Max Robotics — Global Service Robot Directory + US Certification

About

Taku Komura is a distinguished researcher whose work spans robotics, biomechanics, character animation, and motion planning, with particular emphasis on biped locomotion, balance control, and human motion simulation. His foundational contributions in the mid-2000s established new paradigms for humanoid robot stability, most notably through angular momentum-based feedback controllers capable of counteracting large perturbations during gait — work that has garnered nearly 100 citations. His enhanced Linear Inverted Pendulum Model, applied to simulating pathological gait caused by muscle weakness, bridges robotics and clinical biomechanics in a uniquely impactful way. Komura's development of quadratic programming methods for dynamic postural adjustment and C² continuous gait-pattern generation further solidified his reputation as a leading figure in physically plausible motion synthesis. His later research expanded into topology-based motion representations for complex robot interactions and hierarchical deep reinforcement learning for push-recovery behaviors, demonstrating impressive adaptability across evolving paradigms. More recently, his work on predicting human manipulation of large objects reflects a growing interest in human-robot interaction and intent recognition. With cumulative citations exceeding 500 across his top works, Komura's research continues to meaningfully influence both the robotics and computer animation communities.