Design And Implementation Of Service Robot Autonomous Navigation System

In recent years,with the development of science and technology,intelligent robots have been more and more widely used in military,medical,manufacturing,service and other industries.In this context,this paper takes indoor service robots as the object and autonomous navigation of such robots.The system has been designed and researched.The main work is divided into motion control system design,environmental map construction,navigation process optimization,positioning and path planning algorithm research.The specific work is as follows:First,the robot motion control system is designed.Including,the underlying control system circuit design and functional device selection,experimental platform overall construction,robot kinematics analysis,motor control and system communication based on PID algorithm.Second,carry out environmental map construction.By analyzing the common 2D SLAM algorithms of Hector,Gmapping and Cartographer,combined with the actual application environment of the robot,the Gmapping algorithm is selected as the method of positioning and mapping in this paper.Based on the analysis of the working principle and process of the algorithm,The grid map expression is selected,and the algorithm's construction effect is verified by experiments.Again,the positioning research and noise reduction processing in the robot navigation process are performed.The commonly used Monte Carlo positioning(MCL)algorithm is analyzed.For the MCL algorithm,the particle loss problem exists.The adaptive Monte Carlo positioning(AMCL)algorithm is used as the positioning method of the robot.Aiming at the problem of inaccurate positioning caused by bumps and slips in the navigation process,a method of using data fusion algorithm to fuse the inertial navigation and odometer data to correct the pose of the robot in real time is designed.In addition,because the robot is subject to the problem of laser radar noise interference,the noise filtering algorithm is used to reduce the noise by the radius filtering algorithm in the PCL point cloud filtering algorithm.Finally,the global and local path planning algorithms in the autonomous navigation process are studied.Aiming at the problem of excessive number of vertices in the global path planning A* algorithm,the improved A* algorithm for the evaluation function reduces the number of vertices in the global path planning.For the local path planning DWA algorithm path is not the optimal problem,the modified DWA algorithm is used to optimize the local path of the robot and improve the navigation efficiency.The traditional and improved algorithms are simulated by MATLAB and the function and precision of the autonomous navigation system are tested in real environment.The experimental results show that the number of vertices in the global path planning of the autonomous navigation system designed in this paper is reduced by 4 The total length of the path taken is reduced by 0.4m,the total number of local plans is reduced by 11 times,the average deviation on the X-axis is about 0.07 m,the average deviation on the Y-axis is about 0.04 m,and the average angular deviation is about 6.3 degrees.The function completion effect meets the expected goal and can meet the normal working needs of the indoor service robot.The research content in this paper can also provide ideas for the design of other series of wheeled robot navigation systems.