Active perception of material and shape by a walking robot

Abstract

A legged robot can use its legs for more than locomotion. It can use them to identify the material properties and shape of the terrain it traverses. The author formulates active perception techniques based on force/torque sensing to identify terrain stiffness and surface friction, and present first results in applying the techniques to characterize terraine traversability. The author also describes purposeful contact sensing to update geometric terrain maps in a process that might be called 'shape from kicking'.