Max Robotics — Global Service Robot Directory + US Certification

About

Jinze Wu is a leading researcher in legged robotics, specializing in reinforcement learning (RL) for robust and agile locomotion. His most impactful contribution is the development of the first blind locomotion system capable of traversing challenging terrains at high speeds, a breakthrough detailed in his highly cited 2023 work, *Learning Robust and Agile Legged Locomotion Using Adversarial Motion Priors* (90 citations). This work introduced the Adversarial Motion Priors (AMP) framework, enabling robots to dynamically switch between robust and agile behaviors using only proprioception. Wu has further advanced multi-gait learning by designing end-to-end RL frameworks that allow quadruped robots to smoothly transition between gaits based on terrain and velocity commands, as seen in his 2024 paper on skill latent spaces. Beyond locomotion, he has addressed critical challenges in human-robot interaction, notably perspective disambiguation in language-driven manipulation, where his 2022 work on placement optimization ensures robots interpret commands accurately from different viewpoints. Wu’s research consistently pushes the boundaries of what legged robots can achieve, bridging the gap between robust stability and agile, natural movement.