Abstract
A distributed linear quadratic decision problem is considered, where several different controllers act as a team, but with access to different measurements. Previous contributions have shown how to state the optimal synthesis as a finite-dimensional convex optimization problem. This paper shows that the dynamic behavior can be optimized by a distributed iterative procedure, without any need for a globally available model or centralized coordination. An illustrative model with three agents is considered.

This publication has 17 references indexed in Scilit: