Research On Gait Planning Of Humanoid Soccer Robot

Humanoid robot is a very challenging research direction of the robot field. It involves multiple areas, like robotics, artificial intelligence, automatic control and so on. Biped walking is a significant feature of the humanoid robot. So planning stable and fast walking gait has become a hotspot in the research of the humanoid robot. In this paper, a small humanoid soccer robot is used as the research object. According to the hardware of the humanoid soccer robot and the characteristics of the robot soccer competition, the robot gait is planned and simulated. Finally, the planned gait is validated on the humanoid soccer robot.The main work of the paper is listed below:1. Cubic spline interpolation and high order polynomial interpolation are used for gait planning. Firstly, the seven link model is established based on the structure of the robot. Secondly, a complete biped walking cycle of the humanoid soccer robot is analyzed. The motion process is divided into three stages as start, cycle walking and stop. Thirdly, the hip and the ankle are different during the moment. So a cubic spline interpolation and a high order polynomial interpolation are used to plan their trajectories respectively. Finally, the trajectories of the knee joint are deduced by the geometric.2. In order to get stable walking gait, the gait parameters are optimized by genetic algorithm. Firstly, the ZMP criterion of the stable biped walking is discussed. Secondly, the gait optimization objective function is constructed based on the ZMP criterion. Finally, by using GA to optimize the parameters of the hip joint, the planned gait of humanoid soccer robot can get the largest stability margin.3. The simulation experiment of the virtual prototype is made in ADAMS. Firstly, by using the inverse kinematics, the joint angle trajectory of humanoid soccer robot can be calculated in MATLAB. Secondly, the virtual prototype model of humanoid soccer robot is built in ADAMS. And the drive is imported into each joint. Thirdly, the simulation results verify the validity of the gait planning through the dynamics simulation in ADAMS. Finally, the planning gait is used on real humanoid soccer robot. As a result, the robot walks stable and fast.