UAV-assisted Real-time Data Processing using Deep Q-Network for Industrial Internet of Things

Industrial internet of things (IIoT) enables edge computing technology to provide communication between the machines that produce a large amount of data and locate at the edge network. A task scheduling is implemented in the edge node. Furthermore, the real-time data can achieve with the lowest latency that allowed by the edge node near the edge network. However, a mobile machine such as an autonomous guided vehicle can interfere in this situation. Because the vehicle also needs service by the edge node. Over that, quality of service (QoS) performance can decrease. Therefore, this paper deploys an unmanned aerial vehicle (UAV) as an edge node to provide service to the edge network through optimizing the trajectory of UAV, where the edge network request task using a Deep Q-Network (DQN) Learning. The result shows that using machine learning, notably the DQN algorithm, can increase the number of the machine that can be provided service. Subsequently, the real-time data can achieve either the interrupt occurs at the edge node.