Research On Application Of AGV Control System Based On Ant Colony Algorithm

Automated Guided Vehicle(AGV)is widely used in logistics and warehousing,transportation,assembly and manufacturing,and it is a research focus for the control of unmanned vehicle.The requirements for modern industry to control AGV include some basic functions like acceleration and deceleration,forward and backward,and it is also equipped with other functions such as trajectory tracking,path planning,task scheduling,environmental perception.In order to improve automation and reduce production cost,this article designs an AGV control system based on ant colony optimization algorithm.Firstly,research on trajectory tracking of AGV.According to the kinematic and dynamic characteristics of AGV,a hybrid algorithm on the basis of Backstepping and sliding mode control is proposed.The kinematic controller can effectively reduce errors of pose tracking of AGV.The dynamic controller solves the uncertainty of the system.The simulation results show that the algorithm has a fast dynamic response and causes small errors of steady state,and it can effectively eliminate input buffeting.Secondly,research on path planning of AGV.It introduces differential evolution algorithm in ant colony optimization algorithm,ant colony is classified by fitness function,and increasing pheromone concentration of dominant population by pheromone update strategy,a path planning method of AGV based on improved ant colony optimization algorithm is proposed.The algorithm enjoys fast convergence rate and high precision.,when U obstacle appears,the algorithm will not stop and it is suitable for path optimization of AGV in complicated environment.Thirdly,research on scheduling of multiple AGV systems.Based on the path planning of multiple AGV systems,the article proposes using ant colony optimization algorithm to solve the vehicle routing problems with time windows(VRPTW).As for cross-conflict phenomenon,with quadratic path planning algorithm of ant colony,it artificially increases a crossed grid as it seeks to make two AGVs cross the junctions at different time,thereby avoiding cross-conflicts.Compared with genetic algorithm,the method is effective and enjoys excellent performance in solving the VRPTW.Finally,focusing on STM32F103C8T6,an AGV control system is designed.The system adopts double control structure of upper and lower machines,which is convenient to the interaction of data and the driving of instructions,and strengthens AGV's ability to manage complex tasks and multiple sensors.The experimental results show that the AGV system can flexibly achieve steering control,trajectory tracking and obstacle avoidance,and as a result,achieving expected design effects.