A New Trajectory Tracking Control Method for Fully Electrically Driven Quadruped Robot

To improve the accuracy of tracking trunk center-of-mass (CoM) trajectory and foot-end in a fully electrically driven quadruped robot, an efficient practical new control method is designed. The proposed mainly divided into balance controller (TBC) swing leg (SLC). In TBC, robot dynamics model developed to find optimal force that follows CoM based on predictive (MPC) principle. SLC, Bessel curve planned as desired at foot-end, while tracked by virtual spring-damping element driving meanwhile, radial basis function neural network (RBFNN) applied for supervisory performance system. experimental results show can modify robot’s effect, so stability error be eliminated robustness improved, linear circular accurately quickly.