Research On Simultaneous Localization And Mapping Technology Of Mobile Robot

Since the industry 4.0 was put forward,the field of artificial intelligence has been regarded as an important scientific and technological field of the frontier science.Therefore,as a typical representative,mobile robot technology has received more and more attention.The intelligent degree of mobile robot is mainly reflected in its ability to navigate autonomously in ites environment.Simutaneous localization and mapping(SLAM)is the core of autonomous navigation system and is considered to be an important prerequisite for mobile robot to realize intelligence and autonomy.In this paper,based on the research of mobile robot SLAM problem,the SLAM method based on Rao-Blackwellized particle filter(RBPF)is improved accordingly,which improves the accuracy and robustness of positioning and mapping.It was experimentally verified.The main research contents of this paper are as follows:1.Based on the research of RBPF-SLAM proposed distribution optimization,this paper proposes an RBPF-SLAM algorithm based on grey wolf optimizer.The estimation performance of the Rao-Blackwellized particle filter is improved by the local exploration and global development capabilities of the grey wolf optimizer.Low-weight particles move towards the “prey” and the estimates of the pose is further optimized in the process.The algorithm overcomes the problem of particle degradation to a certain extent and reduces the number of particles required for precise localization and mapping.The improved algorithm is compared with the gmapping on different data sets and the results prove that the proposed algorithm is effective in different environments.2.Aiming at the problem of particle diversity degradation caused by the survival of the fittest mechanism in the RBPF-SLAM resampling scheme,an improved strategy based on regularized filtering regional resampling is proposed.In this paper,the particles are divided into three categories: high-weight particles,medium-weight particles and low-weight particles.The improved algorithm adopts different processing strategies for different types of particles,which maintains the diversity of particles to some extent and alleviates the problem of particle dissipation in traditional algorithms.Finally,the simulation experiment is carried out under VS,and the results show the feasibility and effectiveness of the algorithm.3.This parts starts with the chassis software and SLAM ontology software,and briefly introduces the SLAM software system components involved in this topic.The software uses a three-wheeled omnidirectional mobile robot equipped with twodimensional laser radar as the experimental platform,and the ROS system under Linux as the experimental software operating system,which realizes the precise localization and mapping with the SLAM algorithm.The reliability of the algorithm and software was verified by experiments in the real environment.