INTERNATIONAL JOURNAL OF ROBUST AND NONLINEAR CONTROL Int. J. Robust Nonlinear Control 2010; 20:119–122 Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/rnc.1562
Editorial
Industrial control over wireless networks GUEST EDITORS: Maria Domenica Di Benedetto1, ∗, † , Karl Henrik Johansson2 , Mikael Johansson2 and Fortunato Santucci1 1 Center 2 ACCESS
of Excellence DEWS, University of L’Aquila, College of Engineering, L’Aquila, Italy Linnaeus Center, School of Electrical Engineering, KTH Royal Institute of Technology, Stockholm, Sweden
Traditionally, industrial control systems have relied on hardwired information flows among sensor, actuator and control nodes over rather simple network architectures. Over the last decades, distributed control systems have seen a transition to communication buses, such as fieldbus and Ethernet. At present, there is a major interest in taking a further step in this evolution by incorporating wireless connectivity with a concrete perspective of achieving increased efficiency, flexibility, mobility and easier installation. As a matter of fact, wireless technology is already present in several manufacturing and process industries. Wireless links have already been installed for long-range (100–1000 m) monitoring applications, wireless LANs have been used for applications with non-stringent communication requirements (e.g. logistics) and wireless communication has been even used for instrumentation setups (e.g. temperature and pressure measurements). The most challenging application of wireless technology is in distributed control because of requirements involving delay, reliability, availability and stability. This Special Issue is intended to present a snapshot of the recent achievements in control and communication for enabling industrial control over wireless networks. The issue covers critical aspects of control, communication and software design in the following areas: • • • • •
Distributed estimation and control under communication constraints; Communication and network design for distributed control; Control-communication co-design; Software architectures for distributed control; Applications of wireless communication for distributed control.
∗ Correspondence
to: Maria Domenica Di Benedetto, Center of Excellence DEWS, University of L’Aquila, College of Engineering, L’Aquila, Italy. † E-mail:
[email protected] Copyright q
2010 John Wiley & Sons, Ltd.
120
EDITORIAL
Each selected paper covers a different set of areas and may embrace some experimentally oriented content. Accordingly, we thought that it would be useful to the reader to present them as follows.
1. DISTRIBUTED MANAGEMENT We open the issue with a paper by Manfredi that addresses the critical problem of reliable management for monitoring and control in industrial environments. An analysis of the various aspects of this problem is offered, while reviewing the efforts in the field. The contributions of the paper lay in the following areas: (i) investigation of the interactions between contention resolution and congestion control mechanisms in wireless industrial sensor networks with the use of simulation to quantify the impact of network parameters on performance metrics (e.g. packet loss); (ii) a reliable sink resource allocation strategy based on log-utility fairness criteria (iii) the analysis, design and validation of a reliable distributed sink control with a sufficient condition for wireless network stability in the presence of multiple sinks and heterogeneous sensors and (iv) analysis and controller performance evaluation by a Matlab/Simulink-based simulator.
2. DISTRIBUTED ESTIMATION The paper by Rabi et al. investigates the problem of state estimation, where sensor measurements are sent over a medium shared using a contention-based access mechanism. State estimation is a key component of modern automation systems and contention-based medium access is supported in most standards. The estimator performance depends on the dynamics of the individual systems and the nominal sampling interval. More interestingly, it also depends on the distribution of transmission delays and loss rates of sensor packets which, in turn, depend on the applied medium access control scheme and the number of contending nodes. This paper presents analytical expressions for estimation error variance under varying sampling interval, and develops novel formulas for the statistics of packet latencies and losses that arise when nodes contend for medium access. Combining these results, the paper studies optimal sampling interval selection and establishes scaling limits of estimation performance with respect to the number of contending nodes. A basic aspect in the analysis and the design of systems of cooperative agents is related to the effect of the agent information exchange on the coordination performance. A coordination task widely treated in the literature is the so-called average consensus. This is the problem of driving states of a set of dynamic systems to a common state that corresponds to the average of the initial states. The paper by Carli et al. proposes a quantized strategy that reaches the initial average asymptotically. The strategy adapts coding/decoding strategies that were proposed for centralized control and communication problems to the distributed consensus problem. In particular, by resorting to both analytical techniques and simulations, the convergence properties of two coding/decoding strategies are assessed:, the first one is based on the exchange of logarithmically quantized information, the latter one on a zoomin–zoom-out strategy. The paper shows that the convergence factors depend smoothly on the accuracy parameter of the quantizers used and that the quantizer accuracy that guarantees convergence is independent of the network dimension. The paper by Bolognani et al. addresses the modeling and the algorithmic aspects of a popular application for wireless sensor networks (WSNs), namely, localization and target tracking. For this application, the wireless radio in Copyright q
2010 John Wiley & Sons, Ltd.
Int. J. Robust Nonlinear Control 2010; 20:119–122
121
EDITORIAL
each node of the WSN is used not only to communicate but also to measure the radio signal strength associated with the received packet. Since the signal strength is a function of location of the transmitter and the receiver, it can be used to estimate their relative positions. In particular, the paper focuses on the problem of estimating the channel parameters for a generic WSN in a completely distributed manner using consensus algorithms. Specifically, a distributed strategy is proposed to minimize the effects of unknown constant offsets in the reading of the received signal strength indicator due to uncalibrated sensors. Then the computation of the optimal wireless channels parameters, which are the solution of a global least-square optimization problem, can be obtained with a consensus-based algorithm. The proposed algorithms do neither require any knowledge on the global topology of the network nor the total number of nodes.
3. INDUSTRIAL CONTROL EXPERIMENTAL TESTBEDS The paper by Walsh et al. addresses one of the fundamental issues in the practical deployment of a WSN: the limited availability of energy at the sensor nodes. In particular, the paper proposes methods to dynamically adjust, from a finite list of available levels, the transmit power in a distributed manner so that power consumption is minimized while maintaining sufficient transmission quality. The received signal strength indicator is selected as the dynamic variable to manage this objective. A rigorous methodology is presented that quantifies the effect of output power limitations and quantization constraints on bit error rate performance. The methodology uses an approach based on a robust anti-windup scheme. A priori levels of system performance are achieved using a quantitative feedback theory approach on the initial, linear part of the design paradigm. This hybrid design is assessed experimentally using a fully compliant 802.15.4 testbed where mobility is introduced through the use of autonomous robots. The paper by Bemporad et al. presents the design and the experimental validation of model predictive control (MPC) of a hybrid dynamical laboratory process with wireless sensors. The laboratory process, which is motivated by heating processes in the plastic and the printing industries, presents interesting hybrid dynamics and can be used as a general testbed for the validation of control strategies for hybrid systems over wireless networks. By approximating the stationary heat spatial distribution as a piecewise affine function of the position of the wireless sensor along a moving belt, the resulting plant model is shown to have hybrid dynamics. The control architecture is based on the reference governor approach: the process is actuated by a local controller, while a hybrid MPC algorithm running on a remote base station sends optimal commands over a wireless link exploiting the sensor information received through the wireless network. A discrete-time hybrid model of the process is used for the hybrid MPC algorithm and for the state estimator. The physical modelling of the process and the hybrid MPC algorithm are presented in detail, together with the hardware and software architectures. The paper by Witrant et al. describes an industrial automation problem with high environmental impact: the design of a mining ventilation control system. Ventilation control is essential for the operation of a mine in terms of safety (CO and NOx regulation) and energy optimization. A model-based control approach is adopted where distributed sensing capabilities offered by a WSN are used. To develop the control strategy, a new model for underground ventilation is proposed. Two different model-based control approaches are examined: (i) a nonlinear receding horizon control strategy where energy minimization is achieved by continuous operation of the fans, and (ii) a control strategy based on a hybrid description of the plant and fan operations, which provides automatic verification of the wireless control. These control strategies are compared with simulations, in terms of regulation efficiency, energy consumption, and the need for computational capabilities. Copyright q
2010 John Wiley & Sons, Ltd.
Int. J. Robust Nonlinear Control 2010; 20:119–122
122
EDITORIAL
ACKNOWLEDGEMENTS
We are grateful to the authors for their interesting contributions and for their effort in completing the papers on time with the tight deadlines of this Special Issue. We also wish to thank the reviewers for their help and competence offered to make this issue a significant contribution to Industrial Control over Wireless Networks.
Copyright q
2010 John Wiley & Sons, Ltd.
Int. J. Robust Nonlinear Control 2010; 20:119–122
