Max Robotics — Global Service Robot Directory + US Certification

About

Nathan Batchelor is a leading researcher at the intersection of robotics and artificial intelligence, specializing in deep reinforcement learning for agile, full-body robot control. His work focuses on bridging the gap between simulated training and real-world deployment, particularly for legged robots. Batchelor’s most impactful contribution is his pioneering research on teaching a low-cost, miniature humanoid robot to play one-versus-one soccer using deep RL—a project that has garnered 147 citations and demonstrated how complex, dynamic behaviors can be synthesized safely. He also developed the "Imitate and Repurpose" framework, which reuses motion capture data from humans and animals to create reusable locomotion skills for real robots. Batchelor contributed to the "Barkour" benchmark, setting standards for animal-level agility in quadruped robots, and was part of the team behind "RoboCat," a self-improving generalist agent for robotic manipulation. His work on egocentric vision-based robot soccer further pushes the boundaries of onboard computation and perception. With over 200 total citations, Batchelor’s research is shaping the future of autonomous, agile robots capable of navigating and interacting with complex, unstructured environments.