Max Robotics — Global Service Robot Directory + US Certification

About

Shimpei Masuda is a robotics researcher whose work sits at the intersection of humanoid locomotion, dexterous manipulation, and sim-to-real transfer. His primary contributions span three critical areas: bridging the simulation-to-reality gap for bipedal robots, advancing whole-body contact manipulation through tactile sensing, and developing novel hardware for fine-grained object insertion. Masuda’s most cited work, "Sim-to-Real Transfer of Compliant Bipedal Locomotion on Torque Sensor-Less Gear-Driven Humanoid" (11 citations), tackles the persistent challenge of transferring deep reinforcement learning policies to physical hardware, particularly addressing actuator dynamics that often derail real-world performance. His 2024 paper "SAID-NeRF" (8 citations) introduces an innovative approach to depth completion for transparent objects—a notoriously difficult perception problem—by integrating neural radiance fields with segmentation. Masuda’s 2025 work on whole-body contact manipulation (3 citations) demonstrates how humanoid robots can leverage full-body contact for enhanced stability during manipulation tasks, while his "FAAF Hand" (2024) presents a four-axis adaptive finger design that compensates for localization errors during insertion tasks. His research consistently emphasizes practical, hardware-aware solutions that push humanoid robots toward real-world deployment.