Smart Autonomous Vehicles in High Dimensional Warehouses Using Deep Reinforcement Learning Approach

摘要

In this paper, we propose a smart planning andcontrol system for autonomous vehicles in a high dimensionalspace. It is a complete unsupervised scheduler and motion planner. Many warehouses take advantage of using an automatedmaterial handling process for product transshipment to speedup procedures. However, the growth of the space dimensionsbecomes a big issue for the control system that becomesincreasingly complex. The introduced model uses, as input, akernel with a control system based on Deep ReinforcementLearning for the low dimensional space. Besides, it employesa global transition-control system to intelligently coordinatecommunications between the kernels. The global transitioncontrol system creates virtual paths for each product, assignstasks to kernels for handling products in their zones, andensures transitions between blocks to brings each product totheir destination. Our approach yields good performance interms of speed and number of movements. The system is robustto the increase, as well, the size of the warehouse as the numberof products.