Research On Path Planning Of Intelligent Cleaning Robot In Known Environment

Intelligent cleaning robot is currently more popular service-oriented robot,which combines the robot,sensor and artificial intelligence and other technologies.Path planning is one of the key technologies of intelligent cleaning robot,which is an important index to evaluate the merits and demerits of intelligent cleaning robot.Global path planning requires clean robots to plan the best or better path that covers the global collision with the obstacle at the lowest cost(eg,the shortest path,the least time,the lowest energy consumption,etc.).In order to solve the problem of low coverage,high repetition rate and overall traversal efficiency is low of intelligent cleaning robot path planning,this paper studies the environmental model and partition method under the known environmental cleaning task,and presents the solution and optimization model for improving the efficiency of rectangular partition and the efficiency of the regional convergence path.The main work is as follows:Firstly,the working environment is modeled by the grid method.irregular obstacles to for the irregular obstacle in the grid,make the algorithm is easy to fall into the local optimum and the robot into a corner and other issues.The grid element is used as the expansion operator to carry out the edge of the irregular obstacle expansion,so that the edge of the irregular barrier to occupy the entire grid unit,the irregular obstacles rectangular,easy to the next step of the rectangular partition to improve the robot's global path planning efficiency.For the partition of the traditional rectangular partition,the partition structure is not simplified,the improved rectangular partitioning algorithm is used to decompose the partition,and the partition is free grid.,which is convenient for the traversal path planning of the robot in the partition.In order to improve the efficiency of rectangular partition traversal,a solution to the evolutionary algorithm and its optimization model are proposed.For the traditional genetic algorithm,the convergence rate is slow and the optimization result does not satisfy the problem of node adjacency.When the population is initialized,the neighboring nodes are adjacently judged by the adjacent table.The adjacent nodes in the search result are adjacent rectangular partition.The new solution can avoid the repeated traversal of the partition compared with the traditional depth-first and breadth-first search method.Compared with the ant colony algorithm which solves this kind of problem,the number of iterations is reduced and the convergence time is shortened.Search for the traversal order that satisfies the optimization criteria.Aiming at the regional convergence path planning of irregular obstacles between adjacent partitions,and combining with the irregular edge of the obstacle,the expansion area can also be covered and cleaned,so as to improve the overall sweep coverage.In order to reduce the repetition rate of the regional cohesive paths between rule obstructions,the traditional A* algorithm estimation function is improved.First,the rules of reciprocating traversing the starting point,direction and end point in the rectangular partition are given,and the reciprocal artificial cleaning mode is applied to the rectangular partition without obstacle.Based on the combination of Manhattan distance and diagonal distance,the heuristic function is improved on the basis of the traditional A* algorithm,and the corner cost is introduced in the estimation function,which effectively reduces the repetition rate of the regional convergence path.Finally,RobotBASIC is used to simulate the experiment.The experimental results show that the proposed method can make the cleaning robot find the full coverage,low repetition rate and efficiently traverse the clean path.