This paper considers the Model Predictive Control (MPC) set point tracking/regulation problem for a discrete LTI system, which is subject to a class of unbounded disturbances/tracking signals called extended constant signals of unknown structure. Examples of disturbances which belong to this class include constant disturbances as well as unbounded signals such as w[k]=√k and log (k), k=1,2,3,…. A discussion re the choice of window size for MPC is also made; in particular, it is shown that the window size must be larger than a certain lower bound, which can be easily determined, in order to guarantee closed loop stability in MPC control. The main contribution is a formulation of the system's plant equations under which, for output regulation, no knowledge of the structure or magnitude of disturbances is needed in order to achieve set point regulation for this class of extended constant signals. The result is of interest since it also implies that no disturbance observer is necessary in order to solve the set point tracking/regulation problem when full-state feedback is available. The results are experimentally verified.
The present paper addresses an extension of the classical problem of output regulation. It is shown how the availability of supplementary measurement outputs, in addition to the regulated variable, can be exploited to the purpose of overcoming certain current design limitations, extending the analysis to classes of systems with a possibly unstable zero dynamics.
This paper presents a method for digital PID and first order controller synthesis based on frequency domain data alone. The techniques given here first determine all stabilizing controllers from measurement data. In both PID and first order controller cases, the only information required are frequency domain data (Nyquist-Bode data) and the number of open-loop RHP poles. Specifically no identification of the plant model is required. Examples are given for illustration.
Motivated by the recent developments in networked control systems and control over wireless, this paper presents a comparison of five control algorithms that are based on PID, IMC and fuzzy gain scheduling techniques and discusses their performance in varying time-delay systems. The low complexity of the proposed algorithms makes their use attractive in resource-constrained environments such as wireless sensor and actuator networks. The control system consists of a controller, a simple process and an output delay in the feedback loop. Three different delay models are considered in this framework; constant, random, and correlated random delay. In addition to presenting modifications to the control algorithms to better fit the varying time-delay systems a delay-robust tuning method is proposed, and the performance of various controllers is evaluated using simulation. The results show the benefits of adapting the controller parameters based on delay measurement if its amplitude is significant with respect to the time-constant of the process. Nevertheless, the PID algorithm used in the study also performs well in all scenarios, and this is achieved by its careful tuning.
This paper presents a system level optimal design by integrating the feedback control design and sensor/actuator selection. Instead of the traditional approach of designing the feedback control law with predefined sensors and actuators, we determine the precision of sensor/actuator and the output feedback control law simultaneously such that the total sensor/actuator precision is minimized, subject to the specified control performance (output covariance upper bound). In the case of the full order output feedback control, we provide a complete solution to this problem which is converted into an equivalent convex problem. This convex problem is used as the basis of an ad hoc algorithm to reduce the number of sensors and actuators, starting from a large admissible set. The algorithm iteratively deletes instruments until the design specifications can no longer be met.
One of the important examples of mechatronic systems can be found in autonomous ground vehicles. Autonomous ground vehicles provide a series of challenges in sensing, control and system integration. In this paper we consider off-road autonomous vehicles, automated highway systems and urban autonomous driving and indicate the unifying aspects. We specifically consider our own experience during the last twelve years in various demonstrations and challenges in attempting to identify unifying themes. Such unifying themes can be observed in basic hierarchies, hybrid system control approaches and sensor fusion techniques.
Simplified matrix pencil formulae for solution of the H∞ control problem for the case 0≤σ(D11)≤γ are presented. The formulae are useful in developing more numerically robust algorithms in H∞ control.The formulae apply to descriptor form plants.
Proportional-integral-derivative (PID) controller is the most predominant industrial controller that constitutes more than 90% feedback loops. Time domain performance of PID, including peak overshoot, settling time and rise time, is directly relevant to PID parameters. In this work we propose an iterative learning tuning method (ILT) - an optimal tuning method for PID parameters by means of iterative learning. PID parameters are updated whenever the same control task is repeated. The first novel property of the new tuning method is that the time domain performance or requirements can be incorporated directly into the objective function to be minimized. The second novel property is that the optimal tuning does not require as much the plant model knowledge as other PID tuning methods. The new tuning method is essentially applicable to any plants that are stabilizable by PID controllers. The third novel property is that any existing PID auto-tuning methods can be used to provide the initial setting of PID parameters, and the iterative learning process guarantees that a better PID controller can be achieved. The fourth novel property is that the iterative learning of PID parameters can be applied straightforward to discrete-time or sampled-data systems, in contrast to existing PID auto-tuning methods which are dedicated to continuous-time plants. In this paper, we further exploit efficient searching methods for the optimal tuning of PID parameters. Through theoretical analysis, comprehensive investigations on benchmarking examples, and real-time experiments on the level control of a coupled-tank system, the effectiveness of the proposed method is validated.
This paper is concerned with the finite horizon H∞ full-information control for continuous time systems with multiple input delays. The main contributions of the paper are two folds. First, parallel to the duality between the LQR of linear systems without delays and the optimal filtering, we establish the duality between the H∞ full-information control of systems with multiple input delays and an H∞ smoothing estimation of a stochastic backward system without involving delays. The duality allows us to address the complicated multiple input delays system problem via the standard projection and innovation analysis. Secondly, by defining a stochastic indefinite linear space and applying a re-organized innovation analysis, an explicit controller is constructed in terms of two standard Riccati differential equations. As special cases, solutions to the H∞ control problem for systems with single input delay and the H∞ control with preview are obtained.
This paper explores the potential of biological oscillators as a basic unit for feedback control to achieve rhythmic motion of locomotory systems. Among those properties of biological control systems that are useful for engineering applications, we focus on decentralized coordination, that is, the ability of uncoupled neuronal oscillators to coordinate rhythmic body movements to achieve locomotion with the aid of local sensory feedback. We will consider the reciprocal inhibition oscillator (RIO) as a candidate for the basic control unit, and show that uncoupled RIOs can achieve decentralized coordination of a prototype mechanical rectifier (PMR) that captures essential dynamics underlying animal locomotion by a simple arm-disk configuration. Optimality of the induced locomotion is studied in comparison with analytical results we derive for statically optimal PMR locomotion.