AGV (Automatic Guided Vehicle) real-time scheduling method integrated with online learning

Abstract

The invention belongs to the technical field of aviation intelligent manufacturing, and discloses an AGV real-time scheduling method fusing online learning. Comprising the following steps: S1, acquiring position information, environment information and path planning information of an on-site AGV; s2, according to the information obtained in the S1, a genetic algorithm is adopted to match the most suitable trolley to complete a distribution task; s3, the most suitable trolley matched in the S2 is adopted to execute a transportation task; in the process of executing the transportation task, planning a shortest path for the trolley by adopting a reinforcement learning algorithm; and S4, the trolley completes the transportation task according to the shortest path planned in the step S3. By combining the genetic algorithm and the reinforcement learning system, the scheduling problem of the AGV is solved by using the genetic algorithm, the problems of AGV avoidance and path planning are solved by using the reinforcement learning system in the scheduling process, and tasks are reasonably allocated to the AGV. According to the method, the complexity of an algorithm is reduced, the working efficiency of the AGV is improved, and greater economic benefits can be brought.

Description

technical field [0001] The invention belongs to the technical field of aviation intelligent manufacturing, and in particular relates to an AGV real-time scheduling method integrated with online learning. Background technique [0002] At present, domestic intelligent logistics has become a key link in the realization of intelligent manufacturing. AGV undertakes the handling of parts in the logistics system, and sends the corresponding parts to the appropriate place in a timely manner according to the requirements of the system. Therefore, improving the working efficiency of AGV trolleys is of great significance to the development of the company. Compared with foreign countries, the domestic logistics demand is greater, and the research work of real-time regulation is very important. [0003] At present, the real-time scheduling of multiple AGVs in China is greatly affected by the algorithm, and the complex space environment affects the scheduling efficiency of multiple AGVs....

AI Technical Summary

Problems solved by technology

[0004] In view of the above problems, the purpose of the present invention is to provide a real-time scheduling method for AGVs that integrates online learning. By combining genetic algorithms and reinforcement learning systems, genetic algorithms are used to solve the scheduling problem of AGV trolleys, and reinforcement learning systems are used in the scheduling process. Solve AGV avoidance and path planning problems, and reasonably assign tasks to AGV

Images