Multi-objective Hierarchical Optimal Control for Quadruped Rescue Robot

Abstract

Timely and efficient rescue missions in disasters such as nuclear accidents, fires, and earthquakes require operating robots to work under high payload. This paper treats the problem of optimal planning and control of a walking-trot gait for a hydraulically actuated quadruped robot that has been developed in a collaborative project of the Departments of Mechanical Engineering and Automation at Shanghai Jiao Tong University. To mitigate the challenges of working with high-fidelity models of this highly complex mechanical system, a kinematics and approximate dynamics model is proposed that is shown to provide an accurate description of the walking motions we wish to study while at the same time being simple enough to support analysis. The proposed model is used to design limb movements in a walking trot gait that balance the trade off between energy consumption and the speed of execution of a desired mission. Bezier curves of different orders are used to design trajectories of the robot’s limbs. We verify the accuracy of the proposed model through experiments with the actual robot. We apply hierarchical optimization method to minimize the energy consumption, and we discuss a Pareto optimal solution that trades off mission duration and energy consumption. In the end perform several experiments to verify the effectiveness and superiority of the optimal algorithm we proposed.