文档下载 文档下载导航
设为首页 | 加入收藏
搜索 请输入内容:  
 导航当前位置: 文档下载 > 所有分类 > IT/计算机 > 计算机软件及应用 > Quasi-continuous high-order sliding mode controllers.
侵权投诉

Quasi-continuous high-order sliding mode controllers.

Proceedings of the 42nd WEE Conference on Decision md Control Maui, Hawaii USA, December 2003

ThP14-6 h i e Levant

QUASI-CONTINUOUS HIGH-ORDER SLIDING-MODE CONTROLLERS School o f Mathematical Sciences. Tel-Aviv Wniversiy, Ramat-Aviv, 69978 Tel-Aviv, Israel E-mail: levant@post.tau.ac.iI. Tel.: 972-3-6408812, Fer: 972-3-6407543 Abstract. A universal finite-time-convergent controller is developed capable to control the output a of any uncertain SISO system of a known permanent relative degree r. The mode a E 0 (r-sliding mode) is established by means of a control dependent only on a, a, _.., and continuous everywhere except the set a= a= ...= U(,-I1= 0. An output-feedback controller version is also developed. With a being the tracking deviation, the exact finite-timeconvergent output tracking is provided in the absence of output noises, otherwise the tracking accuracy is proportional to the magnitude of the noise. In the latter case a= 0 is not attained producing an unswitched control continuous in time. The resulting performance is significantly improved compared with known r-sliding controllers. discrete sampling and, properly used, totally remove the chattering effect. In order to remove the chattering, the control derivative is to be treated as a new control. The discontinuity set of controllers[8, IO] is a stratified union of manifolds with codimension varying in the range from 1 to r. Unfomately, the complicated structure of the controller discontinuity set causes certain redundant transient chattering. To avoid it one needs an additional artificial increase of the relative degree[SI. A sliding-mode controller of a new type is proposed in this paper, providing for the control being a function of a, 6, 8, ._.,a(-1), continuous everywhere except the manifold. defmed by the r-sliding mode equations a=,. - J=.. a=,,,= a(-1)= 0. (1) The mode a= 0 is established after a finite-time transient. In the presence of errors in evaluation of the output and its derivatives, a motion in some vicinity of (I) takes place. As a result the control is practically a continuous function of time, for the trajectoly does not hit manifold (1). The lacking derivatives are produced in their tum by recently proposed robust exact differentiators with finite time convergence[7, 101. The resulting output-feedback controller provides for exact tracking a= 0 after a finitetime transient when the measurements of the deviation a are exact, and for the tracking error proportional to the maximal measurement error otherwise[9, lo]. Its transient is smoother and the tracking accuracy is higher than those of the known r-sliding controllers[S, IO]. The simulation demonstrates the practical applicability of the new controller. 2.

d”

1 . Introduction

Control under heavy uncertainty conditions remains one of the main subjects of the modem control theory. While a number of advanced methods like adaptation, absolute stability methods or the back-stepping procedu

re are based on relatively detailed knowledge of the controlled system, the sliding-mode control approach requirements are more moderate. The idea is to react immediately to any deviation of the system kom some properly chosen constraint steering it back by a sufficiently energetic effort. Slidingmode implementation is based on its insensitivity to external and internal disturbances and high accuracy[13,14]. The main drawback of the standard sliding modes is mostly related to the so-called chattering effect caused by the high-frequency control switching[ 12,4]. Let the constraint he given by the equation a= s w(f)= 0, where s is some available output variable of an uncertain single-input-single~utput ( W O ) dynamic system and w(r) is an unknown-in-advance smooth input to be tracked in real time. Then the standard sliding-mode control U= - k sign a may be considered as a universal output controller applicable if the relative degree is 1, i.e. if 6 explicitly depends on the control U and 6:> 0. Higherorder sliding mode[6, I] is applicable for controlling SISO uncertain systems with arbitrary relative degree. The correspondent finite-time-convergent controllers (r-sliding controllers)[8, IO] require actually only the knowledge of the system relative degree. The produced control is a discontinuous function of the tracking deviation a and of its real-time-calculated successive derivatives a, 6, 6, .._, a(7-1) . The controllers provide also for higher accuracy with

-

Preliminaries and the problem statement

Consider a smooth dynamic system with smooth output function a, and let the system be closed by some possiblydynamical discontinuous feedback and understood in the Filippov sense[3]. Then, provided that successive total (,-I) time derivatives a, 6, ..., a are continuous functions of the closed-system state-space variables; and set ( I ) is a (,-I) non-empty integral set, the motion on the set a= ...= a= 0 is called r-sliding (rth order sliding) mode[6, 1, IO]. The standard sliding mode used in the most variable structure systems (VSS), is of the fust order (a is continuous, and b is discontinuous). Consider a dynamic system of the form

X

= a(t,x)+ b(t,x)u,

a= o(t, x),

(2)

0-7803-7924-1/03/$17.00 62003 IEEE

4605

Quasi-continuous high-order sliding mode controllers.

第1页

你可能喜欢

热门文档

相关文档

站点地图 | 文档上传 | 侵权投诉 | 手机版
本站所有资源均来自互联网,本站只负责收集和整理,均不承担任何法律责任,如有侵权等其它行为请联系我们.
文档下载 Copyright 2013 doc.guandang.net All Rights Reserved.  -  email