An impact recovery approach for quadruped robot with trotting gait

Abstract

Legged robots are superior in terms of agility and versatility while inferior in stability compared to their wheeled or tracked counterparts. Researchers have developed some algorithms for quadruped robots to achieve impact absorption. Unfortunately those robots all refer to dynamic control and have high demands on the hardware performance. In this paper, we present an approach for quadruped robot with trotting gait to deal with unexpected impacts using position control. The robot could detect the impact and takes two steps to control attitude of the trunk and stabilize it. We demonstrated the practicability of the approach by performing experiments on the servo-actuated quadruped robot LittleCalf and its model in simulation Results from the experiments showed that the control method is able to recover a trotting quadruped robot from unexpected impact of a certain magnitude.