Parameter space design of robust control systems

Find a state or output feedback with fixed gains such that nice stability (defined by a region in the eigenvalue plane) is robust with respect to large plant parameter variations, sensor failures, and quantization effects in the controller. Keep the required magnitude of control inputs small in this design. A tool for tackling such problems by design in the controller parameter spaceKis introduced. Pole placement is formulated as an affine map from the spacePof characteristic polynomial coefficients to theKspace. This allows determining the regions in theKspace, which place all eigenvalues in the desired region in the eigenvalue plane. Then tradeoffs among a variety of different design specifications can be made inKspace. The use of this tool is illustrated by the design of a crane control system. Several open research problems result from this approach: graphical computer-aided design of robust systems, algebraic robustness conditions, and algorithms for iterative design of robust control systems.     

A guided tour through a maze: Computer aided sector design

This paper describes an interactive system that assists a police planner in developing an equitable sector configuration. The system aids in balancing any one of four performance measures: (1) workload, (2) sector travel time, (3) atom travel time, (4) preventive patrol coverage. In addition the user can tradeoff between these performance measures as well as consider precinct-wide average travel time. The system operates in two modes: (1) automatic: the user chooses one of the four goals and the system iteratively modifies an existing configuration until that goal is reached; and, (2) user-directed: the user chooses a goal and the system iteratively suggests several alternative ways of approaching that goal with the user iteratively selecting one of the alternatives.     

Interactive parameter space design for robust performance of MISO control systems

This paper presents a method for the design of nonconservative low-order controllers achieving robust performance in the case of multi-input single-output parallel structure plants subject to unstructured uncertainty. The first step is the analytical generation of gain-phase controller bounds, as in quantitative feedback theory (QFT). Then, to avoid the difficult step of QFT loop shaping, which often produces high-order controllers, these bounds are translated into the controller parameter space where the iterative design of low fixed order controllers takes place. This, as well as the design transparency offered by this technique, constitutes appreciable advantages over the other popular robust performance design method of /spl mu/-synthesis. Other important features are the fact that no extra conservatism is introduced by the method presented and the fact that the method is directly compatible with a sequential loop closing strategy. Finally, the direct search optimization of any additional secondary criteria is possible.     

MRCD: a genetic algorithm for multiobjective robust control design

A genetic algorithm (GA) for the class of multiobjective optimization problems that appears in the design of robust controllers is presented in this paper. The design of a robust controller is a trade-off problem among competitive objectives such as disturbance rejection, reference tracking, stability against unmodeled dynamics, moderate control effort and so on. However, general methodologies for solving this class of design problems are not easily encountered in the literature because of the complexity of the resultant multiobjective problems. In this paper, a recently developed class of GAs, multiobjective GAs, are used to solve robust control design problems. Here, a new algorithm, called multiobjective robust control design, has been proposed. The structure and operators of this algorithm have been specifically developed for control design problems. The performace of the algorithm is evaluated by solving several test cases and is also compared to the standard algorithms used for the multiobjective design of robust controllers.     

Links between the parameter space and frequency domain methods of robust control

Two main approaches to robust control are the parametric and frequency domain approaches. The aim of this paper which starts with tutorial sections on both approaches is to compare and establish links between the parameter space method of the parametric approach and frequency domain methods, with special emphasis on the structured singular value (SSV) method. In an effort to combine the useful properties of both approaches, two SSV minimization based design techniques in controller parameter space are presented here. A simple interpretation of real SSV in plant parameter space and its further implications for analysis and design are also presented. Copyright © 2001 John Wiley & Sons, Ltd.     

A control oriented guided tour in oilwell drilling vibration modeling

In drilling operations, the interaction of the drillstring with the borehole leads to vibrations affecting the performance and increasing the drilling cost. The development of controllers to get a faster and efficient drilling operation is based on a mathematical modeling allowing a proper system characterization and the identification of the vibration sources to avoid them or mitigate their influence. This paper presents an overview of the modeling of axial and torsional self-excited drilling vibrations.     

Using regression analysis: A guided tour

Regression analysis is a useful tool for data analysis, forecasting, and the estimation of fixed and variable components of a data set. Students frequently need to take advantage of data analyses offered by regression methodology. Regression computer packages are readily accessible and easy to use. Without an understanding of the assumptions underlying the linear regression model, it may be difficult for the student to interpret the results of regression analysis in a meaningful manner. The present paper is designed to guide the user through the use and interpretation of multiple regression analysis. To facilitate his understanding of the process, a flowchart of the process is also presented.     

A cascade structure for robust control design

In the context of robust control theory and applications, the well-known variable structure approach and the so-called binary control approach offer both interesting properties and disadvantages. In this paper, the idea of combining the two techniques in order to define a cascade structure exploiting the desirable features of both approaches is discussed, and its natural applicability to the control of a nonlinear electrical actuator is shown by a simulation case     

小型无人直升机的无模型自适应鲁棒控制设计

本文针对小型无人直升机的姿态控制问题,考虑到现有基于模型的控制方法对系统动力学模型的依赖性,以及未建模动态对系统控制性能的影响,设计了一种新的基于数据驱动的无模型自适应鲁棒控制律.通过基于数据驱动的设计方法,降低了控制器对直升机动力学模型先验知识的依赖,补偿了未建模不确定性的影响.仅利用无人直升机的输入输出数据,即可实现对无人直升机系统的稳定姿态控制.然后本文结合离散滑模控制设计,补偿了未知外界扰动的影响,提高了系统的鲁棒性,并通过理论证明了控制误差的收敛性和闭环系统的稳定性.最后,在本研究组自主开发的无人直升机飞行控制实验平台上,进行了无人机实时控制实验.实验结果表明,本文所提出的控制算法取得了很好的姿态控制效果,并对系统不确定性和外界风扰动具有良好的鲁棒性.     

A scheme for robust trajectory control of space robots

This paper presents a scheme for robust trajectory control of free-floating space robots. The idea is based on the overwhelming robust trajectory control of a ground robot on a flexible foundation and robust foundation disturbance compensation presented elsewhere. No external jets/thrusters are required or used in the scheme. An example of a three-link robot mounted on a free-floating space platform is considered for demonstrating the efficacy of the control scheme. Bond graph technique has been used for the purpose of modeling and simulation. Robustness of the control scheme is guaranteed since the controller does not require the knowledge of the manipulator parameters.     

Passification-based robust flight control design

A passification-based robust autopilot for attitude control of flexible aircraft under parametric uncertainty is designed. A high gain controller with forced sliding motions is used to secure good performance over a wide range of the aircraft model parameters. The shunting method is applied to ensure closed-loop system stability under lack of aircraft state information. The series reference model is used to assign the desired closed-loop system performance. An example illustrating a typical design procedure for aircraft attitude control in the horizontal plane for different flight conditions is given. The simulation results demonstrate the efficiency and high robustness of the suggested control system.     

Multivariable robust control of a rotary dryer: Analysis and design

This paper describes the analysis and control of an industrial process usually controlled in manual mode: a continuous rotary dryer. Two process variables are controlled simultaneously: the outlet moisture of the dried material and the output temperature of the exhaust air. To do this, the flow of wet material and the flow of fuel are used as control variables. Thus, this constitutes a MIMO system, with two inputs and two outputs. The system is identified at several operating points, and a controllability analysis is performed in order to find any constraints in its performance. A multivariable robust control based on the $H_{\infty }$ mixed sensitivity problem is proposed. Simulation and experimental results are included, using different adjustments of the weighting matrices based on the performed experiments. Results attained agree with the controllability analysis conclusions.     

Continuous robust control design for nonlinear uncertain systemswithout a priori knowledge of control direction

In this paper, a robust control scheme is proposed for a class of nonlinear systems that have not only additive nonlinear uncertainties but also unknown multiplicative signs. These signs are called control directions since they represent effectively the direction of motion under any given control. Except for the unknown control directions, the class of systems satisfy the generalized matching conditions. Nonlinear robust control is designed to identify on-line unknown control directions and to guarantee global stability of uniform ultimate boundedness without the knowledge of nonlinear dynamics except their size bounding functions. It is also shown that the proposed robust control can be made continuous through utilizing the so-called shifting laws that change smoothly and accordingly the signs of robust controls and that, no matter what time constants and gains of the shifting laws are, global stability is always ensured. The analysis and design is done using Lyapunov's direct method     

Intelligent robust control design of a precise positioning system

This paper addresses an intelligent uncertainty function to improve the robust stability and performance of H _∞ controlled system in terms of reduced conservatism. The system is identified, output performance and control signal requirements are controlled by proper selection of performance and control weighting functions. Adaptive Neuro Fuzzy Inference System (ANFIS) learns the uncertainty bounds of model uncertainty that results from unmodeled dynamics and parameter variations, then the developed uncertainty weighting function will be included in the synthesis of the H _∞ controller. ν-gap measure is utilized to validate the intelligent identified uncertainty bounds and measure the stability of the designed H _∞ controlled system as well. Experimental results on a servo motion system reveal the advantages of combining intelligent uncertainty identification and robust control. Improved performance is achieved. The proposed approach also allows for iterative experiment design.     

Iterative learning control for uncertain systems: Noncausal finite time interval robust control design

In this paper, we present a novel robust Iterative Learning Control (ILC) control strategy that is robust against model uncertainty as given by an additive uncertainty model. The design methodology hinges on ??_∞ optimization, but formulated such that the obtained ILC controller is not restricted to be causal, and inherently operates on a finite time interval. Optimization of the robust ILC (R‐ILC) solution is accomplished for the situation where any information about structure in the uncertainty is discarded, and for the situation where the information about the structure in the uncertainty is explicitly taken into account. Subsequently, the convergence and performance properties of resulting R‐ILC controlled system are analyzed. On an experimental set‐up, we show that the presented R‐ILC control strategy can outperform an existing linear‐quadratic norm‐optimal ILC approach and an existing causal R‐ILC approach based on frequency domain ??_∞ synthesis. Copyright © 2010 John Wiley & Sons, Ltd.     

Modified adaptive robust control system design

The robust control design problem is studied for a class of uncertain dynamical systems. The uncertainty of the system is time varying. The only assumption on the uncertainty is that it is bounded. No statistical property of the uncertainty is ever assumed and utilized. The robust control does not need the a priori estimation on the bound of uncertainty. An adaptive algorithm for the on-line estimation of the bound is constructed and the robust control is only dependent on this estimation. The adaptive algorithm is a modification of previous work by Corless and Leitmann (1983). The advantages of this new algorithm include a constant control design parameter (which should have been time varying previously) and the applicability to linear systems with mismatched uncertainty and measurement noise.