| In recent years,with the continuous innovation of robot technology,the application research of robots has gradually developed from the traditional manufacturing industry to the intelligent industry,and various types of robots emerge one after another.Path planning is the basis for robot to realize autonomous navigation and intelligence.In the actual complex working environment,it is difficult for a single path planning algorithm to meet the requirements of real-time and completeness at the same time.Therefore,it is of practical value to study hybrid path planning.Aiming at the complex and changeable indoor environment where robots work,this paper takes indoor mobile robot AGV as the research object,combines the optimality of global planning and the real-time of local planning,and proposes a hybrid path planning algorithm to solve the path planning problem in complex environment.The main research contents are as follows:(1)The optimized ant colony algorithm is used to search for the global optimal path.In order to solve the problems of poor global path searching ability,slow convergence speed and easy falling into local optimum in the complex and irregular environment with a large number of obstacles,the traditional ant colony algorithm.The heuristic function,transition probability and pheromone update mechanism in basic ant colony algorithm are improved correspondingly,and it is verified by Matlab simulation that the improved algorithm optimizes the shortest path by more than 12% and the iteration times by more than 55% in the presence of different obstacles in the environment,so as to realize the comprehensive optimization of global planning.(2)Aiming at the sudden static obstacles and dynamic obstacles encountered by the robot during driving,the AGV itself carries sensors to detect the state of dynamic obstacles,and uses the optimized rolling window algorithm for local path planning.Firstly,the sub-target points in the current rolling window are selected by heuristic search strategy.For special obstacle environments such as T-shaped areas and concave areas,the sub-target points are searched by the principle of angular separation line regression.Secondly,analyze the moving characteristics of dynamic obstacles,divide the collision types by predictive control theory,and put forward corresponding collision avoidance strategies for each collision type to complete dynamic collision avoidance;(3)Aiming at the complex and changeable indoor environment,a hybrid path planning algorithm based on improved ant colony algorithm and rolling window method is proposed.Firstly,according to the environmental map information established by grid method,an improved ant colony algorithm is used to plan a collision-free optimal path.In the process of traveling,the robot detects real-time environmental information by carrying sensors,analyzes the state of dynamic obstacles,and uses the optimized rolling window method to avoid obstacles and complete local path planning.Matlab simulation verifies the effectiveness of the hybrid algorithm.Finally,an experimental platform based on ROS is built for physical verification,and the hybrid path planning algorithm can complete the search of the optimal path in the shortest time,and the reaction time when encountering obstacles is shorter.It is verified that the hybrid path planning algorithm in this paper is more universal. |
