Decentralized event-triggered control over wireless sensor/actuator networks (1004.0477v2)

Abstract: In recent years we have witnessed a move of the major industrial automation providers into the wireless domain. While most of these companies already offer wireless products for measurement and monitoring purposes, the ultimate goal is to be able to close feedback loops over wireless networks interconnecting sensors, computation devices, and actuators. In this paper we present a decentralized event-triggered implementation, over sensor/actuator networks, of centralized nonlinear controllers. Event-triggered control has been recently proposed as an alternative to the more traditional periodic execution of control tasks. In a typical event-triggered implementation, the control signals are kept constant until the violation of a condition on the state of the plant triggers the re-computation of the control signals. The possibility of reducing the number of re-computations, and thus of transmissions, while guaranteeing desired levels of performance makes event-triggered control very appealing in the context of sensor/actuator networks. In these systems the communication network is a shared resource and event-triggered implementations of control laws offer a flexible way to reduce network utilization. Moreover reducing the number of times that a feedback control law is executed implies a reduction in transmissions and thus a reduction in energy expenditures of battery powered wireless sensor nodes.

• The paper proposes a novel decentralized event-triggered control scheme that reduces reliance on full state information using local sensor data.
• It introduces online adaptation heuristics to balance the frequency of control updates and conserve network resources.
• Results on a nonlinear quadruple-tank process demonstrate improved performance compared to centralized and naive decentralized approaches.

Decentralized Event-triggered Control over Wireless Sensor/Actuator Networks

This paper presents an investigation into decentralized event-triggered control within the field of wireless sensor/actuator networks (WSANs). The authors propose a decentralized approach aimed at implementing nonlinear controllers efficiently over WSANs, emphasizing the economic use of network resources.

Context and Motivation

The transition of industrial automation systems to wireless domains has increased interest in feedback loops over wireless networks. Traditional control systems often assume ideal communication conditions; however, WSANs introduce constraints such as limited bandwidth, delay, and potential packet losses. This paper builds on the concept of event-triggered control as an alternative to conventional periodic control approaches, where control tasks are initiated by specific state-dependent conditions rather than fixed intervals.

Methodology

The research introduces a decentralized event-triggering mechanism which leverages only locally available information at sensor nodes, thereby reducing the communication and computation load. This decentralized approach relies on local triggering conditions that exploit locally measured data to initiate control updates. Unlike centralized implementations, this method avoids the necessity for full-state information, making it feasible for deployment across WSANs.

The authors utilize a structure where each sensor node checks specific conditions and transmits updates if these are met. They propose heuristics to adapt the decentralization parameters online, attempting to balance the sensor nodes' decision-making processes.

Key Contributions

1. Decentralized Event-triggered Scheme: The paper proposes a new framework for event-triggered control specifically designed for WSANs, which focuses on localized decision-making to achieve stability without the need for full system state information.
2. Adaptation Mechanism for Decentralization: A mechanism to adjust the decision parameters online, optimizing the time between control updates and improving the overall resource efficiency of the network.
3. No Weak-Coupling Assumption: The approach does not rely on weak-coupling assumptions, making it applicable to a wide range of control systems beyond those addressed in prior distributed techniques.

Numerical Results

The authors demonstrate the application of their decentralized event-triggered control on a nonlinear quadruple-tank process. It is shown that the proposed method yields time between updates that, while conservative compared to centralized control, provide significant performance benefits over naive decentralized implementations.

Implications and Future Directions

The research highlights the potential of decentralized event-triggered control in reducing the frequency of communications and computations in WSANs, conserving energy and extending the operational life of battery-powered network components. This approach is particularly relevant for applications requiring real-time processing with minimal resource expenditure.

In future developments, the focus may extend towards refining adaptation heuristics further and addressing more complex dynamic controllers. Additionally, ensuring robust operation in the presence of communication delays and other network imperfections remains a critical area for exploration.

Conclusion

The paper offers a significant contribution to the control community by outlining a practical approach to decentralized event-triggered control within WSANs. It presents a promising direction for efficient sensor/actuator network operations, highlighting the importance of prioritizing resource efficiency alongside control performance in evolving wireless infrastructures.