Research On AGV Path Planning And Obstacle Avoidance Based On Fusion Algorithm

Path planning is the core issue of intelligent AGVs,and relevant algorithms have become a research hotspot.Compared with traditional algorithms,intelligent algorithms have stronger spatial search ability,robustness and adaptability.Common ones include ant colony algorithm,genetic algorithm and particle swarm optimization algorithm,among which ant colony algorithm has advantages of simple structure,strong adaptability and robustness.However,there are also problems such as slow convergence speed and not smooth planning path,and dynamic window method can complete the local path planning and obstacle avoidance of AGVs in real time,but there are also problems of low adaptability to the environment.In this paper,aiming at the path planning and obstacle avoidance of AGVs,a fusion algorithm based on improved ant colony algorithm and adaptive dynamic window method is proposed to study it.Firstly,the ant colony algorithm is improved to solve the problem that the path planning is not smooth enough when the traditional ant colony algorithm is used for global path planning.A smooth function is introduced into the state transition probability,and a multi-objective evaluation function based on entropy weight is proposed to evaluate the planned path.By improving the pheromone updating rules,the convergence rate of the algorithm is improved.Node optimization rules and bezier curve are used to optimize the initial optimal path.Through simulation experiments in grid environments with different complexity,it can be seen that the improved ant colony algorithm has shorter path length and better path smoothness,which verifies the effectiveness of the improved ant colony algorithm.Secondly,for AGV local path planning and obstacle avoidance problems,an adaptive dynamic window method is proposed,introducing two values of the average distance from the end of the predicted trajectory of AGV to the nearest obstacle and the distance from the current position of AGV to the target point as the independent variables affecting the size of the weight coefficient of the evaluation function,so that the weight coefficient can change adaptively with the environment,and proposing an obstacle avoidance strategy for dynamic obstacles.The simulation results show that the path planned by the adaptive dynamic window method is smoother,and the AGV can avoid the dynamic obstacles in the environment well,which has better performance;while the adaptive dynamic window method cannot avoid the "C" type obstacles in the environment well,the fusion of improved ant colony algorithm and adaptive dynamic window method is proposed.The performance of the fusion algorithm is verified by simulation experiments..Finally,a priority-based conflict resolution strategy is proposed for the conflict problem in multi-AGV path planning.Each AGV in the environment uses the fusion algorithm for path planning,and uses the conflict resolution strategy to resolve the conflicts between AGVs.The simulation experimental results show that the AGVs in the environment can operate in a coordinated manner and judge and resolve the conflict type in time for the occurrence,which verifies the effectiveness of the conflict resolution strategy.