsenseai/librealsense
livox viewer: https://www.livoxtech.com/downloads
nomachine: https://www.nomachine.com/
fast-livo2-deep-robotics: https://github.com/DeepRoboticsLab/fast-livo2-deep-robotics
We didnt use the third cable coming out of the lidar to enable hard synchronization and because of that, the SLAM algorithm was unable to continue. This is not a problem of the algorithm. We encourage you guys to perfect this project by adding synchronization. Lite3| [Perception] [HardwareExtension] Compute and Sensor Extension,based on reproducing FAST-LIVO2](https://i.ytimg.com/vi/hwtDnuVs-OI/mqdefault.jpg)





