Research On Path Planning Of The Mobile Robot In The Generalized Obstacle Environment

Path planning is one of the main research contents of robot motion planning.The task is to make the robot seek an optimal path from the start point to the end point in the working environment.In the existing research on path planning,the environment is usually divided into passable regions and non-passable regions.Corresponding to the common grid modeling method,these areas are free grids and obstacle grids.However,in the real environment in which mobile robots perform their tasks,there are not only "freedom" and "obstacle" areas with a single attribute,many regions are between them hazily.As free grids,because of the different environmental characteristics such as terrain,the cost of passage may be different.As obstacle grids,when robots carry out tasks,some obstacles can and should be broken through,while others must be avoided.In this paper,based on the fuzzy characteristics of mobile robot operating environment,the concept of generalized obstacle is proposed.Further,the modeling method and planning algorithm for path planning of mobile robot under generalized obstacle environment are studied.The main research contents and conclusions are as follows:(1)In view of the fuzzy characteristics of mobile robot operating environment,the definition and modeling method of generalized obstacles are given.Based on the characteristic parameters of special terrain,the breaking cost is obtained by fuzzy membership function,and the generalized obstacle model is established by grid modeling method.(2)For special terrain and obstacle-breaking task in static environment,an off-line planning algorithm for improved ant colony algorithm with the potential field under generalized obstacles is put forward.The artificial potential field is used to improve the related factors of ant colony algorithm.And combined with the iterative characteristics of the swarm intelligent optimization,a strategy of population crossover optimization is presented.(3)For unknown dynamic environment,an on-line planning algorithm based on biologically inspired neural network under generalized obstacles is proposed.Mobile robot takes virtual target as the basis of optimization and uses rolling optimization method;the threat assessment model is established by intuitionisticfuzzy set and multi-attribute evaluation method.And a transmission mode about guided scanning is provided to transfer the neuron excitation better.(4)The method of generalized obstacle modeling and path planning algorithm are tested and verified.The visual interface for path planning simulation is built on Matlab2014 b platform,then the related experiments and analysis is carried out.The testing results show that compared with the result of the traditional obstacle model,the off-line path planning algorithm with generalized obstacles proposed in this paper can get shorter and smoother paths,and can also obtain more reasonable and better planning paths;the on-line path planning algorithm under generalized obstacles can effectively avoid dynamic obstacles with different speeds and has better planning ability under the environment without motivation.