Research On Indoor AGV Path Planning Algorithm Based On Swarm Intelligence Optimization And RRT

Against the background of declining birth rate and increasingly serious aging population,AGV can solve the problem of labor shortage in the production process well.After years of development,AGV has become an important tool in the construction industry,industrial production,hotel service and other industries.Through the research of AGV path planning algorithm,it is helpful to reduce labor cost,improve efficiency,and ensure the safety of personnel in the production processPath planning algorithm is the key technology to improve the performance of AGV.As the production environment becomes increasingly complex,the research on THE PATH planning algorithm of AGV has been a hot issue.In this paper,RRT series algorithms and swarm intelligence optimization algorithms are studied.Based on previous studies,a variety of path planning algorithms are improved and simulated.The main work is as follows:Firstly,this paper improves the RRT* algorithm,and proposes an efficient PB-RRT*path planning algorithm based on map partition sampling and target bias expansion strategy.The map was divided into two regions according to the starting point and target point,and then the appropriate sampling probability parameters were set according to the proportion of obstacles to the map.In the process of sampling,sampling is carried out to the region where the target point is as much as possible with the set sampling probability parameter,so as to speed up the search speed.Secondly,although PB-RRT * algorithm improves the efficiency of AGV path planning,the planned path trajectory is not optimal.Therefore,the swarm intelligence algorithm is studied deeply in this paper,and an improved particle swarm algorithm is proposed.According to the density of obstacles in the environment,the population was divided into several small populations to increase the diversity of the population.Then,according to the accuracy required by the algorithm,the appropriate judgement interval and the minimum optimization value are set to guide the particles to get rid of the poor population.Then,in order to further improve the ability of PSO to find the global optimal solution,this paper introduces Cuckoo Search(CS)algorithm for local path planning.The ability of global particle swarm optimization is enhanced,and the CS-PSO algorithm is proposed to optimize local path planning.At the same time,the particle swarm optimization(PSO)algorithm is improved adaptively so that it can converge quickly when the optimal solution is not found,and improve the search ability when the convergence is near to find the asymptotic global optimal solution successfully.Finally,the PB-RRT * algorithm and CS-PSO algorithm proposed in this paper are summarized,and the development trend of AGV path planning is forecasted.