Study On Complete Path Planning Algorithm Of Community Intelligent Sweeper

As the inevitable development trend of community and industrial park cleaning in the future,the application of community intelligent sweeper is an indispensable factor for building an environmentally friendly and civilized city.Due to the different application scenarios,the path planning and control of the community intelligent sweeper is a different research field from other unmanned vehicles.This paper is based on the National Key Research and Development Program(No.2020YFB1313400),the National Natural Science Foundation of China(No.U1864204),and the Fundamental Research Funding Project of Central Universities(No.300102220204).This paper will study the complete path planning algorithm of the community intelligent sweeper,and study the control algorithm suitable for the community intelligent sweeper.The completion of the article is expected to play a positive role in the development of domestic cleaning equipment.Aiming at the problem that the number of nodes in the OPEN list of the A* algorithm affects the efficiency of finding the optimal path,the JPS algorithm is selected as the basic algorithm for global path planning in this article.However,the traditional JPS algorithm can only search for jump points according to the eight directions of the current node,and does not consider the influence of the number of turns in the path on the motion efficiency.This paper uses the angle evaluation function and the method of eliminating invalid jump points to evaluate the JPS The algorithm is improved.Aiming at the problems of low computational efficiency,poor practicability,and high repetition rate in the traditional full coverage path planning algorithm that combines the unit decomposition method and the template traversal method,this paper proposes a directed traversal algorithm.This algorithm mainly determines the priority of the traversal direction according to the starting point position and direction,which can improve the computational efficiency and ensure a small path repetition rate and a higher coverage rate.Then,the DWA local path planning algorithm is researched.Based on the classic and complex obstacle avoidance environment,simulation is performed for different weighted parameter combinations,and finally the DWA algorithm parameter combination suitable for community intelligent sweeping vehicles is selected.Aiming at the problem of turning points in the initial path generated by the path planning algorithm,the approximate B-spline curve is used to smooth the initial path.For the special differential driving of the sweeper,the relationship between linear velocity,angular velocity,the front wheel steering angle,and the speed of the two driving wheels is deduced to control the differential wheel.And verify the effectiveness of the path tracking algorithm based on LQR control system.In order to verify the effectiveness of the algorithm in the paper,the ROS platform is used to model and simulate the community intelligent sweeper and the working environment,and the Adaptive Monte Carlo Algorithm is used for positioning.The effectiveness and applicability of the DWA algorithm and the directed traversal algorithm are verified.The feasibility of differential drive control is verified too.