Research On Robot Path Planning Problems Based On Improved Fruit Fly Optimization Algorithm

At present,with the rapid development of robotics,the applications of robots have become more and more extensive.The variability and complexity of the robots' mtotion environment determine that robot path planning is not only a research focus in the field of robots,but also a crucial technology to intelligent robots.This paper explores and researches the following aspects of robot path planning.Firstly,the paper briefly summarizes the development history of robots and path planning methods,especially introduces research status and development trend of robot path planning in details.Thus,the solution strategy of this paper is made explicitly.In the second place,an effective and feasible algorithm is proposed to solve path planning problems.On the basis of a novel fruit fly optimization algorithm(FOA),a parallel adaptive improved fruit fly optimization algorithm(IFOA)based on simulated annealing is proposed to overcome FOA's shortcomings,such as easily trapping in local optima and weak balance ability of global search and local search.The basic strategies are as follows:First,chaos theory is used to initialize the initial population so that they can be better distributed in the solution space,which provides a great basis for the subsequent optimization process of IFOA.Second,an adaptive strategy is introduced to make the value of search distance change adaptively in a certain range,which avoids negative effects on the overall performance from improper values in the standard fruit fly optimization algorithm.Third,the concept of density factor is employed to judge the density of fruit fly population,and then corresponding location updating strategy is adopted according to this value.Fourth,the simulated annealing mechanism is conducted to effectively reduce the probability of the algorithm falling into local optima.Fifth,the parallel evolution strategy is to divide the population into exploration subpopulation and exploitation subpopulation,so that two subpopulations adopt different parameters to balance global search and local search ability of the whole algorithm.Then,the proposed algorithm is applied to classical function optimization problems.The feasibility and effectiveness of the proposed algorithn are verified by comparing with standard fruit fly optimization algorithm and other algorithms in references.Finally,the proposed algorithm is applied to robot path planning problems.The paper adopts a combination strategy of rough planning based on graph theory and detailed planning based on intelligent optimization algorithm.First,MAKLINK Graph method is used to model environments,and then Dijkstra algorithm can obtain the best rough planning path from the starting position to the target position in the MAKLINK Graph.On this basis,the proposed algorithm in this paper is applied to the detailed planning.In this paper,simulation experiments are carried out in many cases,which include six working conditions of single robot global path planning,multi-robot global path planning and robot path planning in the environment with dynamic obstacles.The experiment results show that the accuracy and robustness of the proposed algorithm are greatly improved,and which can solve this kind of path planning problems well.For the research of robot path planning problems,this paper makes some beneficial exploration and attempt.The proposed algorithm can also be extended to other engineering optimization problems.