Loop‐Shaping Feedback and Feedforward Control for Networked Systems With Saturation and Delay

Abstract

The paper proposes a new robust networked controller, for example suitable for packet data transfer control in wireless sensor networks with uncertain delays. The method adjusts a linear H2 loop‐shaping design towards consistency with a long delay and a saturation in the feedback loop, where the saturation results because of the one‐directional packet flow. Feedforward from a measurable disturbance is integrated in the design. By formally extending recently derived sensitivity function constraints to include feedforward, a subset of the scriptL2 stability region is pre‐computed using the stability inequality of the Popov criterion. A robustly stable controller is then selected in the stability region. The new controller design method is applied to internet data flow control supporting 5G wireless connectivity. The results indicate a possible enhancement of the robustness against feedback delay uncertainty with a factor of 3‐4. Such robustness should be beneficial for example for packet data relying with variable routing over ad hoc wireless sensor networks.

Guaranteed cost control of distributed parameter networked control systems with time‐varying delay

“…and then applying free weighted technologies (18), (19), (23) and inequalities (16), (17), (22) to equality (15), it can be obtained that…”

“…Different from lumped parameter NCS studied in, [24] for distributed parameter systems, the boundary conditions given in (9) and Poincare inequality theory as (16) play important roles in deriving the stable condition with guaranteed cost problem considered. Moreover, in equations (18), (19) and (23), the weighting matrix technology is involved. With the introduces of some extra matrices such as M i and N i (i = 1,⋯, 6), although it increases the computational complexity to some extent, it indeed helps to reduce the conservatism by reducing the number of zero elements in the whole matrix.…”

“…However, in practical NCS, there inevitably exist time delay, quantization, data packet dropouts and other uncertainties because of the introduction of the communication network, which could cause negative impact on control systems, even lead to instability. Thus, increasing research interests have recently been paid to the study of the stability and stabilization with uncertainties and time‐varying delays, such as …”

Output‐feedback‐based sliding mode control for networked control systems subject to packet loss and quantization

“…On another research front, networked control systems (NCSs) have received a great deal of attentions over the past two decades, in which system information is transmitted by communication networks with finite capacity. Compared with traditional control systems, the structure of the network itself has brought a series of detrimental phenomena, including packet dropouts [6,7], random delays [8–10], quantization errors [11], fading in channels [12,13], and so on, which inevitably results in performance degradation or even instability of the controlled systems. Hence, it is quite interesting and necessary to investigate the design of SMC when it is subject to incomplete information.…”

The free‐weight matrix approach to event‐driven output H∞ control of networked control systems

“…System that using network as a medium for real-time information exchange is called NCSs. Thanks to low cost, light weight, easy installation and maintenance, NCSs has gradually become a hot topic in the international control field [1][2][3][4][5][6][7][8]. In the past years of research, a large amount of information exchange and data sampling in NCSs have been conducted according to time-driven manner.…”