Deep Reinforcement Learning-Based Controller for Autonomous Guided Vehicles (AGVs) in a Multi- Department Production Plant

With the advent of advanced industrial facilities, there arises a pressing demand for swifter and smarter production methods. This is precisely why material handling plays a pivotal role in shaping modern manufacturing practices. The swift and efficient material handling is achieved by implementing Automated Guide Vehicles (AGVs). In our study, we delve into the examination of routing and scheduling predicaments encountered by AGVs, employing reinforcement learning (RL) techniques with PPO algorithm, using AnyLogic and ALPyne. The findings we have obtained exhibit intriguing prospects for resolving the real-case study we have presented, while also demonstrating reliability in addressing various alterations and scaling operations associated with the problem.