Max Robotics — Global Service Robot Directory + US Certification

About

E. Schuitema is a robotics researcher whose work tackles one of the most persistent challenges in autonomous systems: enabling real robots to learn from experience through Reinforcement Learning (RL). Her research focuses on bipedal walking, motor control, and the practical application of RL to physical hardware—a notoriously difficult problem where simulated successes often fail to transfer to the real world. Her most influential contribution is a memoryless approach to handling control delay in RL for real-time dynamic systems (56 citations), a fundamental issue that arises from the lag between sensing and acting in physical robots. She also demonstrated RL-based control for a passive dynamic walking robot (46 citations), addressing the stability challenges that arise from uneven terrain. Schuitema designed LEO, a 2D bipedal walking robot specifically built for online autonomous RL (45 citations), establishing key hardware and software requirements for such experiments. Her later work on parallel online temporal difference learning (16 citations) offers a faster alternative to policy search methods for motor control. Through these contributions, Schuitema has advanced the frontier of bringing RL out of simulation and onto real, walking robots.