Research On The Path Planning Of Mobile Robot Based On Intelligent Algorithm

In recent years, mobile robots have been widely used in various fields, so the robot path planning problem has attracted more and more researchers’ attention. With the emergence of intelligent robots, the calculation of path planning has been more and more complex. Moreover, the environment of robot can be dynamic and the problem can be multi-objective which generate more difficulties for the classical optimization algorithms.Intelligent computing is a special computing method which has aroused extensive attentions. It uses the heuristic search strategy to complete the random and parallel optimization of feasible solution, and it is widely applied on some optimization problems.Thus this thesis focuses on solving the path planning problem using the intelligent algorithms and the main contents are described as follows:1. Firstly, the dissertation provides a detailed literature review about the path planning. Then the definitions, general characteristics, common classifications of the robot path planning problem and some common algorithms to solved the problem are introduced in detail. The biological principles and the main feature of artificial bee colony are also described in this section.2. Then this paper takes use the Improved Artificial Bee Colony(IABC) algorithm to handle the path planning problem. For this problem, the proposed algorithm modifies the search methods of employed bees and onlooker bees, it combines with the search methods of particle swarm optimization and adopts the global best and the personal best.The IABC further enhances the algorithm’s search capabilities, and also overcome the disadvantages of the standard artificial bee colony algorithm. The experimental results demonstrates IABC is able to generate better solutions and can get shorter collision-freepath.3. Based on the above, IABC can only get one optimal path, for this problem, a newly developed niching DE algorithm called Neighborhood based Crowding Differential Evolution(NCDE) is used to handle path planning problem in this paper.The crowding niching technology can solve multimodal problems better, it can obtain a optimal solution and some suboptimal solutions in one run. The experiment shows that NCDE is effective in solving the problem, the algorithm is able to generate multiple optimal paths simultaneously, and the best path is shorter than other algorithms’.