Mobile Robot Path Planning Based On Improved Particle Swarm Optimization

In recent years,as a cutting-edge direction and key topic of artificial intelligence technology,mobile robots have been widely applied in practical production and life with the development of artificial intelligence technology.As one of the core technologies of mobile robots,path planning technology dominates their work efficiency and affects their feasibility.Although traditional path planning algorithms meet the path planning needs of mobile robots to a certain extent,they still have problems such as poor adaptability and low planning quality.This article takes the particle swarm optimization with good global planning ability as the research object,and focuses on the path planning of mobile robots in static and dynamic scenarios,ultimately ensuring efficient and safe autonomous path planning for mobile robots in high complexity and dense working environments.The main research content of this thesis is summarized as follows:(1)Research on static path planning of particle swarm optimization based on multi module strategy.The main research contents of the multi module strategy include: 1)Aiming at the problem of low convergence speed of the particle swarm optimization,this thesis proposes a reverse learning strategy,which obtains the optimized initial particle swarm by judging the fitness of the initial particle swarm and its reverse particle swarm,so as to improve the convergence speed of the algorithm;2)In response to the contradiction between the global search breadth and local search accuracy of particle swarm optimization,this thesis proposes a parameter adaptive strategy based on search degree indicators,which iteratively optimizes the modifiable parameter groups and current conditions of each particle to achieve particle level search quality improvement;3)Aiming at the problem that particle swarm optimization is easy to fall into local optimization,this thesis proposes a hierarchical optimization strategy,which divides particles into high-quality particle swarm,ordinary particle swarm and lowquality particle swarm according to the fitness value,and designs different update strategies for each level of particles to improve the optimization ability of the algorithm by expanding the breadth of available strategies.(2)Research on dynamic path planning based on hybrid planning algorithm with separation of motion and static.The highly complex dynamic environment poses strict requirements and challenges for the real-time adaptability and planning timeliness of path planning algorithms.To address these issues,this thesis proposes a two-stage hybrid planning algorithm.1)In the static stage,a multi module strategy particle swarm optimization is used to provide the planned global optimal path for the static environment of the current frame and serve as the planning basis for the dynamic stage;2)In the dynamic phase,in view of the dynamic changes of the local environment,after the environmental information of each node is updated,the dynamic real-time obstacle avoidance strategy based on the joint dynamic obstacle avoidance strategy of improved artificial potential field is used to avoid obstacles.After successfully avoiding obstacles,continue to travel along the static optimal path.This thesis conducted optimization function testing,static and dynamic environment testing on the improved multi module strategy particle swarm optimization algorithm.Specifically,it includes: 1)Optimization function testing shows that compared to traditional particle swarm optimization algorithms,the multi module strategy particle swarm optimization algorithm has stronger global search ability;2)Static environment testing shows that the improved multi module strategy particle swarm optimization algorithm exhibits better path planning ability in various scenarios;3)Dynamic environment testing has shown that the hybrid planning algorithm combining motion and static can perform high-quality path planning in dynamic environments.