Fabian Canas

Florida Institute for Human and Machine Cognition, University of Michigan–Ann Arbor

Papers

Total Citations

115

H-Index

About

Fabian Canas is a leading figure in the field of bipedal locomotion and humanoid robotics, best known for his pioneering work on push recovery and dynamic balance. His most significant contribution is the development of "Capture Points"—a groundbreaking concept that identifies the precise location on the ground where a biped must step to come to a complete stop after a disturbance. This framework, introduced in his highly influential 2007 paper (79 citations), provides a computationally efficient method for robots to recover from pushes, mimicking human-like stepping reflexes. Canas further advanced this work through a 2008 study (15 citations) that applied machine learning techniques to predict capture points, enabling more robust and adaptive responses to unexpected forces. He also played a key role in the development of the Yobotics-IHMC Lower Body Humanoid Robot (2009, 21 citations), a 12-degree-of-freedom platform with series elastic actuators. This robot demonstrated the practical application of capture point theory using virtual model control, showcasing how theoretical insights could be translated into real-world stability. Canas’s research has been instrumental in bridging the gap between human balance strategies and robotic control, with his capture point framework now a standard tool in humanoid robotics. His work continues to inspire new generations of researchers tackling the challenge of making bipedal robots as agile and resilient as humans.