Research And Application Of Path Planning Problem By Beetle Antennae Search Algorithm In 2D Environment

With the development of technology,mobile robots have been used more and more widely in daily life and industrial production,such as: sweeping robots,game AI automatic path finding,express automatic sorting and so on.Therefore,how to quickly plan a reasonable and short path becomes the pain point of the path planning problem.After researching and analyzing the current mainstream path planning algorithms,we select the beetle antennae search algorithm with better performance in convergence speed and optimization performance to study mobile robot path planning and obstacle avoidance.In view of the path planning problem in different environments,we propose two improved beetle antennae search algorithm,and carry out simulation under various environments.The main contents of this paper are as follows.Aiming at the path planning problem in two-dimensional static environment,considering the deficiency of the beetle antennae search algorithm in terms of escape of the local extremum.This paper proposes an improved algorithm,named fallback beetle antennae search algorithm.The improved algorithm records each iteration as a collision when the algorithm falls into a local extremum.According to the number of collisions,it is automatically determined whether to switch the search strategy,and the automatic switching of the two search strategies of back and forward avoids the local extremum.Simulation experiments show that the improvement can improve the optimization success rate of the algorithm and the obstacle avoidance ability for complex obstacles in path planning applications.Aiming at the path planning problem in two-dimensional dynamic environment,considering the high requirements of real-time performance of dynamic path planning,a hybrid algorithm based on the beetle antennae search algorithm and artificial potential field algorithm is proposed.In addition,due to the particularity of path planning in a dynamic environment,the path planning method based on the algorithm is divided into two phases in actual operation.In the first stage,the initial potential field is first established based on the artificial potential field algorithm,and then the path is planned based on the beetle antennae search algorithm with hierarchical tentacles and the designed multi-segment objective function to obtain the initial path.In the second stage,trajectory tracking is performed on the initial path.Based on the detection window and partial re-planning strategy,the real-time update of the path and the pre-judgment and avoidance of the dynamic obstacle are realized.The simulation results show that the algorithm can realize real-time path planning in two-dimensional dynamic environment,and significantly reduce the planning time and the path length of the final travel.