There exist bounded transmission delay and data packet dropout in the networked control systems (NCSs). When the sensors and actuators are time-driven and controllers are event-driven, the NCSs can be modelled as a class of discrete-time systems with time-varying input delay. Most of similar articles simply combine delay and packet dropout to analyse and synthesise NCSs without distinguishing their different impacts, which leads to conservative results. In this study, the authors summarise that the number of consecutive data packet dropout increases gradually in case of packet dropout. A novel Lyapunov–Krasovskii functional is constructed based on this increment property, so less conservative results are obtained through the Lyapunov–Krasovskii functional approach. In addition, the upper bound of a Lyapunov functional difference cross term is reasonably estimated to further reduce the conservativeness. Stability and stabilisation criteria which are separately related to the transmission delay and data packet dropout are presented. The obtained conditions are based on linear matrix inequalities, which can be solved easily by MATLAB or other numerical software.

In this study, the authors consider a coordinated target assignment problem where a group of unmanned flight vehicles cooperate with each other to combat with a set of target vehicles which may perform various manoeuvres. Each flight vehicle is expected to be optimally assigned to a manoeuvring target vehicle such that the maximal possibility of successful interception can be guaranteed. Considering the course of the engagement, the target assignment should be achieved in terms of the performance over a limited time interval. The authors establish a linear time-varying system in a finite-time horizon to study the target assignment problem. A cost function is proposed based on the finite-time generalised H ₂ gain measurement and the finite-time final value performance measurement which can be viewed as the metric of the maximal required acceleration and the zero miss distance. For computation, the properties and performance criterion of the finite-time generalised H ₂ gain are proved. The computation of the finite-time final value performance measurement is also treated. The targets are assigned by minimising the value of the overall assignment evaluation function. Finally, the authors give some simulation results to illustrate the effectiveness of the proposed target assignment method.

This article studies the problem of stability analysis for neural networks (NNs) with two additive time-varying delay components. By taking both the independence and the variation of the two delay components into consideration, a more general Lyapunov functional is defined. By estimating the upper bound of the derivative of the Lyapunov functional more tightly, a less conservative delay-dependent stability criterion is established in terms of linear matrix inequalities. To reduce the computational complexity, a method for eliminating slack variables is provided, and then a simplified stability criterion is obtained. Some numerical examples are given to illustrate the effectiveness of the proposed method and the significant improvement over the existing results.

This work addresses the tracking and rejection problem of periodic signals for discrete-time linear systems subject to control saturation. To ensure the periodic tracking/rejection an internal model-based controller is considered. Based on this control structure, conditions allowing the simultaneous synthesis of a stabilising state feedback gain and an anti-windup gain are proposed. Provided that references and disturbances belong to the interior of a certain admissible set, these gains guarantee that the trajectories of the closed-loop system starting in an ellipsoidal set contract to the linearity region of the closed-loop system, where the presence of the internal model ensures the periodic tracking/rejection. To compute the gains, an LMI-based optimisation problem aiming at the maximisation of the set of admissible states and/or the set of admissible references/disturbances is proposed.

In this study, the authors are concerned with the observer-based quantised H _∞ control problem for a class of discrete-time stochastic systems with random communication delays. The system under consideration involves signals quantisation, state-dependent disturbance as well as random communication delays. The measured output and the control input quantisation are considered simultaneously by using the sector bound approach, while the random communication delays from the sensor to the controller and from the controller to the plant are modelled by a linear function of the stochastic variable satisfying Bernoulli random binary distribution. It is aimed at designing an observer-based controller such that the dynamics of the closed-loop system is guaranteed to be exponentially stable in the mean square, and a prescribed H _∞ disturbance attenuation level is also achieved. Finally, a simulation example is given to illustrate the effectiveness of the proposed method.

The problem considered here is output regulation of a large-scale hydraulic network where the structure of the network is subject to change and where the actuation is subject to a non-negativity constraint. Previous results have shown that semi-global practical output regulation is achievable using a set of decentralised, logarithmic quantised and constrained control actions with properly designed quantisation parameters. That is, an attractor set with a compact basin of attraction exists. Subsequently, the basin can be increased by increasing the control gains. In our work, this result is extended by showing that an attractor set with a global basin of attraction exists for arbitrary values of positive control gains, given that the upper level of the quantiser is properly designed. Furthermore, the proof is given for general monotone quantisation maps. Since the basin of attraction is global and the result is independent on the size of the network, this shows that structural changes can be implemented without destabilising the closed-loop system.

The problem of automatic administration of vasoactive drugs to a patient can be treated as a regulation problem of a system, which is characterised by large parametric uncertainty, non-Gaussian disturbances and unmodelled dynamics, yet carries strict requirements in terms of robustness and performance. A number of approaches have been proposed in the past to tackle this problem, particularly for the postoperative management of blood pressure in cardiac surgery patients. We describe the design of a robust multiple-model adaptive control (RMMAC) architecture and investigate whether this can overcome some issues observed with earlier methods. Key features of RMMAC are robust optimal controller design using an iterative μ synthesis algorithm and improved system estimation. Simulation results indicate that RMMAC is capable of avoiding transient instability and delivering performance in the face of significant parameter changes over time and large disturbances including non-zero-mean signals. The findings support further research into RMMAC as a potentially viable approach to the design of safer automatic closed-loop drug administration technologies capable of operating under challenging clinical conditions such as may arise in an intraoperative setting.

Many robotic applications involve the tasks in which multiple coordinated mobile manipulators system interacts with rigid or non-rigid environments. In this study, we study the topic of multiple mobile manipulators grasping a rigid common object in contact with a deformable working environment, for which few investigations have been made. To solve the difficulty. We first develop the decentralised dynamics of the inter-connected system, including the dynamics of the mobile manipulators, the internal forces between end-effectors and the object and the interaction forces between the object and the environments. Then, we design decentralised adaptive fuzzy control based on impedance approach for coordinated multiple mobile manipulators using Lyapunov synthesis. Rigourous stability and performance analysis have been established, and simulation are presented to figuratively demonstrate that the motion/force-tracking errors converge to zero, whereas the internal force-tracking error remains bounded and can be made arbitrarily small.

The aim of this study is to provide a full comprehensive study of the singular linear parameter-varying (LPV) systems and observer synthesis in a real process. The states and unknown inputs estimation problem for a binary distillation column is solved by using an LPV proportional-integral observer (PI-Observer). A singular LPV model of the process is developed. This model facilitates the design of the PI-Observer for singular LPV systems. The observer proposed is tested using experimental data of a distillation column separating an ethanol–water mixture. The results indicate that the observer is able to accurately reconstruct the product composition dynamics and the unknown inputs of the process.

Sliding mode control is a simple and powerful technique for the robust control of uncertain systems. However, despite theoretical promise, real systems using sliding mode controllers have a tendency to chatter at higher frequencies. Second order or two-sliding mode controllers, can be used to suppress chatter while retaining the desirable properties of a classical sliding mode controller. These more complex controllers increase the number of parameters in the system. The affect of these new parameters on system performance is unclear, and this can make it difficult to tune the controllers to meet a desired performance criteria. This study addresses the problem of estimating the performance of a system controlled by two-sliding mode controllers. Describing functions are proposed to estimate chattering amplitude and settling time, and a new method is presented, based on invariant ellipsoid sets, to predict the maximum output bound of the system. The aim is assist in producing a systematic approach for practitioners wishing to directly apply sliding mode controllers with chatter suppressing properties to real applications.

In many applications, add-on type repetitive controllers have been reported to have prominent capability of attenuating periodic disturbances and/or tracking periodic reference commands. However, the effective information such as performance weighting functions for the design of feedback controllers has not been considered sufficiently on the design of repetitive controllers. In this study, we deal with a problem of a robust repetitive controller design for an uncertain feedback control system using its explicit performance information. We first show that a robust stability condition of repetitive control systems has a similar form with the well-known robust performance condition of general feedback control systems. The repetitive controller is designed using the performance weighting function for the design of the robust feedback controller. It is also shown that a steady-state tracking error of the repetitive control system is described in a simple form without time-delay term. This result yields that the repetitive control system has a much larger loop gain in the steady state than the feedback control system. Moreover, this paper provides sufficient conditions ensuring that the power of the steady-state tracking error in the repetitive control system is less than or equal to that of the feedback control system. Based on the obtained results, we present repetitive controller design method using the design information of the feedback control system. Finally, application studies on the track-following control system of optical disk drives are performed to show the validity of the proposed method.

The delay-dependent stability problem for systems with time-delay varying in an interval is addressed in this study. Based on Lyapunov–Krasovskii theory the proposed methods formulated as linear matrix inequality problems are able to check the stability interval when the time-varying delay d(t) belongs to an interval [τ₁, τ₂]. The Lyapunov–Krasovskii functional (LKF) selected in the present paper is simpler than some ones considered in the literature. However, the criteria obtained, based on this simple LKF, outperformed the similar existing ones in all numerical tests accomplished in this paper.

Notation: Through out this paper the superscript T stands for transpose. 0 refers to a null matrix with appropriate dimensions. For a real symmetric matrix M, the notation M > 0 ( < 0) means that M is the positive (negative) definite. The symmetric term in a matrix is denoted by *.

Monotone dynamical systems are those with solutions preserving specific orderings, relative to the associated initial states. This study shows that, under negative output feedback control, a monotone single-input single-output system with well-defined static characteristics is able to globally asymptotically regulate its solution at desired constant set-points, while the boundedness of all solutions are successfully guaranteed. This valuable result is obtained only through limited amount of qualitative and quantitative data, which may be provided from relatively simple experiments for real applications. The design procedure, in the cast of a new theorem, is mainly derived from the specific version of Small Gain Theorem for autonomous monotone systems; which ensures the global attractivity of the desired set-point. A biological simulation example illustrates the effectiveness and applicability of the proposed control strategy.

In this study, the problem of fault detection (FD) system design for linear discrete time-varying systems is addressed. Different from most existing methods which separately handle residual generator and evaluator, this study focuses on integrated design of residual generator and evaluator. Instead of traditional trade-off design between sensitivity to faults and robustness to disturbances, this study aims at maximising FD rate given a predefined false alarm rate. The proposed approach in this study follows two steps: first, in the norm-based framework, the parity relation-based offline FD system achieves maximal FD rate by minimising the set of undetectable faults; then the optimal offline FD system is equivalently transformed into a recursive FD algorithm for online use. Relationship between the obtained solution in this study and the existing ones is analysed. The performance improvement of the proposed approach is illustrated by comparing against existing methods through Monte Carlo simulations.

This study investigates the feedback control for a class of switched positive linear systems (SPLSs). By means of the linear programming approach, output-feedback and state-feedback controllers for the underlying systems with average dwell time are designed, respectively. Under these controllers, the closed-loop systems are positive and asymptotically stable. These results obtained provide a way to solve the control synthesis problems of SPLSs by multiple linear copositive Lyapunov functions. Finally, an example is given to illustrate the validity of the present design.

In this study, the problem of iterative learning control for a class of non-linear switched systems with arbitrary switched rules is considered. It is assumed that the considered switched systems are operated during a finite-time interval repetitively, and then the iterative learning control scheme can be introduced. It is also shown that under some given conditions, the D-type iterative learning control law can guarantee the asymptotic convergence of the tracking error between the desired output and the actual output for the entire time interval through the iterative learning process. Two examples are given to illustrate the effectiveness of the proposed approach.