Model-Based Compensation for Burst Message Loss in Wireless Networked Control Systems: Experimental Results

Abstract

A recent trend in networked control systems (NCSs) is the use of wireless networks enabling interoperability between existing wired and wireless systems. One of the major challenges in these wireless NCSs (WNCSs) is to overcome the impact of the message loss that degrades the performance and stability of these systems. Moreover, this impact is greater when dealing with burst or successive message losses. This paper discusses and presents the experimental results of a compensation strategy to deal with this burst message loss problem in which a NCS mathematical model runs in parallel with the physical process, providing sensor virtual data in case of packet losses. Running in real-time inside the controller, the mathematical model is updated online with real control signals sent to the actuator, which provides better reliability for the estimated sensor feedback (virtual data) transmitted to the controller each time a message loss occurs. In order to verify the advantages of applying this model-based compensation strategy for burst message losses in WNCSs, the control performance of a motor control system using CAN and ZigBee networks is analyzed. Experimental results led to the conclusion that the developed compensation strategy provided robustness and could maintain the control performance of the WNCS against different message loss scenarios.

The Benefits of Soft Sensor and Multi-Rate Control for the Implementation of Wireless Networked Control Systems

“…The block diagram of the soft sensor strategy based on the process mathematical model with multi-rate control is shown in Figure 2 . Unlike the strategies described in [ 28 , 38 ] in which the model data was used to compensate packet losses and the model is directly updated, in this paper an online network update scheme is developed and the model data is used mainly for reducing energy consumption. In this network update scheme, the soft sensor running in parallel with the physical process is updated in real-time with the same control signal data transmitted on the network to the actuators (different to the case when the control signal update is done directly without considering the network transmission).…”

“…The mathematical model of the WNCS for DC motor velocity control used for implementation of the soft sensor is given by Equation (1) . This model was experimentally identified using the scheme proposed for WNCS in [ 38 ]: …”