Research On Indoor Mobile Robot SLAM Based On Laser Radar

In recent years,with the emergence of artificial intelligence as a hot research direction in various fields,the intelligence of mobile robot has also developed rapidly.Mobile robots perceive the surrounding information and are able to move autonomously into the basic requirements of intelligence.Applied sensor aware information to achieve reliable positioning is the most basic and important capability of autonomous mobile robots.It is also a research topic that has attracted much attention and challenge in robot research.Simultaneous Localization and Mapping(SLAM)is the premise of robotic intelligent movement,and it is also a difficult and hot topic in the field of robot research.Therefore,SLAM is an important part of robot research.In this paper,Lidar is used as the main sensor to study the SLAM method and path planning problem in the case of unknown robot pose in indoor environment.Main tasks as follows:Firstly,build the hardware platform required by SLAM,establish the motion model and sensor observation model of mobile trolley,analyze the principle of common laser sensor in detail,and give the data acquisition method of RPLidar A2 laser sensor used in the paper;introduce the theory of SLAM.Two filtering methods are used: Extended Kalman Filter(EKF)and Particle Filter(PF)to solve the SLAM problem.By implementing the simulation in MATLAB and comparing the simulation results of the two methods,the extended Kalman filter is a better filtering method in the practical application of this paper.Then,the algorithm of filtering,region division,fitting and road sign extraction in point clouds processing is designed and implemented.When matching the point clouds in the front and back time,this paper proposes to select the landmark points in the point cloud to match,which reduces the algorithm complexity of point cloud matching.The Kalman filtering method is used to realize the SLAM process in the indoor environment.In the indoor environment,the robot platform is built,and the robot car moves at a fixed speed and performs the SLAM process.The system results are in line with expectations.Finally,the path planning algorithm is implemented in the result map obtained by applying the SLAM process.Give map representation methods in path planning: raster maps,straight maps,topological maps,and landmark-based maps.The A* algorithm path planning is applied to the grid map.The steps of path planning of the A* algorithm in the grid map are given,and the corresponding algorithm codes are written.The advantages and disadvantages are summarized.On the basis of the traditional A* algorithm,it carried out the improvement of eight neighborhoods,simulated with MATLAB,effectively realized path planning,and provided technical basis for robot path planning.