Research on Gait Planning and Inverse Kinematics Solving of Biped Walking Robots

Gait planning and inverse kinematics are becoming more complex with the development of biped walking robots. In this paper we propose a new method to simplify these processes. By applying this method, the angle variation based on the walking pose at each moment can be figured out simultaneously and straightforward without complex calculations. The method starts from a linear inverted pendulum model in three-dimensional Cartesian coordinates which is used for planning the centroid and the walking pose of the robot. Then the kinematics equations in each discrete joint are established to obtain the trajectory of each joint, followed by adopting the space vector angle to solve the robot´s inverse kinematics. We finally establish a simulation model of the biped robot using Matlab to test the validity of our method. The results indicate that stable walking and simplified inverse kinematics solving can be realized using our method.