Max Robotics — Global Service Robot Directory + US Certification

About

Jie Tan is a pioneering researcher in robotic locomotion and deep reinforcement learning, whose work has fundamentally advanced how legged robots learn to move autonomously. Operating at the intersection of machine learning and robotics, Tan's research focuses on automating the acquisition of complex motor skills—eliminating the need for tedious manual controller design that has historically constrained the field. His most influential contribution, "Learning to Walk Via Deep Reinforcement Learning" (2019, 434 citations), demonstrated that deep RL could enable robots to map sensory inputs directly to low-level actions, bypassing explicit programming. This work, alongside his landmark sim-to-real transfer research (2018, 114 citations), established foundational frameworks for training quadruped robots in simulation and deploying them successfully in the real world—a notoriously difficult challenge. Tan has further pushed boundaries by drawing inspiration from biology, developing systems that imitate animal locomotion strategies and benchmarking quadrupedal agility against natural movement standards through the Barkour framework. His explorations of data-efficient learning, meta-learning for rapid adaptation, and loco-manipulation reflect a sustained commitment to making robots more versatile and autonomous. Collectively, his body of work has shaped modern expectations for what agile, learning-driven robotic systems can achieve.