Adaptive gait control for a walking robot

Abstract

Abstract Walking vehicles have the potential to emulate the superior off‐road mobility of biological systems. However, it is important to make the walking machine terrain adaptive and versatile, and to minimize man's role as an operator in order to realize this potential. Terrain adaptive locomotion involves intelligent foothold selection and the control of gait to produce the desired motion. This requires a departure from the idealized, structured stepping patterns for statically stable gaits which have been the object of considerable research. A modified wave gait is used to demonstrate that it is possible for the vehicle velocity to be varied continuously in accordance with higher level commands even with irregular, asymmetric, and changing support patterns, A varying duty factor is employed to enable optimal leg cycling frequencies. Implementation of gait control algorithms and results from a computer simulation are also presented.