Research On Re-entry Complete Coverage Path Planning In Public Places For Cleaning Robots

Service robots are gradually accessing to people's daily life,and playing increasingly more important roles in family service,restaurant service,medical nursing and so on.Cleaning robots are most widely used to keep people away from heavy and boring physical laboring.It can be used for the ground cleaning either in domestic place or pubic place.According to the environmental characteristics of public places,this paper deeply researches the path planning technology of cleaning robots to improve its coverage efficiency.Complete coverage path planing and re-entry path planning are discussed in this paper.The main contents focused in this dissertation are listed as follows:(1)Research on environmental modeling and path planning.The kinematic model of a two-wheel differential driving cleaning robot is established;Several common environmental modeling methods are studied to determine that the grid map is used to express the working space information of the cleaning robot;Analysis and comparison the advantages and disadvantages of biologically inspired neural network,cell decomposition method and template model method to lay a foundation for subsequent algorithm improvement.(2)Research on efficient complete coverage path planning algorithm.According to the situation that the existing template algorithm cannot meet the requirements of complete coverage cleaning of cleaning robots in large-scale public places,an efficient template algorithm is proposed.This algorithm redefines four efficient templates,introduces the backtracking to escape from the dead zone,and uses A * algorithm for backtracking path planning.In terms of dynamic obstacle avoidance,a rolling window is constructed using sensor information,and a DWA algorithm is integrated to avoid obstacles based on the efficient template algorithm.Finally,MATLAB simulation shows that the algorithm can effectively reduce repetitive path,reduce repetition rate,and improve the coverage efficiency of the cleaning robot.(3)Research on re-entry path planning algorithm.According to the omitting areas caused by autonomous obstacle avoidance in complete coverage path planning,the problem of re-entry path planning is proposed.Considering the dynamic characteristics of the omitting areas,this paper summarizes the problem into the category of dynamic traveling salesman,and proposes a re-entry path planning method based on the dynamic Inver-Over evolutionary algorithm to solve it.Finally,MATLAB simulation shows that the algorithm can plan the path traversing all the omitting areas in real time and has fast response.(4)Experimental research for public places.The design of the ROS software system and the construction of the robot platform are completed.Simulation test and actual test verify the feasibility and effectiveness of the re-entry complete coverage path planning algorithm proposed in this paper.