Research On Path Planning Of Kiva Robots

With economic globalization and the rapid development of Internet technology,ecommerce gradually penetrates into people's daily lives which becomes more and more convenient.E-commerce not only promotes the overall development of China's logistics industry,but also puts forward higher requirements and more challenges.Data shows that China's total cost of logistics accounted for the proportion of total GDP is still high.As one of the core aspects of logistics,slow development of logistic distribution has a serious hindrance to the level of China's logistics industry.In face of the boom increase both in orders' number and species,existing layout and sorting mode of distribution center faces tough difficulties.Traditional "people arrive" sorting mode wastes a lot both in walking time and labor costs.In view of the shortcomings of the traditional logistics and distribution system,Kiva mobile robot logistics and distribution system came into being.the appearance and application of which greatly improve the Amazon distribution center's sorting accuracy and efficiency,thereby effectively reducing the cost of labor,greatly improving competitiveness of enterprises.After reviewing the current situation of domestic and foreign research,this paper studies the path planning problem of multi-robot complex dynamic environment,treating Kiva system and Kiva robots as objects.Through simulating working environment of the distribution center,using the improved PSO algorithm,avoiding robots' collision as the main constraint,the robots walking path is planned reasonably,so that the system operation efficiency is optimized.The main contents of this paper are as follows: according to the working principle and characteristics of Kiva robots,modeling the logistics distribution center environment through the grid method;comparing existing path planning algorithms,finally choosing PSO as the main algorithm to study the multi-robot path planning problem;aiming at the defects of PSO,introducing a judgment threshold to avoid premature convergence,proposing the nonlinear dynamic weight and the boundary constraint to optimize the influence of the core parameters and the particle boundary problem,finally making the overall performance of PSO to achieve optimal;avoiding the multi-robots' road conflicts with establishment of reasonable priority rules and fitness function;using MATLAB as a tool to simulate the path selection of single Kiva robots,dual Kiva robots,and multirobots in complex dynamic environment.Final simulation results show that the improved PSO algorithm effectively prevents various collisions while choosing the optimal path for multiple robots,finally making system run efficiently and orderly.