Research On Path Planning Based On Improved A-star Algorithm Of Humanoid Robot

Compared with other types of robots,humanoid robot has obvious human like characteristics,unique advantages,can better integrate into the human society,and is more convenient for serving humans.With the development of artificial intelligence and automation,the research of humanoid robot is becoming more and more in-depth.“Human-robot commensalism” is the most important way for humanoid robot to serve human beings.In the“Human-robot commensalism” residence,the most urgent problems to be solved are the understanding of the working environment and the path planning.Thanks to the improvement of machine vision technology and image processing techniques,autonomous mobile operation based on machine vision is more advantageous to the identification and judgment of indoor environment,so the visual sensor used in humanoid robot in the working environment to understand.In the path planning,the characteristics of humanoid robot need to be considered and combine them with algorithms,better results will be got in path planning when we consider path planning as a whole system.The purpose of this thesis is to improve the efficiency of path planning based on improved A-star algorithm of humanoid robot.This thesis describes the structure of the system,the modeling of operation environment and the improvement of path planning algorithm.Firstly,the way of the visual sensor to obtain the environment map had improved,and the global environment mathematical model was established based on the environment information.Secondly,the evaluation function and search strategy of the traditional A-star algorithm are improved.The improved algorithm is more efficient in dealing with the special environment map.Finally,the operation environment modeling and path planning was verified by experiments.In the experiment,the system successfully completed the task of identification,establishing the environment mathematical model and planning the path of the operating environment.The minitype humanoid robot successfully reached the target location along the best path.Experiments show that the environment recognition and the algorithm of path planning in this thesis are effective,and the task of research on path planning based on improved A-star algorithm of humanoid robot is completed.