Biologically Inspired Planning and Optimization of Foot Trajectory of a Quadruped Robot

Foot trajectory plays an important role in high speed running of cheetahs, so it is very significant to design an appropriate foot trajectory by mimicking cheetahs to improve the performance of quadruped robots. This paper presents a Bezier curve based foot trajectory planning approach for a quadruped robot. Four 5th-order Bezier curves are employed to plan the robot's foot trajectory in the sagittal plane, which is inspired by the foot trajectory characteristics of cheetahs. Particle swarm optimization algorithm is used to determine the foot trajectory parameters with the aim of minimizing the jerk of the trajectory and impact as the feet touch the ground. Finally, the joints trajectories are computed through the inverse kinematic of the quadruped robot. Moreover, the foot trajectory is parametrized for applying to robots with different sizes. The dynamic simulation results show that the generated foot trajectory is effective to approximately achieve the desired speeds.