Michael S. Koehle

University of British Columbia, Autodesk (United States)

Papers

Total Citations

H-Index

About

Michael S. Koehle is a roboticist whose research bridges human motor control and autonomous construction, with a focus on enabling robots to perform complex, contact-rich tasks. His work is centered on two key areas: understanding the biomechanical principles of human balance and applying those insights to develop adaptive, long-horizon robotic manipulation. In his highly cited 2014 study, Koehle experimentally evaluated human balance control models, demonstrating how intermittent muscle activation and predictive mechanisms overcome sensorimotor delays—a foundational contribution to both rehabilitation robotics and humanoid stability. More recently, he has pioneered adaptive robotic construction of wood frames (2024) and introduced the ARCH hierarchical hybrid learning framework for long-horizon, contact-rich robotic assembly (2024). This latter work tackles the critical challenge of generalizable assembly by combining imitation learning with hierarchical reasoning, reducing the need for massive demonstration datasets while achieving high precision. With over 20 citations across his most prominent papers, Koehle’s impact lies in translating biological control principles into scalable robotic systems, advancing the frontier of autonomous manufacturing and assistive robotics.