Abstract
Wireless sensor-actuator networks (WSANs) are gaining momentum in industrial process automation as a communication infrastructure for lowering deployment and maintenance costs. In traditional wireless control systems the plant controller and the network manager operate in isolation, which ignore the significant influence of network reliability on plant control performance. To enhance the dependability of industrial wireless control, we propose a holistic cyber-physical management framework that employs run-time coordination between the plant control and network management. Our design includes a holistic controller that generates actuation signals to physical plants and reconfigures the WSAN to maintain desired control performance while saving wireless resources. As a concrete example of holistic control, we design a holistic manager that dynamically reconfigures the number of transmissions in the WSAN based on online observations of physical and cyber variables. We have implemented the holistic management framework in the Wireless Cyber-Physical Simulator (WCPS). A systematic case study has been presented based on two 5-state plants sharing a 16-node WSAN. Simulation results show that the holistic management design has significantly enhanced the resilience of the system against both wireless interferences and physical disturbances, while effectively reducing the number of wireless transmissions.
Introduction: The Silo Problem in Industrial Automation
Modern industrial automation increasingly relies on wireless sensor-actuator networks (WSANs) to reduce costs. However, a fundamental weakness persists: the "siloed" nature of wireless control. Traditionally, the plant controller (managing physical dynamics) and the network manager (managing data packets) operate in isolation. This lack of coordination leads to fragility during physical disturbances and resource waste during stable operations.
The Holistic Management Framework
The proposed framework bridges the gap between control theory and network management. By allowing the plant controller to dynamically reconfigure network parameters in real-time based on the plant's physical state, the system ensures high reliability when critical and saves energy when stable.
Core Components and Methods
The architecture centers on a Holistic Controller that integrates several key principles:
- Model Predictive Control (MPC): Instead of single commands, the controller calculates a sequence of optimal future inputs. These are buffered at the actuator; if a packet is lost, the actuator executes the next best move from the previous successful transmission.
- Lyapunov-based Adaptation: A Lyapunov function serves as a real-time "health meter" for the physical plant. When the plant state deviates significantly from the target, the controller triggers an increase in wireless retransmissions (#TX) to guarantee delivery.
- Asymmetric Scheduling: The system prioritizes the actuation path (controller to actuator) for adaptive retransmissions, as sensing data is less sensitive to loss due to the use of Kalman Filter-based state observers.
- Piggybacking Protocol: Configuration changes are embedded in standard actuation packet headers, avoiding the overhead of dedicated management traffic.
Evaluation and Results
The framework was evaluated using the Wireless Cyber-Physical Simulator (WCPS), seeded with real-world wireless traces (RSSI and noise) from a 16-node testbed.
Key Insights
- Stability over Reliability: The study demonstrates that 100% Packet Delivery Ratio (PDR) is not always necessary for dependability. The system remains stable even with high data loss, provided the network can ramp up reliability the moment the physical state degrades.
- Resource Recapture: By reducing retransmissions when the plant is stable, the system recaptures bandwidth and extends battery life.
Performance Metrics
- Resilience: Under heavy interference, the Holistic Controller (HC) stabilized 100% of executions, whereas fixed-retransmission strategies failed in 25% of cases.
- Efficiency: HC achieved control performance (Mean Absolute Error) comparable to a fixed 4-retransmission strategy but with significantly lower resource consumption.
- Longevity: The adaptive strategy extended projected system lifetime by over one month compared to traditional network-only adaptation.
Limitations and Future Outlook
While promising, the current design assumes "revivable" conditions where at least some packets get through. In cases of total jamming, more aggressive recovery methods like flooding may be required. Additionally, scaling the framework to handle hundreds of control loops in complex, non-linear industrial environments remains a key area for future research.
Summary updated