Search

This article addresses the problem of distributed Nash equilibrium (NE) seeking in strongly contractive aggregative multipopulation games subject to partial-decision information. In particular, we consider the scenario where the so-called payoff providers of the multiple populations communicate through a possibly noncomplete network, and we formulate some consensuslike dynamics for the payoff providers to distributedly compute their payoff signals using local information only. Moreover, by exploiting the notions of $\delta$-passivity and $\delta$-antipassivity, we provide a unified analysis for several classes of evolutionary game dynamics. As the main contributions, we provide sufficient conditions to guarantee the $\delta$-antipassivity of a class of continuous-time dynamical systems, and we exploit such results to design distributed NE seeking dynamics for strongly contractive aggregative population games, as well as for a class of merely contractive aggregative population games. To the best of our knowledge, this is the first article to consider the problem of distributed NE seeking for such classes of population games and from a unifying passivity-based perspective.

In agriculture, most of the water is transported by networks of open channel irrigation systems (OCIS). To ensure an optimal operation of this type of systems, it is very important to have accurate monitoring of all the system state variables. In order to perform robust estimation mitigating the effect of unknown user demands, an unknown input observer for one sided Lipschitz (OSL) quadratically inner bounded (QIB) systems is proposed. The proposed UIO scheme is assessed in a case study based on the Corning channel in California that is modelled uisng material and energy balances. The modeling methodology includes inflows and outflows of matter, along with a transition flow which models the effect of potential energy (channel slope) and kinetic energy (velocity in the transport of matter and frictional losses). The objective of this paper is to show that with the UIO for QIB-OSL systems, costs could be reduced in the purchase of sensors, since the observer is capable of reconstructing the upstream heights from the measurements of the downstream heights, which will be subject to noise and unknown inputs (in this case, user demands).

While humans and animals learn incrementally during their lifetimes and exploit their experience to solve new tasks, standard deep reinforcement learning methods specialize to solve only one task at a time. As a result, the information they acquire is hardly reusable in new situations. Here, we introduce a new perspective on the problem of leveraging prior knowledge to solve future tasks. We show that learning discrete representations of sensory inputs can provide a high-level abstraction that is common across multiple tasks, thus facilitating the transference of information. In particular, we show that it is possible to learn such representations by self-supervision, following an information theoretic approach. Our method is able to learn abstractions in locomotive and optimal control tasks that increase the sample efficiency in both known and unknown tasks, opening a new path to endow artificial agents with generalization abilities.

In open-channel irrigation systems (OCIS), most of the water is lost by seepage and leaks. However, the reported control strategies for OCIS are not focused on leaks and seepage minimization. Therefore, aiming to contribute to the reduction of waste of water, in this paper, an efficient control approach for OCIS, capable to reduce the waste of water by up to 50%, is presented. The proposed strategy is a nonlinear model predictive control that operates under request mode, which has been designed using a simplified modeling approach capable to describe the behavior of the OCIS along a broad operation region. The proposed strategy has been tested using a well-known test bed of the literature, and the results show that, by using this strategy, an appropriate amount of water can be supplied to the users, and the waste of water can be significantly reduced.

This paper proposes a coalitional model predictive control method for temperature regulation in parabolic-trough solar fields. The global optimization problem is divided into a set of local subproblems that will be solved in parallel by a set of coalitions. However, these local (smaller) problems remain coupled by a common global resource constraint. In this regard, we present a population-dynamics-assisted resource allocation approach to fully decouple the local optimization problems. By doing this, each coalition can address its corresponding optimization problem without relying on the solutions of the other coalitions. To illustrate the proposed methodology, we provide simulation results for a 100-loop parabolic-trough solar collector field.

This paper studies the joint production and maintenance scheduling in microalgae manufacturing systems comprised of multiple machines, which are subject to coupled production demand agreements and operational maintenance constraints. Namely, there are some microalgae production demands to be met over a given horizon, and the maintenance of each microalgae manufacturing unit must be done before a given deadline. Moreover, the number of units whose maintenance can be done simultaneously over the same day is limited, and the units that undergo maintenance cannot contribute to microalgae production during their maintenance day. To solve the considered problem, we design a mixed-integer nonlinear model predictive controller, which is implemented in two optimization stages. The former regards a mixed-integer model predictive control problem, while the latter considers a nonlinear model predictive control problem. The proposed approach allows us to decouple the mixed-integer and nonlinear parts of the whole problem, and thus provides more flexibility on the optimization solvers that can be employed. In addition, the first stage also evaluates the attainability of the demand agreements, and provides a mechanism to minimally adjust such constraints so that their satisfaction can be guaranteed at the second stage. The overall model predictive control approach is based on experimental data collected at VAXA Technologies Ltd., and the effectiveness of the proposed method is validated through numerical simulations including multiple manufacturing units and uncertainties.

As an envisioned technology for future smart city networks, this article studies the real-time decentralized charging coordination of a fleet of plug-in electric vehicles (PEVs) under feeder capacity constraints. In particular, inspired by some ideas in the field of population games and payoff dynamics models, we propose a novel form of continuous-time primal-dual gradient dynamics and develop a real-time control method for the charging coordination of PEVs in smart city networks. The proposed method is able to coordinate the charging profiles of multiple PEVs in a decentralized fashion under a general convex optimization objective, and guarantees the satisfaction of the operational constraints of the PEVs and the feeder lines of the distribution network for all times. The optimality and asymptotic stability of the proposed dynamics are formally proven, and the advantages of the proposed method are illustrated through numerical simulations considering a fleet with several PEVs.

This paper studies the distributed formation control of multiple differential-drive robots. To solve such problem, we propose a novel class of distributed population dynamics, formulated in discrete-time, and we obtain sufficient conditions to guarantee asymptotic stability in practical implementations where computations are necessarily discrete. Moreover, we apply the proposed dynamics to a real multi-robot platform where robots achieve geometric formations under partial information and limited communication capabilities. Our proposed method achieves comparable and even better performance than other distributed methods, and displays some invariance properties that make it attractive for several other engineering applications.

In irrigation, most of the water is transported by networks of open-channel irrigation systems (OCIS). In most cases, the OCIS are manually operated showing low efficiency. Then the incorporation of control strategies is one of the most practical ways to increase the efficiency of these systems. However, in order to design an appropriate control strategy, an accurate control-oriented model that can be used for analyses, simulation, design, and test of the OCIS under realistic scenarios is necessary. In OCIS, obtaining a control-oriented model is a challenge that has aroused interest in the related literature. In spite of the multiple research in modeling of OCIS, the development of control-oriented modeling approaches that describe the dynamic and nonlinear behaviors of OCIS with gate regulation structures, remains an open problem. In this paper, a modeling approach that describes the nonlinear and dynamical behaviors of OCIS using a mass and energy balance by channel is proposed, which is compared with two modeling approaches. The comparison has been performed with a test case proposed in the literature. The results show that the proposed modeling approach is better describing the nonlinear behavior of the OCIS and presents a simpler structure that can be used in the design and test of control, prediction, and estimation strategies for these type of systems.

A hierarchical approach for the energy management of geographically close microgrids connected through a dedicated AC power network is proposed in this paper. The proposed approach consists of a two-layer energy management system (EMS) for networked microgrids. In the lower layer, each microgrid solves its own economic dispatch problem through a distributed model predictive control approach that respects capacity limits and ramp-rate constraints of distributed generation. In the upper layer, the energy trading in the network of microgrids decides how to optimally trade the energy based on the marginal cost information from the lower layer in order to improve global optimization objectives, e.g., social welfare. In order to solve the trading problem, a consensus-based algorithm and a replicator dynamics algorithm are proposed assuming that the marginal cost function of the microgrid is known and linear. It is shown that both algorithms converge to the same solution, which is equivalent to the minimization of operation costs. The consensus-based algorithm is extended in order to tackle more general marginal cost functions and trading network constraints. Moreover, the effect of ramp constraints and network limits is studied. Simulations show the effectiveness of the proposed algorithms for three interconnected microgrids with different characteristics.

A weakness of most distributed resource allocation algorithms proposed in the literature is that they assume that all nodes of the underlying graph are involved in the resource allocation problem. Such assumption does not hold in some applications. This manuscript deals with the problem of resource allocation among a subset of nodes of a graph. We propose a fully distributed algorithm based on consensus that provably reaches the optimal solution of a class of resource allocation problem that only involves certain nodes of a graph. Usefulness of the proposed algorithm is illustrated by means of two applications: distributed computation of a minimum-covariance estimate in sensor networks under failures, and distributed node counting.

By combining different control methods, we can obtain a desired performance in a control system. In this work, we introduce a novel technique that uses a controller that is based on an energy system for the inverted pendulum, combined dynamically with an optimal controller. The dynamic combination of these two controllers is based on a population dynamics approach, which gives more weight to the controller that is designed to perform better in a given region. The controller is tested on a classical experiment used to illustrate decentralized controllers: the two inverted pendulums coupled by a spring (TPCS) system. Some simulations comparing our novel technique and the one that has been used are shown.

This paper solves a data-driven control problem for a flow-based distribution network with two objectives: a resource allocation and a fair distribution of costs. These objectives represent both cooperation and competition directions. It is proposed a solution that combines either a centralized or distributed cooperative game approach using the Shapley value to determine a proper partitioning of the system and a fair communication cost distribution. On the other hand, a decentralized non-cooperative game approach computing the Nash equilibrium is used to achieve the control objective of the resource allocation under a non-complete information topology. Furthermore, an invariant-set property is presented and the closed-loop system stability is analyzed for the non-cooperative game approach. Another contribution regarding the cooperative game approach is an alternative way to compute the Shapley value for the proposed specific characteristic function. Unlike the classical cooperative-games approach, which has a limited application due to the combinatorial explosion issues, the alternative method allows calculating the Shapley value in polynomial time and hence can be applied to large-scale problems. [ABSTRACT FROM AUTHOR]