Local Environmental Perception And Obstacle Avoidance System For Mobile Robot

The autonomy of robots has always been the goal pursued by people in the field of mobile robots.In recent years,while the working environment of more and more autonomous mobile robots continues to expand,people's requirements for their autonomous movement are also increasing.As the underlying support for autonomous behavior,mobile robots efficiently sense complex three-dimensional environments and generate trajectories that avoid obstacles,which is the prerequisite for ensuring their safe operation.This paper takes the problem of local environment perception and obstacle avoidance of mobile robots in complex environments as the starting point,and conducts relevant research on the efficient perception expression and obstacle avoidance planning of mobile robots in local complex environments.First,it analyzes the current research status at home and abroad,and expounds the research background and significance of the thesis.After that,the concept and basic principles of mobile robot are introduced,the mobile robot motion model,odometer model and distance sensor model are deduced,and their respective calibration procedures are introduced.Secondly,the advantages and disadvantages of environmental information perception and expression methods commonly used in mobile robots are analyzed.In order to efficiently and compactly represent the obstacle information in the local three-dimensional environment of the mobile robot,the updating idea of sliding windows is introduced into the probabilistic elevation mapping method and related principles and specific steps are introduced.Furthermore,combined with the local environment elevation map generated by the local elevation map,a corresponding mobile robot traversability detection method is proposed,and finally the calculated traversability map is converted into a local pass cost map directly used for obstacle avoidance planning.Thirdly,after the dynamically updated local traffic cost map is known,the local obstacle avoidance problem of the robot is redefined as how to generate a link between the local initial state and the target state on the basis of considering the global planning path and the local traffic cost map Feasible collision-free trajectory.This paper introduces agraph-based local trajectory planning method TEB(Time Elastic Band)algorithm.The global path planned by the global planning algorithm is used as the initial optimization value of the TEB local trajectory planning method,and a dynamically updated local cost map is used to generate a Local collision-free trajectory.The trajectory is re-planned with the dynamic update of the local traversability map,so that the mobile robot has flexible obstacle avoidance and navigation capabilities in a complex dynamic environment.Finally,the local environment awareness and obstacle avoidance system of the mobile robot proposed in this paper is constructed in the environment based on ROS(Robot Operation System),and the system proposed in this paper is verified in a complex environment in a simulated environment and a real experimental environment.Feasibility and effectiveness.Compared with traditional perception methods and independent obstacle avoidance strategies,the system is more effective in expressing information in locally complex environments.At the same time,local trajectory planning based on local dynamic environment maps also makes mobile robots more flexible in obstacle avoidance ability.