A convex characterization of gain-scheduled H/sub /spl infin// controllers
Citations
1,621 citations
Cites background from "A convex characterization of gain-s..."
...We should note that LPV approaches o!er another way to interpolate controllers (Packard, 1994; Apkarian & Gahinet, 1995 )....
[...]
1,439 citations
Cites methods from "A convex characterization of gain-s..."
...A first technique for parameter-dependent controller synthesis is based on the small gain theorem and applicable to LPV plants with an LIT (linear fractional transformation) dependence on the parameter 0 (Packard, 1994; Apkarian and Gahinet, 1995)....
[...]
...Once adequate matrices R and S have been computed, the Lyapunov matrix Xce common to all inequalities (29) and the vertex controllers Ri are obtained along the lines of Gahinet and Apkarian (1994) and Apkarian and Gahinet (1995)....
[...]
1,099 citations
Cites background or methods from "A convex characterization of gain-s..."
...Since in (4) is available online, system (4) also belongs to the more general class of gain-scheduling control systems intensively studied in control theory in the past decade (see, e.g., [12], [ 1 ], and [2])....
[...]
...Only recently, however, this technique has received a systematic treatment within the framework and tools based on LMIs [12], [ 1 ], [2]....
[...]
...our presented results based on a general theory of gain-scheduling control [ 1 ], [2], [12] have the following essential advantages over those of [14] and [15]: 1) The resulting optimization formulations are much simpler with much fewer variables involved....
[...]
933 citations
887 citations
Cites methods or result from "A convex characterization of gain-s..."
...In addition, these controllers have an LFT representation in terms of the nonlinear functions, i(:), and hence are computationally comparable to those of the LFT gain-scheduling approaches in [7, 8]....
[...]
...Therefore, if the structure of the problem is ignored, the techniques in this paper or [7, 8] may o er no advantages over existing LFT gain-scheduling techniques....
[...]
...Basic Technique with X( 2) and Y ( 2) 3:82 Projected Technique with X( 2) and Y ( 2) 3:82 LFT Technique in [8] 3:82 Table 2: Performance comparisons ( ) with ignored structure....
[...]
...The corresponding levels are compared with the LFT gain-scheduling technique in [8], a technique that puts no bound on the parameter variation rates....
[...]
...The Linear Fractional Transformation (LFT) gain-scheduling techniques in [7, 8, 9, 10] or the so-called quadratic gain-scheduled techniques in [11, 12] make use of a xed Lyapunov function, as opposed to one which depends on the scheduled variables, to characterize stability and performance....
[...]
References
3,200 citations
1,439 citations
1,309 citations
"A convex characterization of gain-s..." refers methods in this paper
...To apprehend this problem with Small Gain Theory,we must rst gather all parameter-dependent components into a single uncertainty block....
[...]
...The synthesis of gain-scheduledH1 controllers relies on the Small Gain Theorem [35, 9]....
[...]
...Su cient conditions for solvability are then provided by Small Gain Theory [35, 9]....
[...]
...If there exists a scaling matrix L 2 L and an LTI controlstructure K( ) such that the nominal closed-loop system Fl(Pa( );K(( )) is internally stableand satis es L1=2 00 I Fl(Pa( ); K( )) L 1=2 00 I 1 < ; (2:15)then Fl(K( ); ) is a -suboptimal gain-scheduled H1 controller.7 Proof: The proof is a straightforward application of the Small Gain Theorem....
[...]
...Speci cally, consider the set of positive de nite similarity scalingsassociated with the structure in (2.3):L = fL > 0 : L = L; 8 2 g Rr r with r = KXi=1 ri: (2:13)This set enjoys the following immediate properties:(P1) Ir 2 L (P2) L 2 L ) LT 2 L (P3) L 2 L ) L 1 2 L (P4) L1 2 L ; L2 2 L ) L1L2 = L1L2; 8 2 (P5) L is a convex subset of Rr r.Given L , the set of scalings commuting with the repeated structure is readily deducedas: L = L1 L2LT2 L3 > 0 : L1; L3 2 L and L2 = L2;8 2 : (2:14)From Small Gain Theory, a su cient condition for robust performance in the face of theuncertainty , or equivalently for the existence of gain-scheduled controllers, is asfollows....
[...]
1,218 citations