The body as a reservoir: locomotion and sensing with linear feedback

Abstract

It is known that mass-spring nets have computational power and can be trained to reproduce oscillating patterns. In this work, we extend this idea to locomotion and sensing. We simulate systems made out of bars and springs and show that stable gaits can be maintained by these structures with only linear feedback. We then conduct a classification experiment in which the system has to distinguish terrains while maintaining an oscillatory pattern. These experiments indicate that the control of compliant robots can be simplified if one exploits the computational power of the body’s dynamics.