Motion Control Of Humanoid Robot In RoboCup Standard Platform League

Robotics is one of the hotspots of artificial intelligence research at present.As an important branch of robotics,humanoid robot research covers multiple disciplines and has a wide range of applications in different fields.This paper takes the RoboCup standard platform league as the research background,takes the NAO robot used in the competition as the research object,and mainly studies the motion control of the robot in the process of the standard platform group competition.carry out research.The main work is as follows:Firstly,in order to solve the problem of slow and unstable walking speed of the robot in the competition environment,a single-support phase walking gait based on centroid control and sensor feedback is proposed.In this method,the NAO robot is modeled as a 3D inverted pendulum model to analyze the movement of the torso and feet during the walking process of the robot.On this basis,the walking gait of the single support phase is generated and the inverse kinematics is used to calculate the angles of each joint.At the same time,fall detection is added based on the allowable ZMP region(AZR)method.Finally,through simulation and real robot experiments,it is proved that this method can effectively improve the walking speed and robustness of the robot.Secondly,for robot football games like the standard platform league,the kicking action design of the humanoid biped robot is a very important part,and an excellent kicking action can often change the whole situation.Aiming at this situation,a robot kicking algorithm based on motion engine is proposed.The approximate kicking trajectory is fitted by the Bezier curve,and whether the action is continuous or not is judged by the continuous differentiable condition.Then,the stability of the action is judged by the balance of the center of mass and the feedback of the gyroscope,and then the parameters of the kick are optimized by the improved CMA-ES algorithm.Finally,the experiment proves that the kicking algorithm is efficient and feasible.Finally,for the robot path planning problem in the dynamic environment of the competition,a path planning method based on synchronous visible view and improved A~* is proposed.This method firstly optimizes the visible view method,introduces crossing lines to reduce the number of generated visible edges,and at the same time intelligently selects target points in different situations according to team communication and competition rules,thereby reducing the computational complexity of the algorithm and improving the execution speed of the robot.Experiments are carried out in static and dynamic environments,respectively,to demonstrate the effectiveness of the method.