XLeRobot 0.2.3 (XLeVR) out!. • VR (Quest 3) Whole-body Control.• Reads everything: head and hand poses, joysticks, and all the buttons.• Minimal dependencies, web-based.• Modular, can use it on other robots.• Open-source, easy install, play in 20min

XLeRobot 0.2.4: VR control on a real LeRobot SO101 arm. Wireless 🛜, Quest 3, and built-in analytical IK for the fastest computation. Seldom jerkiness due to WiFi network lag (will also happen to leader-follower control). Open-sourced at:

XLeRobot Sim2real RL:. Only 1hr RL training in pure simulation (Maniskill) with a 4090 GPU and then we got this. Faster and smoother than most of the VLA models I have seen on LeRobot. Codes and tutorials fully open-sourced:. Many thanks to @Stone_Tao!

Next step

Upgrade to rerun for visualization and data management to be compatible with LeRobot dataset. Showing XLeRobot brings the salt and pepper to your table and push over your bike.

A short update on XLeRobot:. A fully functional simulation system for the new 0.2.0 hardware and integrated wrist RGB cameras and a RGBD head camera. Get started in 10min at the new documentation website:.

RT @YutongBAI1002: What would a World Model look like if we start from a real embodied agent acting in the real world?. It has to have: 1)….

Many thanks to the Bambu Lab official for giving me the H2D combo for free! .@BambulabGlobal . This will help me to develop XLeRobot hardware upgrades faster and work on hard-soft hybrid printing.

RT @Stone_Tao: Code/tutorial released for🤗LeRobot zero-shot visual sim2real Reinforcement Learning!. Train your SO100 robot in ManiSkill's….

Caging in Time is officially on IJRR! . We proposed a novel theory framework that deals with uncertainties by cages in time dimension for robust object manipulation such as random object pushing and tennis balancing and catching, all in pure open loop.