摘要:

In this paper, a global tracking control problem of a constrained robot is considered to be formulated by the hybrid coordinate frames of the joint-space coordinate for the motion of the robot's joints, and the task-space coordinate for the constraint forces on the robot's end-effector. After model reduction and a proper change of control variable, reduced state-space error dynamics are explicitly obtained and used to derive the nonlinear quadratic optimal tracking controller for the constrained robot. The advantage of the controller is that it provides an asymptotic tracking solution to minimize, simultaneously and systematically, the tracking error about the desired position and velocity of the joint-space motion, with a minimum of applied torques which affect the kinetic energy only. In addition, it is interesting that our controller can achieve the exponential tracking of the task-space constraint force.

ISSN:0020-7179    

DOI:10.1080/00207179308923001

出版商:Taylor & Francis Group    

年代:1993

数据来源: Taylor

摘要:

Recent theoretical advances in nonlinear systems theory insist on the relevance of dynamics where control derivatives appear. We discuss here a simplified model of a crane which exhibits such a description.

ISSN:0020-7179    

DOI:10.1080/00207179308923002

出版商:Taylor & Francis Group    

年代:1993

数据来源: Taylor

摘要:

Performance characteristics of a sliding mode controller depend essentially on a selected sliding manifold. Customarily, the variable-structure controllers operate with linear sliding surfaces. However, using nonlinear or time-varying manifolds may provide definite advantages; in particular, they may provide faster convergence. In this paper, a control algorithm is presented which guarantees the occurrence of the sliding mode on nonlinear or time-varying manifolds. The bounds needed for calculating the switching gains are determined in terms of parameter variations rather than bounds on the system matrices. The parameter bounds are usually evident and can be easily established or estimated, while matrix bounds can rarely be justified in practical problems. In this paper, we also present an adaptive version of the controller, which provides an automatic estimation of the parameter bounds. The robustness of the controller with respect to a class of state-dependent uncertainties is analysed by the Lyapunov method. Application of the developed theory is illustrated by an example of robotic manipulator control with a nonlinear sliding manifold.

ISSN:0020-7179    

DOI:10.1080/00207179308923003

出版商:Taylor & Francis Group    

年代:1993

数据来源: Taylor

摘要:

This paper is concerned with the problem of designing a robust output tracking controller forMIMOnonlinear systems with weakly non-minimum phase. Based on our system formulation, control plants with uncertainties and/or with actuator dynamics fall into the class under consideration. The controller design here is divided into two phases: fast feedback control and slow feedback control, so that a final composite control is obtained. The former is chosen to stabilize the boundary layer system, whereas the latter essentially handles the mismatched uncertainties after the system is reformulated. Under some mild assumptions, it is shown that the overall states are bounded and the tracking errors converge to a residual set whose size is a classkfunction of ϵ. As ϵ → 0, the residual set shrinks to the origin. An interesting application to a simplified aircraft model with fast actuator dynamics, which turns out to be weakly non-minimum phase, is given. The computer simulation has verified the expected satisfactory performance.

ISSN:0020-7179    

DOI:10.1080/00207179308923004

出版商:Taylor & Francis Group    

年代:1993

数据来源: Taylor

摘要:

Multilayer neural networks are used in a non-linear adaptive control problem. The plant is an unknown feedback linearizable continuous-time system with relative degree ⩾1. The single-input:/single-output system is studied first and then the methodology is extended to control square multi-input/multi-output systems. The control objective is for the plant to track a reference trajectory, and the control law is defined in terms of the outputs of the neural networks. The parameters of the networks are updated on-line according to an augmented tracking error and the network derivatives. A local convergence theorem is given on the convergence of the tracking error. This control algorithm is applied to control a two-input/two-output relative-degree-two system.

ISSN:0020-7179    

DOI:10.1080/00207179308923005

出版商:Taylor & Francis Group    

年代:1993

数据来源: Taylor

摘要:

A set of necessary and sufficient conditions under which a generalH2-optimal control problem has a unique solution is derived. It is shown that the solution for anH2-optimal control problem, if it exists, is unique if and only if (i) the transfer function from the control input to the controlled output is left invertible, and (ii) the transfer function from the disturbance to the measurement output is right invertible.

ISSN:0020-7179    

DOI:10.1080/00207179308923006

出版商:Taylor & Francis Group    

年代:1993

数据来源: Taylor

摘要:

A generalization of polynomial interpolation to the matrix case is introduced and applied to problems in systems and control. It is shown that this generalization is most general and it includes all other such interpolation schemes that have appeared in the literature. The polynomial matrix interpolation theory is developed and then applied to solving matrix equations; solutions to the diophantine equation are also derived. The relation between a polynomial matrix and its characteristic values and vectors is established and it is used in pole assignment and other control problems. Rational matrix interpolation is discussed; it can be seen as a special case of polynomial matrix interpolation. It is then used to solve rational matrix equations including the model matching problem.

ISSN:0020-7179    

DOI:10.1080/00207179308923007

出版商:Taylor & Francis Group    

年代:1993

数据来源: Taylor

摘要:

This paper characterizes the degrees of freedom in theH∞1-block problem and subsequently uses these degrees of freedom to obtain desirable properties in the Hankel approximation. In particular, for a special class of transfer function matrices, the degrees of freedom can be deployed to obtain station-arity of the approximation system gramians which, in turn, imposes flatness onallof the error-system singular values as well as providing completely stable projections in the context of a super-optimal algorithm. Subsequently, the class constraints are relaxed in order to generalize the results.

ISSN:0020-7179    

DOI:10.1080/00207179308923008

出版商:Taylor & Francis Group    

年代:1993

数据来源: Taylor

摘要:

The stabilization of uncertain dynamical systems on ℝnthat admit a decomposition into two coupled subsystems of dimensionncandnrrespectively, is studied. We refer to thenc-dimensional sybsystem as the reduced-order system and to thenr-dimensional subsystem as the residual system: the overalln-dimensional system is termed the full system—the prototype for which is a mechanical model arising in the study of the active control of structures. Each subsystem is modelled by a differential equation with a linear nominal part and a nonlinear perturbation of a specified class. The output available for feedback purposes is an ℝnc-valued linear combination of the state components of the full system and is subject to bounded measurement noise (with known bound). An output feedback strategy is described and sufficient conditions are obtained for the existence of a (calculable) global uniform compact attractor (containing the state origin) for controlled systems of this class.

ISSN:0020-7179    

DOI:10.1080/00207179308923009

出版商:Taylor & Francis Group    

年代:1993

数据来源: Taylor

摘要:

In this paper, a finite impulse response (FIR) prefilter design for an FIR system is proposed. The proposed method assures that the step response overshoot and undershoot of the given system and the prefilter in series do not exceed two prespecified bounds. This design is then applied to two deadbeat controllers, the first in the form of a unity feedback controller and the second in the form of a two-degree-of-freedom controller. Several numerical examples are given to illustrate the effectiveness of the proposed design.

ISSN:0020-7179    

DOI:10.1080/00207179308923010

出版商:Taylor & Francis Group    

年代:1993

数据来源: Taylor