Mobile Robot Hybrid Path Planning Based On Improved Ant Colony Optimization And Pigeon-inspired Optimization Algorithm

Path planning is an important part of autonomous navigation.For path planning,it can be divided into the global path planning and the local path planning based on whether the surrounding environment information is known.In a real environment,a global path is obtained through the global path planning.When a mobile robot encounters sudden obstacles,it needs to use the local path planning to avoid obstacles.This dissertation proposes a hybrid path planning that combines global and local path planning.The main research content is as follows:(1)Aiming at the problems of slow convergence speed,difficulty in finding the global optimal solution,and interference from non-optimal path pheromone for the Ant Colony Optimization(ACO)algorithm in the global planning,this dissertation proposes an improved ACO algorithm to achieve the global path planning.First,an improved synchronous bidirectional A* algorithm is proposed,the direction evaluation function is introduced to accelerate the convergence of the algorithm.The initial pheromone distribution of the ACO algorithm is optimized by the improved synchronous bidirectional A* algorithm.Then,the transition probability of the ACO algorithm is introduced into a random mechanism to enhance the global search ability of the algorithm.A reward and punishment factor is introduced into the pheromone update mechanism of the ACO algorithm to reduce the number of search times near the non-optimal path and to accelerate algorithms converge.The performance analysis experiment of the improved ACO algorithm is carried out through the test function,and the simulation experiment verifies that the algorithm can obtain better results in the global path planning,and the proposed improved strategy is proved to be effective.(2)Aiming at the problems of low execution efficiency,insufficient security,and too few populations in the later stage for the Pigeon-Inspired Optimization(PIO)algorithm in the local path planning,this dissertation designs a new evaluation function judgment strategy and an evaluation function indicator.Furthermore,this dissertation proposes an improved PIO algorithm to achieve the local path planning.First,the Logistic mapping is utilized to initialize the PIO algorithm to improve the efficiency of the algorithm.In the map operator,the Beetle Antennae Search algorithm is introduced to make the PIO algorithm consider the influence of the group and the individual.The Metropolis guideline is introduced into the global optimal location to make the algorithm jump out of the local optimal solution.In the landmark operator,the logsig function is introduced as an adaptive factor of the population size to solve the problem of too few populations in the later stage.The performance analysis experiment of the improved PIO algorithm is carried out through the test function,and the simulation experiment verifies that the algorithm can better achieve local obstacle avoidance,and the proposed improved strategy is proved to be effective.(3)Aiming at the situation of sudden obstacles in complex environments,this dissertation proposes a hybrid path planning combining the global and local path planning.The improved ACO algorithm is used for global path planning,the improved PIO algorithm is used for local path planning,and the cubic B-spline curve is introduced to smooth and re-plan the polyline segments on the path to meet the motion characteristics of mobile robots,and to get a smooth and feasible path.The experimental results show that the mobile robot can safely avoid obstacles when encountering sudden obstacles.The hybrid path planning proposed in this dissertation is practical and can be used for path planning of mobile robots in complex environments.