| With the rapid development of science and technology,indoor service robots are more and more used in the field of medical treatment.It has important research value and market prospect to design a nurse assistant robot that can navigate autonomously in the room.However,the autonomous navigation and positioning of nurse assistant robot is the focus and difficulty in the current research.Therefore,this paper makes theoretical research and experiment on the real-time localization and mapping(SLAM)based on lidar technology and the autonomous navigation technology of mobile robot,and builds a set of autonomous navigation technology based on laser using the robot operating system(ROS)Radar slam and autonomous navigation system.First of all,the paper introduces the development status of nurse assistant robot at home and abroad.Aiming at the navigation function of nurse assistant robot,it designs the autonomous navigation system of nurse assistant robot,analyzes the chassis and hardware configuration of nurse assistant robot,and studies the motion model of differential drive wheel mobile robot according to the driving mode of the chassis.Aiming at the construction of sensor and software system of nurse assistant robot,the principle of lidar ranging and robot operating system(ROS)are studied.Secondly,the principle of SLAM algorithm based on particle filter and the principle of SLAM algorithm based on graph optimization are studied,and the gmapping algorithm based on improved Rao Blackwellized particle filter and the cartographer algorithm based on graph optimization are analyzed.The simulation environment and nurse assistant robot model are built by using the ROS simulation software gazebo.The robot model is simulated in the simulation environment by running the gmapping algorithm and cartographer algorithm.By comparing the mapping effect of gmapping algorithm and cartographer algorithm,the optimization scheme of each algorithm is proposed,and the map established by the optimized cartographer algorithm is determined for the robot navigation simulation experiment of nurse assistant.On the basis of determining the navigation map,the global path planning method based on the a * algorithm and the local path planning method based on the dynamic window method(DWA)are studied,and the function of autonomous navigation and obstacle avoidance of nurse assistant robot in the simulation environment is verified by gazebo.Finally,the nurse assistant robot equipped with navigation system is used to carry out the experiment of drawing and path planning in the third floor of Yifu Building.Through the control robot using the optimized gmapping algorithm and cartographer algorithm respectively,and analyzing the effect of the two algorithms,the experiment verifies that the map built by theoptimized cartographer algorithm is more in line with the actual environment,so the map built by the algorithm is selected as the map for the autonomous navigation test of nurse assistant robot in the real environment.According to the map environment,the ability of nurse assistant robot to plan global path based on a * algorithm is tested by setting target points,and the ability of nurse assistant robot to plan local path based on DWA is tested by setting multiple obstacles on the planned path.Experiments show that the system can make nurse assistant robot achieve the ability of map building and autonomous navigation. |
PDF Full Text Request