Research On Mobile Robot Positioning And Navigation System Optimization

In recent years,with the continuous expansion of the scale of artificial intelligence applications,robotics has also undergone significant development.Mobile robot positioning and navigation technology is applied in the fields of unmanned vehicles and drones.As an important research direction of robots,simultaneous localization and mapping(SLAM)technology is the key to solving navigation problems.Due to the shielding of signals,traditional positioning technologies such as GPS cannot accurately position the robot indoors.This paper proposes a mobile robot system based on two-dimensional lidar.The robot system modeling,mapping and path planning are studied.The research content includes:(1)The overall architecture of the system is designed and a mathematical model is constructed for the system,including the robot's coordinate system model,motion model,and observation model.Based on the established mathematical model,the problem of simultaneous positioning and mapping of mobile robots in indoor environments is studied.Two filtering-based SLAM algorithms are introduced: SLAM algorithm based on Extended Kalman Filter(EKF?SLAM)and Particle Filter SLAM algorithm(PF).The basic theory of these two algorithms is deduced,and the steps and implementation methods of the algorithms are discussed.The simulation experiments are performed on the algorithm in matlab.The experiments show that by improving the proposed distribution and using adaptive resampling,the RBPF-based SLAM algorithm has higher positioning accuracy.It is used as a basic method for solving the SLAM problem of mobile robots.(2)Discuss the path planning of the robot in the current environment.The advantages and disadvantages of the dijikstra algorithm and the A * algorithm are analyzed and the algorithm simulation is compared.The A* algorithm is used as the global path planning algorithm of the machine,and the Dynamic Window Approach(DWA)is used as the local path planning algorithm.Combining the two algorithms of global path planning and local path planning,and adding the expansion interval,the overall path planning scheme of the system is designed to ensure that the robot reaches the designated target site safely and smoothly.(3)Use the simulation tools in ROS to build a virtual environment and simulate a robot,and perform simulation experiments on the robot to verify the effectiveness of the algorithm in the previous chapter.According to the functions that the robot needs to implement,the relevant hardware configuration and software platform are built.Lidar,odometer,inertial measurement unit and other sensors are selected for the robot to sense the strange environment.Experiment with a built mobile robot in a real environment.The observation results in Rviz show that the improved algorithm has higher mapping accuracy,and using the path planning scheme proposed above,the robot can smoothly and safely complete the navigation task along the optimal path.