Task-oriented Scheduling for Networked Control Systems: An Age of Information-Aware Implementation on Software-defined Radios
Networked control systems (NCSs) are feedback control loops that are closed over a communication network. Emerging applications, such as telerobotics, drones and autonomous driving are the most prominent examples of such systems. Regular and timely information sharing between the components of NCSs is essential to fulfill the desired control tasks, as stale information can lead to performance degradation or even physical damage. In this work, we consider multiple heterogeneous NCSs that transmit their system state over a shared physical wireless channel towards a gateway node. We conduct a comprehensive experimental study on selected MAC protocols using software-defined radios with state-of-the-art (SotA) solutions that have been designed to increase information freshness and control performance. As a significant improvement over the SotA, we propose a novel contention-free algorithm that is able to outperform the existing solutions by combining their strengths in one protocol. In addition, we propose a new metric called normalized mean squared error that maps the age of information to a dimensionless quantity that captures the expected value of a control system's next transmission. We demonstrate its adoption and effectiveness for wireless resource scheduling in a case study involving multiple inverted pendulums. From our experimental study and results, we observe that value-aware prioritization of the sub-systems contributes to minimizing the negative effects of information staleness on control performance. In particular, as the number of devices increases, the benefit of control-awareness to the quality of control stands out when compared to protocols that focus solely on maximizing information freshness.
Code (0)
등록된 구현이 없습니다.
Tasks
Autonomous DrivingSchedulingSimilar Papers 제목 키워드 기반
Deep reinforcement learning for scheduling in large-scale networked control systems
This work considers the problem of control and resource scheduling in networked systems. We present DIRA, a Deep reinforcement learning based Iterative Resource Allocation algorithm, which is scalable and control-aware. …
Deep Reinforcement Learningreinforcement-learningReinforcement LearningReinforcement Learning (RL)+1When Mean Age Is Not Enough: Distribution-Aware Scheduling for Networked LQR Control
Age of Information (AoI) has become a central metric for the design of wireless update systems, especially in applications where fresh measurements support tracking, estimation, and control. Despite its popularity, the u…
A probabilistic algorithm for scheduling networked control systems under data losses
This paper deals with the design of scheduling logics for networked control systems (NCSs) whose communication networks have limited capacity and are prone to data losses. Our contributions are twofold. First, we present…
SchedulingA probabilistic scheduling algorithm for networked control systems
This paper deals with the design of scheduling logics for Networked Control Systems (NCSs) whose communication networks have limited capacity. We assume that only a subset of the plants can communicate with their control…
SchedulingOptimal Transmission Power Scheduling for Networked Control System under DoS Attack
Designing networked control systems that are reliable and resilient against adversarial threats, is essential for ensuring the security of cyber-physical systems. This paper addresses the communication-control co-design …
Scheduling