Decentralized control information structures preserved under feedback

Abstract

We consider the problem of constructing decentralized control systems. We formulate this problem as one of minimizing the closed-loop norm of a feedback system subject to constraints on the controller structure. We define the notion of quadratic invariance of a constraint set with respect to a system, and show that if the constraint set has this property, then the constrained minimum norm problem may be solved via convex programming. We also show that quadratic invariance is necessary and sufficient for the constraint set to be preserved under feedback.We develop necessary and sufficient conditions under which the constraint set is quadratically invariant, and show that many examples of decentralized synthesis which have been proven to be solvable in the literature are quadratically invariant. As an example, we show that a controller which minimizes the norm of the closed-loop map may be efficiently computed in the case where distributed controllers can communicate faster than the propagation delay of the plant dynamics. University, Stanford CA 94305-4035, U.S.A. The first author was partially supported by a Stanford Graduate Fellowship. Both authors were partially supported by the Stanford URI Architectures f . 7 Secure and Robust DIstnbuted Infmstructures, AFOSR DoD award number 49620-01-1-0365. f ( P , K ) = P I I + S z K ( I -P z z K ) -' P z I 0-7803-7516-5/02/$17.00 a2002 IEEE

A convex characterization of distributed control problems in spatially invariant systems with communication constraints

“…The same structure of controllers as the plant is imposed and a relaxation is used to obtain stability and performance conditions via LMIs. More recently, an interesting algebraic characterization of plant-controller structures that leads to convex maps has been provided in [6]. The work in [6] nicely generalizes the classes of convex structural problems presented earlier in [7].…”

Convex risk averse control design

“…In fact, it has been shown that the optimal control policy may not even be linear [Wit68] and that the general problem of designing linear feedback gains subject to constraints is NPhard [BT97]. Previous approaches to synthesizing structured controllers can be broadly categorized into three types: Frequency Domain [RL02] [BHLO06]. The first two classes of approaches find exact solutions to structured control problems for special cases.…”

A scalable method for continuous-time distributed control synthesis

“…Some special types of team problems were solved in the 1970's [15], [6], but the problem area has recently gain renewed interest. Spatial invariance was exploited in [1], [2], conditions for closed loop convexity were derived in [14], [13] and methods using linear matrix inequalities were given in [7], [12], [5].…”