Distributed stochastic model predictive control synthesis for large-scale uncertain linear systems

Abstract

This paper presents an approach to distributed stochastic model predictive control (SMPC) of large-scale uncertain linear systems with additive disturbances. Typical SMPC approaches for such problems involve formulating a large-scale finite-horizon chance-constrained optimization problem at each sampling time, which is in general non-convex and difficult to solve. Using an approximation, the so-called scenario approach, we formulate a large-scale scenario program and provide a theoretical guarantee to quantify the robustness of the obtained solution. However, such a reformulation leads to a computational tractability issue, due to the large number of required scenarios. To this end, we present two novel ideas in this paper to address this issue. We first provide a technique to decompose the large-scale scenario program into distributed scenario programs that exchange a certain number of scenarios with each other in order to compute local decisions. We show the exactness of the decomposition with a-priori probabilistic guarantees for the desired level of constraint fulfillment. As our second contribution, we develop an inter-agent soft communication scheme based on a set parametrization technique together with the notion of probabilistically reliable sets to reduce the required communication between each subproblem. We show how to incorporate the probabilistic reliability notion into existing results and provide new guarantees for the desired level of constraint violations. A simulation study is presented to illustrate the advantages of our proposed framework.