In this paper, we address finite-horizon control for a stochastic linear system subject to constraints on the control and state variables. A control design methodology is proposed where the appropriate trade-off between the minimization of the control cost (performance) and the satisfaction of the state constraints (safety) can be decided by introducing appropriate chance-constrained problems depending on some parameter to be tuned. From an algorithmic viewpoint, a computationally tractable randomized approach to find approximate solutions which are guaranteed to be feasible for the original chanceconstrained problem is proposed. A numerical example concludes the paper.

Abstract:We consider the problem of optimal charging of heterogeneous plug-in electric vehicles (PEVs). We approach the problem as a multi-agent game in the presence of constraints and formulate an auxiliary minimization program whose solution is shown to be the unique Nash equilibrium of the PEV charging control game, for any finite number of possibly heterogeneous agents. Assuming that the parameters defining the constraints of each vehicle are drawn randomly from a given distribution, we show that, as the number of agents tends to infinity, the value of the game achieved by the Nash equilibrium and the social optimum of the cooperative counterpart of the problem under study coincide for almost any choice of the random heterogeneity parameters. To the best of our knowledge, this result quantifies for the first time the asymptotic behaviour of the price of anarchy for this class of games. A numerical investigation to support our result is also provided.

We consider multi-agent, convex optimization programs subject to separable constraints, where the constraint function of each agent involves only its local decision vector, while the decision vectors of all agents are coupled via a common objective function. We focus on a regularized variant of the so called Jacobi algorithm for decentralized computation in such problems. We first consider the case where the objective function is quadratic, and provide a fixed-point theoretic analysis showing that the algorithm converges to a minimizer of the centralized problem. Moreover, we quantify the potential benefits of such an iterative scheme by comparing it against a scaled projected gradient algorithm. We then consider the general case and show that all limit points of the proposed iteration are optimal solutions of the centralized problem. The efficacy of the proposed algorithm is illustrated by applying it to the problem of optimal charging of electric vehicles, where, as opposed to earlier approaches, we show convergence to an optimal charging scheme for a finite, possibly large, number of vehicles.

We consider the problem of optimal charging of plug-in electric vehicles (PEVs). We treat this problem as a multi-agent game, where vehicles/agents are heterogeneous since they are subject to possibly different constraints. Under the assumption that electricity price is affine in total demand, we show that, for any finite number of heterogeneous agents, the PEV charging control game admits a unique Nash equilibrium, which is the optimizer of an auxiliary minimization program. We are also able to quantify the asymptotic behaviour of the price of anarchy for this class of games. More precisely, we prove that if the parameters defining the constraints of each vehicle are drawn randomly from a given distribution, then, the value of the game converges almost surely to the optimum of the cooperative problem counterpart as the number of agents tends to infinity. In the case of a discrete probability distribution, we provide a systematic way to abstract agents in homogeneous groups and show that, as the number of agents tends to infinity, the value of the game tends to a deterministic quantity.

Constrained control for stochastic linear systems is generally a difficult task due to the possible infeasibility of state constraints. In this paper, we focus on a finite control horizon and propose a design methodology where the constrained control problem is formulated as a chance-constrained optimization problem depending on some parameter. This parameter can be tuned so as to decide the appropriate trade-off between control cost minimization and state constraints satisfaction. An approximate solution is computed via a randomized algorithm. Precise guarantees about its feasibility for the original chance-constrained problem are provided. A numerical example shows the efficacy of the proposed methodology