The Research And Design Of Obstacle Avoidance Complete Coverage Path Planning Algorithm For Robot Vacuums

Robot vacuums have been widely used in various households,and complete coverage path planning is an important indicator for evaluating the performance of robot vacuums.Complete coverage path planning requires robot vacuums to plan a route that covers all areas which can be cleaned,meanwhile reduces duplication of routes.Since the obstacle avoidance process of most robot vacuums is relatively simple,it leads to low coverage rate and high repetition rate.To deal with the problem,this paper proposes an obstacle avoidance complete coverage path planning algorithm,which combines obstacle avoidance strategies of different obstacles with modified complete coverage path planning algorithm based on grid activity to improve the coverage rate and reduce the repetition rate.Firstly,this paper studies the motion model of robot vacuum,map modeling and complete coverage path planning algorithm,and analyzes two typical complete coverage path planning algorithms: unit decomposition method and biologically inspired neural network method,pointing out the shortcomings of the two algorithms.Aiming at the problems of low coverage rate and high repetition rate caused by simple obstacle avoidance,this paper designs obstacle avoidance strategies based on different size of obstacles.Robot vacuums adopt surround obstacle avoidance strategy when it detects small obstacles,and border obstacle avoidance strategy for large obstacles.The obstacle grids in the grid map are marked according to the size of obstacles,and the robot adopts a right obstacle avoidance strategy when it moves to the vicinity of the obstacle grid.Then,aiming at the problem that there are some grids which have same value of activity by using the biologically inspired neural network method,an improved method of calculating the value of grid activity is given.At the same time,the energy consumption function is added to correct it,and use the A* algorithm to help robot escape from the dead zone.Finally,the improved algorithm is simulated and analyzed.Compared with the unit decomposition method and the biologically inspired neural network method,the coverage rate and the number of turns are reduced,and the efficiency of the complete coverage path planning is improved.In the end of this paper,the experimental platform of the robot vacuum is built.The hardware circuits of motor drive and obstacle avoidance are designed,and the software is designed based on the ROS and Keil development platform.In order to verify the feasibility of the algorithm,the experimental test of the robot vacuum is carried out.The experimental results show that the coverage rate of the robot is 90.37% and the repetition rate is 11.42%.The reasons for the missing area and the repeating area are analyzed,and the robot can complete the complete coverage cleaning task.