Learning-Enabled Network-Control Co-Design for Energy-Efficient Industrial Internet of Things

In the Industrial Internet of Things (IIoT), energy efficiency is critical for effective management of physical systems. To achieve stable control of IIoT with minimal energy consumption, it is essential to co-design the controller and the wireless network. In this paper, we present a novel reinforcement learning (RL) approach called the Learning-enabled Self-triggered Wireless Networked-Control System (LS-WNCS). LS-WNCS learns complex interdependence between control and network systems, generating near-optimal control commands and sampling periods simultaneously to minimize energy consumption and maximize control performance. Compared with conventional RL algorithms, LS-WNCS reduces network energy consumption by up to 66% while maintaining a high level of control performance.