Key Algorithm And Implement Of Indoor Ground Robot Autonomous Navigation System

Since the 21st century,robot technology has gradually entered the stage of techno-logical frontiers,and human research on robots has gradually become intelligent.Robot navigation is a basic study of robot technology,and its significance lies in:autonomous movement in the environment where it is located is a prerequisite for many types of robots to complete other complex tasks.The robot navigation system includes mapping function,positioning function and planning function.The mapping function collects environmental data through sensors and processes the data using relevant algorithms to obtain a map model consistent with the real environment;The positioning function is based on the environment model that has been built,and the robot senses the sur-rounding environment and recognizes the characteristics of the environment during the movement process,and then determines its position in the environment;The planning function plans a feasible trajectory from the starting position to the target position based on the known environment model and positioning information,and issues control in-structions to guide the robot to move according to the planning resultsThis article uses ground mobile robots as the application background.It first ex-plains the problems encountered by robots in large-scale scene mapping,and introduces the background of SLAM algorithm and analysis of common SLAM problem solving methods.Proposed distribution to reduce the number of particles and mitigate the prob-lem of particle degradation through selective resampling strategiesSecondly,the robot autonomous positioning technology in complex scenes is stud-ied,the definition of mobile robot positioning and the representation based on proba-bilistic methods are given,and the positioning methods based on wheel odometer,IMU,lidar and other positioning methods are introduced.In order to solve the problem of in-accurate positioning in the process of using single sensor information,a combined posi-tioning algorithm was proposed.The UKF fusion wheel odometer and IMU data were used to obtain the robot posture and posture information,and then the posture was used as the initial MCL Value fusion laser data is used for secondary positioning,and finally the positioning results with higher accuracy and better robustness are obtained,and the effectiveness of the improved algorithm is verified through simulation and actual robot test.Third,the path planning technology of mobile robots is studied.The problem of low efficiency in calculating global paths by Dijkstra algorithm is discussed.A*search algorithm based on improved evaluation method is proposed as a global path planning algorithm.At the same time,we study the DWA dynamic window algorithm of local path planning,optimize and adjust the parameters of the evaluation function of the DWA algorithm,and conduct simulation experimentsFinally,according to the research methods of mapping,positioning and planning in this paper,experiments are carried out on the mobile robot platform,the methods proposed in this paper are implemented based on the ROS robot operating system,and corresponding experiments are carried out to verify the construction of maps in larger scale scenes.The positioning,planning and obstacle avoidance functions,the improved methods proposed in this paper,have achieved the expected results and have certain practical application value.