Research On Obstacle Avoidance Control Algorithm Of Mobile Robot In Complex Environment

With the advancement of science and technology and the rapid development of industry and manufacturing,robots are gradually known and play an increasingly important role.In recent years,mobile robot has become a research focus for its advantages such as simple structure,easy control and suitable for secondary development.Among the many research directions in the field of mobile robots,obstacle avoidance and safety research are very basic and important.In this thesis,the obstacle avoidance,motion,safety and practicability of mobile robots in a variety of obstacle environments are deeply studied.Our work is based on the Turtle Bot3 robot platform,and the algorithm is implemented by ROS operating system.This thesis mainly contributes to the following aspects.Firstly,the robot operating system and its application are introduced.The frame and related hardware of Turtle Bot3 mobile robot system are described.The mobile robot kinematics model is established on the basis of coordinate transformation and motion principle.At the same time,the communication mechanism of ROS,and the release and reception of ROS topics are also described.Secondly,considering the safety of mobile robot in practical application environment and the influence of unknown obstacles,the evaluation function of the Dynamic Window Algorithm(DWA)is improved and the mean evaluation method of motion trajectory is proposed.This method combines the advantages of the terminal value method and the cumulative sum method.It does not eliminate the high speed,but also maintains a suitable distance from the surrounding obstacles.In addition,in view of the influence of obstacles with different densities on obstacle avoidance of mobile robots,an Adaptive Dynamic Window Algorithm with variable speed weights is proposed.The validity of the designed scheme is verified through experiments.Thirdly,in the study of dynamic Obstacle avoidance algorithm for multi-mobile robots,the advantages and disadvantages of Reciprocal Velocity Obstacle(RVO)algorithm were analyzed.In view of the problem that reciprocal velocity obstacle avoidance velocity theory is optional and practical application is not accessible,considering the characteristics of reciprocal velocity obstacle algorithm and dynamic window algorithm,the two algorithms are fused to constrain the optional obstacle avoidance speed.Furthermore,in the process of multiple mobile robots encounter in complex environment,Adaptive Reciprocal Velocity Obstacle algorithm(ARVO)with variable expansion radius is proposed for the problem that conventional expansion radius may lead to the failure of obstacle avoidance.ARVO can adjust the expansion radius according to the complexity of the surrounding environment,which not only ensures the safe distance between the mobile robot and the surrounding obstacles,but also increases the optional speed range.In addition,the concept of fuzzy end point is proposed to solve the problem of "swinging" of mobile robot in specific environment.The validity of the algorithms is verified through experiments.Finally,ROS robot operating system and Turtlebot3 mobile robot are used to further verify the feasibility of the proposed algorithm in the real environment of the laboratory.