Deadlock-free supervisory control for automated manufacturing systems (AMSs) has been a popular research subject. However, for numerous researchers, their studies are merely practicable on the assumption that resources cannot fail. In practice, resource failures can occur unexpectedly. In this study, the authors take into consideration deadlock and blocking problems in systems with assembly operations embedded in flexible routes. Their objective is to develop a robust supervisor that controls resource allocation and selects flexible routes such that stagnant parts requiring failed resources do not block the movement of parts not necessarily requiring failed resources. That is to say, it must ensure that parts not necessarily requiring any failed resource can continue their operations. The proposed policy tries to take full advantage of the shared resource capacity to improve systems' performance. By modelling AMSs with Petri nets, their robust supervisor predicts in advance whether the currently-available resources are sufficient to support a token to advance into the desired destination in a single process. The method is achieved in an online and distributed way and avoids enumerating states. At last, several simulation examples show the correctness of their proposed method.

This study investigates the problem of global finite-time attitude tracking for spacecraft subject to actuator constraints and attitude measurements only. A velocity-free saturated hybrid proportional–derivative (PD) plus spacecraft dynamics (PD+) control is proposed. Benefiting from the hybrid control technique, the unwinding phenomenon is completely avoided and the global stability is achieved. Lyapunov stability theory and homogeneous method are employed to show global finite-time tracking. Advantages of the proposed control include simple and intuitive structure and the absence of velocity measurements, and thus it is ready to implement. An additive feature is that the proposed control is explicitly upper bounded and hence it can assure that actuator constraints will not be violated by selecting the control gains a priori. Simulations are performed to illustrate the effectiveness and improved performance of the proposed approach.

This study considers the consensus tracking problem of singular multi-agent systems by using an iterative learning control approach. Here, the communication among the followers is described by a directed graph, and only a portion of the followers can receive the leader's information. For such singular multi-agent systems, a unified iterative learning algorithm is proposed in both continuous-time domain and discrete-time domain. Furthermore, the convergence condition of the algorithm is presented and analysed. In this study, the main contribution is to extend the iterative learning control theory from multi-agent systems to singular multi-agent systems. It is shown that the algorithm can guarantee the outputs of the followers converge to the leader's trajectory on a finite time interval along the iteration axis. Finally, the provided examples illustrate the effectiveness of the theoretical results.

Due to the long-range data communication and complex Mars environment, the Mars lander needs to promote the ability to autonomously adapt uncertain situations ensuring high precision landing in future Mars missions. Based on the analysis of multiple disturbances, this study demonstrates an enhanced predictor–corrector guidance method to deal with the effect of atmospheric uncertainties during the entry phase of the Mars landing. In the proposed method, the predictor–corrector guidance algorithm is designed to autonomously drive the Mars lander to the parachute deployment. Meanwhile, the disturbance observer is designed to onboard estimate the effect of fiercely varying atmospheric uncertainties resulting from rapidly height decreasing. Then, with the estimation of atmospheric uncertainties compensated in the feed-forward channel, the composite guidance method is put forward such that both anti-disturbance and autonomous performance of the Mars lander guidance system are improved. Convergence of the proposed composite method is analysed. Simulations for a Mars lander entry guidance system demonstrates that the proposed method outperforms the baseline method in consideration of the atmospheric uncertainties.

This study addresses the consensus problem for linear multi-agent systems subject to external disturbances under the leaderless framework. A novel distributed dynamic output feedback control protocol is proposed, which utilises not only relative output information of neighbouring agents but also relative state information of neighbouring controllers. Through model transformation, the consensus control problem of multi-agents network is reduced to a set of independent stabilisation subproblems for n-dimensional linear systems. Sufficient analysis conditions are derived using the Lyapunov method. An important contribution of this work lies in that the leaderless output-feedback consensus synthesis conditions are convexified without introducing any conservatism and formulated as linear matrix inequalities, which can be solved efficiently via convex optimisation. This is achieved by using a novel dynamic output-feedback controller structure. A numerical example has been used to demonstrate the advantage of theoretical results.

In this study, the authors propose a framework for inversion-based estimation of certain categories of faults in discrete-time linear systems. The fault signal, as an unknown input, is reconstructed from its projections onto two subspaces. The first projection is achieved through an algebraic operation, whereas the second projection is obtained by a dynamic filter whose poles coincide with the transmission zeros of the system. Feedback is then introduced to stabilise this filter as well as to provide an unbiased estimate of the unknown input. Their proposed methodology has two distinctive and practical advantages. First, it represents a unified approach to the problem of inversion of both minimum and non-minimum phase systems as well as systems having transmission zeros on the unit circle. Second, the feedback structure ensures that the proposed scheme is robust to noise. They have shown that the proposed inversion filter is unbiased for certain categories of faults. Finally, they have demonstrated the capabilities and performance of the proposed methodology through several numerical simulation case studies.

This study is concerned with the minimal controllability problem of a connected threshold graph following the Laplacian dynamics. The goal is to find the minimum number of controllers and a small set of vertices for the controllers to connect to render the graph Laplacian controllable. A simple algorithm is provided to generate a spanning set of orthogonal Laplacian eigenvectors of the graph from a straightforward computation on its Laplacian matrix. A necessary and sufficient condition for the graph to be Laplacian controllable is then proposed. The condition suggests that the minimum number of controllers to make a connected threshold graph Laplacian controllable is the maximum multiplicity of Laplacian eigenvalues of the graph, and this minimum can be achieved using a binary control matrix. If a controller can be connected to one vertex only, the minimum number is the difference between the number of vertices in the graph and the number of vertices with different degrees. The condition also implies that the controllers ensuring the Laplacian controllability should be connected to the vertices with repeating degrees to break the symmetry of the network topology. Several examples are provided to illustrate the authors' results.

Parameter estimation of Markov-switching Hammerstein systems is presented in this study. The random switching mode is described by a hidden Markov model. An input non-linear controlled autoregressive Hammerstein model is applied to describe the dynamic process of each local model. By applying the decomposition technique to the non-linear mapping of the input signal, each local model is written as a linear-in-parameter form. The variational Bayesian approach is applied to identify the switched Markov Hammerstein models. Both the discrete-valued switching mode state and the unknown parameters in each local model are estimated. Moreover, instead of the point estimation, posterior distributions of unknown parameters for each local model are obtained. Numerical simulation examples and a hybrid tank experiment are presented to verify the effectiveness of the proposed approach.

To obtain an effective state estimator for industrial processes, estimator needs to be designed to match the characteristics of noise. In this study, a new filter is proposed focusing on asymmetric measurement noise with probable outliers. By learning a time-varying skew t distribution using the variational Bayesian technique, the authors' method can estimate the system state and update the noise statistics simultaneously. A numerical simulation, as well as an experiment on the hybrid tank system, is conducted to demonstrate the performance. It shows that the proposed filter is superior to the existing solutions, especially when the statistics of measurements noise are unknown.

This study deals with the fault-tolerant control problem for a class of two independent wheeled mobile robot systems with actuator faults and unknown robot parameters. Partial loss of control effectiveness and bias-actuator faults are addressed without knowing eventually faulty information of actuators. Adaptive schemes are developed to estimate unknown faulty parameters of actuator faults and unknown robot parameters of viscous friction factor and driving gain. Then, the backstepping control technique is utilised to construct the state feed-back control strategy based on the adaptive estimations. By using the Lyapunov stability theory, it is shown that the forward speed and azimuthal angle of wheeled mobile robots can track the given trajectory asymptotically in the case of actuator faults and unknown parameters. The effectiveness of the proposed designs is illustrated via a mobile robot system.

This study investigates the design problem of the linear sliding variable-based sliding mode controllers of descriptor systems via output information. A generalised regular form, which is the counterpart of that for normal systems, is first recalled. Then a static output feedback sliding mode control strategy is developed to stabilise the descriptor system under consideration and the result is further improved by a reduced-order observer-based sliding mode control strategy. Necessary and sufficient conditions are derived such that the resulting sliding motion is regular and asymptotically stable. It is shown that the main merit of the static output feedback sliding mode control method is the simplicity and that of the reduced-order observer-based sliding mode control method is the elimination of a restrictive assumption frequently used in the static output feedback sliding mode control method. Finally, two connected one-mass oscillators model is numerically simulated to testify the validity of the proposed methods.

This study concerns the parameter identification issues for a class of multivariable systems with moving average noise. The main contributions are to transform a multivariable system into several subsystems for reducing the computational complexity by means of the decomposition technique and to deal with the coloured noise for improving the estimation accuracy by using the maximum likelihood principle and the iterative estimation theory. As the maximum likelihood gradient-based iterative algorithm makes sufficient use of all the observed data at each iteration, the parameter estimation accuracy can be enhanced. The numerical simulation results demonstrate that the proposed algorithm has faster convergence rates and better tracking performance than the compared multivariable extended stochastic gradient algorithm.

In this study, a dynamic (infinity-norm) constrained control allocation scheme is designed for attitude tracking control of combined spacecraft with inertia uncertainties and external disturbances. A disturbance-observer-based constrained backstepping control law is developed to generate the control command signals considering non-symmetric constraints on control input, where the lumped disturbance containing inertia uncertainties and external disturbances is compensated by the output of stable non-linear disturbance observer. The control scheme can guarantee that the attitude and angular velocity tracking error converge to small neighbourhood of zero by appropriately tuning the control parameters. With the consideration of physical amplitude and rate constraints on actuators, the dynamic constrained control allocation problem is solved by linear programming technique. Numerical examples demonstrate the effectiveness of the proposed disturbance-observer-based constrained backstepping control method and the dynamic constrained control allocation algorithm.

In the pressure management of large scale water distribution network (WDN), a distributed zone model predictive control (MPC) is proposed to keep the terminal water head in thedesired pressure range for satisfying the customer's demand, avoiding frequently operating of the actuator and reducing the correlation between subsystems. To ensure the existence of feasible solutions of constrained distributed zone MPC, a new reference trajectory of the tank level is introduced as an optimised variable. With the consideration of the desired pressure range constraints on the new reference trajectory and some physical constraints on the corresponding physical variables, a distributed zone MPC is presented to minimise the weighted sum of the three terms in the proposed performance index. To achieve the convergence of distributed zone MPC optimisation problem, an augmented Lagrangian formulation is applied to the distributed coordinated strategy. The proposed distributed zone MPC is applied to the WDN in the Shinan district of Shanghai, and the effectiveness of the method is illustrated.

A modification to the control framework for uncertain systems with actuator delays is presented. Specifically, a time delay is introduced in the control input of the state predictor to compensate for the destabilising effect of an input delay in the plant. For this modified framework, the analysis shows that the output of the adaptive system closely follows the behaviour of a suitably defined, non-adaptive, stable reference system provided that a delay-dependent stability condition is satisfied and the adaptive gain is chosen sufficiently large. The set of combinations of input delay and compensation delay for which the stability condition is satisfied contains an open set of pairs of positive values provided that a filter bandwidth is chosen sufficiently large. The efficacy of the delay compensation is illustrated by a simple example. A numerical continuation is also performed to explore the stability region for a case where this can be approximated a priori.

In this study, the problem of finding state bounds is considered, for the first time, for a class of positive time-delay coupled differential-difference equations (CDDEs) with bounded disturbances. First, the authors present a novel method, which is based on non-negative matrices and optimisation techniques, for computing a like-exponential componentwise upper bound of the state vector of the CDDEs without disturbances. The main idea is to establish bounds of the state vector on finite-time intervals and then, by using the solution comparison method and the linearity of the system, extend to infinite time horizon. Next, by using state transformations, they extend the obtained results to a class of CDDEs with bounded disturbances. As a result, componentwise upper bounds, ultimate bounds and invariant set of the perturbed system are obtained. The feasibility of the obtained results is illustrated through a numerical example.

This study aims at investigating the reachable set estimation for delayed memristive neural networks (MNNs) with bounded disturbances. Under the framework of Filippov's solution and differential inclusion theory, a reachable set estimation criterion is established by means of Lyapunov functional. The obtained condition guarantees that all the states of MNNs are bounded. Moreover, a delay-dependent feedback controller design problem for MNNs is also considered. The exponential stability and stabilisability conditions are obtained. Two numerical examples are presented to demonstrate the usefulness of the proposed results.

This study is to estimate the bounds for the maximum allowable sampling interval (MASI) and the coarsest quantisation density (CQD) that guarantee the stability of non-linear systems with multi-input signals sampled and logarithmic quantised. First of all, hybrid feedback systems are proposed to describe the non-linear control systems. Then a sufficient condition is provided to ensure that the systems are uniformly globally exponentially stable. A crucial step is to find a novel Lyapunov function to verify the stability conditions in the sense of the hybrid framework. Meanwhile, explicit bounds for the MASI and CQD are obtained to guarantee stability. Furthermore, in the case of no quantisation or no sampling, some special stability criteria can be also obtained. Finally, some examples are given to illustrate the effectiveness of the proposed theorem.

In this study, the problem of stochastic stability of non-linear impulsive semi-Markov jump systems is investigated. Using the method of stochastic Lyapunov functions, the authors develop some sufficient conditions about stochastic stability for a class of non-linear impulsive semi-Markov jump systems with unbounded transition rates. Particularly, the obtained results generalise and complement some published literatures. Finally, some examples are given to show the effectiveness and advantages of the proposed results.

This study investigates the distributed fixed-time consensus tracking for high-order uncertain non-linear multi-agent systems with switching topologies. Each follower is assumed to be in strict-feedback form with both unknown parameters and mismatched disturbances. Different from traditional centralised tracking control problem, only a subset of the agents can acquire the desired trajectory information. By adopting backstepping method and dynamic surface control technique, the distributed fixed-time consensus protocol and appropriate adaptive laws are designed without requiring knowledge of the upper bounds of disturbances. Moreover, the problem of ‘explosion of complexity’ in standard backstepping design is avoided. Under the proposed control protocol, it is shown from the Lyapunov-based analysis that the tracking errors between all followers and the leader can converge to a small neighbourhood of the origin within a fixed time, even though the communication structures among agents dynamically change over time. Finally, a numerical example is carried out to illustrate the effectiveness of the proposed method.