Research On Path Planning Method Of Wheeled Robot

With the development of science and technology,more and more intelligent robots are used in our daily life and industrial production.However the path planning is the vital guarantee for the robots to finish the advanced and complicated tasks.It solves the problem of how mobile robots walk in the environment and it has become a key topic in the study of robots.Therefore,the research to the path planning of robots is of great significance.This thesis takes the wheel robots as the object of study and does an intensive research on the methods of the path planning,mainly including the following three parts:Firstly,it analyzes the robots' path planning through the artificial potential field method.Concerning about the problem of destination unreachable and local minimum,it suggests to adopt the node-shift method on the basis of building a new potential field function and optimize the parameter by genetic algorithm.Thus the optimal path can be planned.That is tested and verified by MATLAB simulation experiment.Secondly,the thesis suggests to plan the robot's planning by intelligent ant colony algorithm.It shows the modeling of the robot's working space by grids.Analyzing the basic ant colony algorithm,the thesis designs the self-adaption but the results are not quite satisfactory.Hence it uses elitist strategy,fallback strategy,immune operator and heuristic probability formula to improve the ant colony algorithm.Based on the simulation comparison through MATLAB,it utilizes wheel robots to do the experiments,confirming the practicability of the proved algorithm.At last the modified ant colony algorithm is led into the dynamic path planning.On the base of using Dijkstra algorithm to search the whole path quickly,interference of the dynamic and static obstacles is added to the process.And then the robots successfully reach the target point by applying the modified ant colony algorithm to seek the partial obstacle avoidance.Therefore the path planning is realized and the superiority of the combination of global and local path planning can be tested and verified by simulation experiment.