| With the concept of Made in China 2025 and industrial automation,advanced technologies such as digital factories,industrial IoT,and artificial intelligence have been rapidly developed and applied,and these technologies have gradually been applied in intelligent assembly workshops.In the intelligent assembly workshop,how to realize the automation and flexibility of material transportation has become a bottleneck to improve the efficiency of the assembly workshop.However,the application of automatic guided vehicles(AGV)has become an effective way to solve the problem.The application of AGV vehicles can improve the Transportation efficiency,maintaining the stability of system assembly and processing.This thesis mainly studies single AGV path planning and multi-AGV coordination strategies.Firstly,the single AGV path planning algorithm is improved,and then load balancing strategies are used to avoid conflicts between AGVs.In the process of AGV path planning,the ant colony algorithm is used to plan the path.Due to the low search efficiency of the algorithm,it is easy to fall into the local optimum.This thesis proposes the ant colony algorithm segmented search and adaptive pheromone volatility coefficient.Among them,the ant colony algorithm segmented search is to divide the entire search process into two stages.In the first stage,the original algorithm search step is improved,and the search is performed at a double step.In the second stage,the original ant colony algorithm is used to search.The size of the working environment area and the complexity of the environment,and the determination of the boundary between the first and second stage searches has become the main research content.In the process of using ant colony algorithm to search the path,an adaptive pheromone volatility coefficient is proposed.The coefficient is a function of the number of iterations.At the beginning of the algorithm,the value of the pheromone volatility coefficient is large,which is conducive to the global divergent search path.The number of iterations increases,the volatility coefficient gradually decreases,and the amount of pheromone on the path gradually increases,which enhances the guiding role of pheromone,speeds up algorithm convergence,and improves algorithm stability.When multiple AGVs are transporting tasks at the same time,in order to ensure the smooth execution of AGV tasks and reduce the number of AGV stops and starts due to conflicts,this thesis proposes a multi-AGV collision avoidance method based on load balancing,which records the road segment AGV traffic according to the road record points the number of times and the section’s approaching map information,and the penalty terms and penalty coefficients are used to update the global map section weight information in real time.The AGV path is planned from the perspective of overall system execution task efficiency,to achieve a uniform distribution of section weight values,to avoid relatively concentrated AGV driving routes,and reduce the number of conflicts,and improve the efficiency of the system to complete the task.This thesis mainly improves the single AGV path planning algorithm,and proposes a load balancing avoidance strategy for multi-AGV systems.In the intelligent assembly-oriented system,the AGV dispatching system rationally plans transportation routes according to the actual operating environment and improves the efficiency of transportation tasks,which has important application value. |
