Research Of Actions Plan And Adaptive Arithmetic Trajectory Tracking For Humanoid Soccer Robot

Humanoid robot is the top class of robots. Being the centralized embodiment of intelligent robot theory and technology,it can lead to the intersect development and advancement of many related subjects and technologies.After nearly ten years of robot international football match began, humanoid robots finally appear in robot soccer games. The research of humanoid robots is just beginning, so it has a very important significanceIn this thesis, mini-type humanoid soccer robot is taken as the main subject of research and action plan through simulation. According to the unpredictable outside force in competition circumstance, the adaptive controller is designed and the arithmetic trajectory tracking of robot joints is done.Firstly, the model is established. Takes the ROBONOVA-I as prototype and draw the robot, ground and sphere with Solidworks. The model gets higher comparability in the shape basing on the dimension of reality.Secondly, single joint actions plan. Design some actions through simulation and modify the contact parameter among robot,ground and sphere by taking advantage of virtual prototype technology of ADAMS. Ensure that the effect is similar to reality when the data from simulation apply to the reality. The data stored from single joint actions plan can be used in the complex actions of the follow-up process.The relative complex actions were planded based on the planned actions above. The actions were planded with the co-simulation of ADAMS and Simulink by making use of the perfect control technology of Simulink.To ensure the stability, the theory of ZMP is imported. By putting forward the concept of the ideal and actual track of ZMP simulation, quantize the degree of stability. Taking the actual track of ZMP simulation as feedback into the control system and enhancing the stability by modifying the joint angle of coxa, finally, the stability of robots is enhanced on condition that the prescribed actions are finished.The adaptive controller is designed aimed at the uncertainty force existed in the racing circumstance and the error between the track of the joint action and the ideal track. This way don't need the configuration and parameter of the robot, also don't need to calculate regression matrix. The controller has simple configuration and better controlling capability. It can make the error of trajectory tracking close to zero, and finally the whole control system will receive a better performance.