Mechanical and Civil Engineering Seminar

A major challenge in understanding biological movement and advancing robotic mobility in natural and artificial environments is the relative lack of advanced understanding of the physics of locomotor-ground interaction in complex terrain. In this talk, I discuss my research at the interface of engineering, physics, and biology to create terradynamics, analogous to the well-established aero- and hydrodynamics which describe fluid-structure interaction and allow quantitative predictions of forces and movement in fluids. Using controlled granular media as the next step to define flowable ground (e.g., sand, soil, mud, Martian soil), I systematically study legged movement of robots and animals to reveal capabilities and limits, and discover transitions from rapid movement to failure to move which depend sensitively on ground properties, leg kinematics, and leg morphology. Using physics experiments to systematically measure intrusion forces, I elucidate the principles of legged movement on granular media and develop a resistive force theory—the first terradynamic model capable of accurately and rapidly predicting forces and movement in complex terrain. In expanding terradynamics into 3D multicomponent terrain (e.g., dense vegetation, cluttered rooms, building rubble), I study insects moving through controlled, grass-like, cluttered beam obstacles, and discover the new concept of terradynamic shapes that facilitate movement and provide new bio-inspired design principles for robots. Finally, I posit that, in concert with aero- and hydrodynamics, an emerging field of terradynamics will not only advance understanding of how animals move in nature, but also facilitate development of robots with locomotor capabilities approaching those of organisms in real-world environments.