Search NASA⌕ Search

SEARCH · Search NASA

Results for “Multi-agent control”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 37 records · Page 2
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Self-Aware Local Autonomous and Semi-Cooperative Control for Cross-Layered Resilience (SLAC3R)

The objective of this work is to develop and demonstrate novel, adaptive, lightweight algorithms that enable the decision-making agents in a large cyber-physical network to act both autonomously and in collaborative harmony to enforce assured resilience across spatiotemporal layers, even under unforeseen adversarial scenarios (e.g., high- impact-low-probability events). Towards this end, the proposed solution will serve as minimally invasive add-on layers that bridge the existing (faster, reactive) local myopic controls and (slower, predictive) centralized optimization. Importantly, the proposed algorithms will enable the multi-agent network to autonomously and collaboratively enforce resilient operation under no or limited communication environment typical of severe cyber- physical adversarial events. The expected outcome of this effort is a suite of prototype, open-source, software algorithms for safety-aware local autonomous and semi-cooperative control (SLAC3R), demonstrated on networked microgrids (via RD2C/Thrust-1 OPAL-RT testbed).

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Optimization-based approaches to control of connected and automated vehicles: Principles, complexities, applications, challenges, and outlook

Safe and optimal motion control for connected and automated vehicles (CAVs) poses a fundamental optimization challenge at the intersection of system complexity, environmental uncertainty, and stringent real-time constraints. Existing surveys address this challenge in isolation – focusing either on specific control techniques or individual uncertainty sources – without providing a unified framework that characterizes the trade-offs among computational tractability, performance verifiability, and adaptive generalization across paradigms. This review addresses that gap by presenting a cohesive analytical framework concentrated on the decision-making and trajectory optimization layers of the CAV autonomy stack. We systematically analyze three major optimization paradigms – first-principles model-based optimization, data-driven methods, and hybrid synergistic architectures – evaluating each against four core complexity axes: problem formulation, constraint handling, optimality guarantees, and robustness. Key applications including platooning, trajectory planning, collision avoidance, and cooperative control are examined to reveal recurring methodological patterns and critical operational constraints that limit real-world performance. Our synthesis identifies verifiable hybrid architectures, incentive-aligned multi-agent cooperation, and hardware-algorithm co-design as the defining research frontiers, and distills a targeted agenda for developing CAV control systems that are simultaneously safe, computationally efficient, and deployable in the full complexity of real-world traffic environments.

Muzahid, Abu Jafar Md [University of Tennessee, Kn↗
DOE OSTI2025

LC-Opt: Benchmarking Reinforcement Learning and Agentic AI for End-to-End Liquid Cooling Optimization in Data Centers

Liquid cooling is critical for thermal management in high-density data centers with the rising AI workloads. However, machine learning-based controllers are essential to unlock greater energy efficiency and reliability, promoting sustainability. We present LC-Opt, a Sustainable Liquid Cooling (LC) benchmark environment, for reinforcement learning (RL) control strategies in energy-efficient liquid cooling of high-performance computing (HPC) systems. Built on the baseline of a high-fidelity digital twin of Oak Ridge National Lab's Frontier Supercomputer cooling system, LC-Opt provides detailed Modelica-based end-to-end models spanning site-level cooling towers to data center cabinets and server blade groups. RL agents optimize critical thermal controls like liquid supply temperature, flow rate, and granular valve actuation at the IT cabinet level, as well as cooling tower (CT) setpoints through a Gymnasium interface, with dynamic changes in workloads. This environment creates a multi-objective real-time optimization challenge balancing local thermal regulation and global energy efficiency, and also supports additional components like a heat recovery unit (HRU). We benchmark centralized and decentralized multi-agent RL approaches, demonstrate policy distillation into decision and regression trees for interpretable control, and explore LLM-based methods that explain control actions in natural language through an agentic mesh architecture designed to foster user trust and simplify system management. LC-Opt democratizes access to detailed, customizable liquid cooling models, enabling the ML community, operators, and vendors to develop sustainable data center liquid cooling control solutions.

Naug, Avisek [Hewlett Packard Enterprise]↗
NASA NTRS2000

Optimal Wonderful Life Utility Functions in Multi-Agent Systems

The mathematics of Collective Intelligence (COINs) is concerned with the design of multi-agent systems so as to optimize an overall global utility function when those systems lack centralized communication and control. Typically in COINs each agent runs a distinct Reinforcement Learning (RL) algorithm, so that much of the design problem reduces to how best to initialize/update each agent's private utility function, as far as the ensuing value of the global utility is concerned. Traditional team game solutions to this problem assign to each agent the global utility as its private utility function. In previous work we used the COIN framework to derive the alternative Wonderful Life Utility (WLU), and experimentally established that having the agents use it induces global utility performance up to orders of magnitude superior to that induced by use of the team game utility. The WLU has a free parameter (the clamping parameter) which we simply set to zero in that previous work. Here we derive the optimal value of the clamping parameter, and demonstrate experimentally that using that optimal value can result in significantly improved performance over that of clamping to zero, over and above the improvement beyond traditional approaches.

Wolpert, David H.↗
NASA NTRS2000

Optimal Reward Functions in Distributed Reinforcement Learning

We consider the design of multi-agent systems so as to optimize an overall world utility function when (1) those systems lack centralized communication and control, and (2) each agents runs a distinct Reinforcement Learning (RL) algorithm. A crucial issue in such design problems is to initialize/update each agent's private utility function, so as to induce best possible world utility. Traditional 'team game' solutions to this problem sidestep this issue and simply assign to each agent the world utility as its private utility function. In previous work we used the 'Collective Intelligence' framework to derive a better choice of private utility functions, one that results in world utility performance up to orders of magnitude superior to that ensuing from use of the team game utility. In this paper we extend these results. We derive the general class of private utility functions that both are easy for the individual agents to learn and that, if learned well, result in high world utility. We demonstrate experimentally that using these new utility functions can result in significantly improved performance over that of our previously proposed utility, over and above that previous utility's superiority to the conventional team game utility.

Wolpert, David H.↗
DOE OSTI2025

Enabling Real-Time Communication in Multi-Agent Systems: A Graph Neural Network Based Approach

Global connectivity enables effective coordination in Multi-Agent Systems (MAS). Solving these connection problems under hardware constraints is an NP-hard non-Euclidean Degree Constrained Minimum Spanning Tree (DCMST) problem. Prior MAS controllers coordinate team movement for task completion and collision avoidance; some considering Line-of-Sight (LOS) maintenance but prioritizing flexibility over guarantees. Evolutionary Algorithms (EA) have been shown to find good solutions for DCMST, but their performance degrades with larger populations required to support a large MAS. We present a method based on edge graph attention networks, trained offline to reduce online computation times. Empirical comparisons with greedy polynomial-time solvers and EA show that our method leverages latent graph information to consistently find constraint-satisfying solutions in less time.

connectivity maintenance↗
NASA NTRS1999

A Cognitive Game Theoretic Analysis of Conflict Alerts in Air Traffic Control

The current research was motivated by the recommendation made by a joint Government/Industry committee to introduce a new traffic control system, referred to as the Free Flight. This system is designed to use recent new technology to facilitate efficient and safe air transportation. We addressed one of the major difficulties that arise in the design of this and similar multi-agent systems: the adaptive (and slippery) nature of human agents. To facilitate a safe and efficient design of this multi-agent system, designers have to rely on assessments of the expected behavior of the different agents under various scenarios. Whereas the behavior of the computerized agents is predictable, the behavior of the human agents (including air traffic controllers and pilots) is not. Experimental and empirical observations suggest that human agents are likely to adjust their behavior to the design of the system. To see the difficulty that the adaptive nature of human agents creates assume that a good approximation of the way operators currently behave is available. Given this information an optimal design can be performed. The problem arises as the human operator will learn to adjust their behavior to the new system. Following this adjustment process the assumptions made by the designer concerning the operators behavior will no longer be accurate and the system might reach a suboptimal state. In extreme situations these potential suboptimal states might involve unnecessary risk. That is, the fact that operators learn in an adaptive fashion does not imply that the system will become safer as they gain experience. At least in the context of Safety dilemmas, experience can lead to a pareto deficient risk taking behavior.

Erev, Ido↗
NASA NTRS1994

Multi-Agent Flight Simulation with Robust Situation Generation

A robust situation generation architecture has been developed that generates multi-agent situations for human subjects. An implementation of this architecture was developed to support flight simulation tests of air transport cockpit systems. This system maneuvers pseudo-aircraft relative to the human subject's aircraft, generating specific situations for the subject to respond to. These pseudo-aircraft maneuver within reasonable performance constraints, interact in a realistic manner, and make pre-recorded voice radio communications. Use of this system minimizes the need for human experimenters to control the pseudo-agents and provides consistent interactions between the subject and the pseudo-agents. The achieved robustness of this system to typical variations in the subject's flight path was explored. It was found to successfully generate specific situations within the performance limitations of the subject-aircraft, pseudo-aircraft, and the script used.

Johnson, Eric N.↗
DOE OSTI2026

Optimal CO 2 storage management considering safety constraints in multi-stakeholder multi-site GCS projects: A Markov game perspective

Geological carbon storage (GCS) projects could involve a diverse array of stakeholders or players from public, private, and regulatory sectors, each with different objectives and responsibilities. Given the complexity, scale, and long-term nature of GCS operations, determining whether individual stakeholders can independently optimize their interests — or whether collaborative coalition agreements are needed — remains a central question for effective GCS project planning and management. To access large, high-quality storage resources, future GCS deployment may increasingly occur in geologically connected sites, where shared geological features such as pressure space and reservoir pore capacity can lead to competitive behavior among stakeholders. In this work, we propose a paradigm based on Markov games to quantitatively investigate how different coalition structures affect the goals of stakeholders. We frame this multi-stakeholder multi-site problem as a multi-agent reinforcement learning problem with safety constraints. Our approach enables agents to learn optimal strategies while complying with safety regulations. We present an example where multiple operators are injecting CO 2 into their respective project areas in a geologically connected basin. To address the high computational cost of repeated simulations of high fidelity models, a previously developed surrogate model based on the Embed-to-Control (E2C) framework is employed. Our results demonstrate the effectiveness of the proposed framework in addressing optimal management of CO 2 storage when multiple stakeholders with different objectives and goals are involved.

58 GEOSCIENCES↗
NASA NTRS1994

Space/ground systems as cooperating agents

Within NASA and the European Space Agency (ESA) it is agreed that autonomy is an important goal for the design of future spacecraft and that this requires on-board artificial intelligence. NASA emphasizes deep space and planetary rover missions, while ESA considers on-board autonomy as an enabling technology for missions that must cope with imperfect communications. ESA's attention is on the space/ground system. A major issue is the optimal distribution of intelligent functions within the space/ground system. This paper describes the multi-agent architecture for space/ground systems (MAASGS) which would enable this issue to be investigated. A MAASGS agent may model a complete spacecraft, a spacecraft subsystem or payload, a ground segment, a spacecraft control system, a human operator, or an environment. The MAASGS architecture has evolved through a series of prototypes. The paper recommends that the MAASGS architecture should be implemented in the operational Dutch Utilization Center.

Grant, T. J.↗
NASA NTRS2000

Adaptivity in Agent-Based Routing for Data Networks

Adaptivity, both of the individual agents and of the interaction structure among the agents, seems indispensable for scaling up multi-agent systems (MAS s) in noisy environments. One important consideration in designing adaptive agents is choosing their action spaces to be as amenable as possible to machine learning techniques, especially to reinforcement learning (RL) techniques. One important way to have the interaction structure connecting agents itself be adaptive is to have the intentions and/or actions of the agents be in the input spaces of the other agents, much as in Stackelberg games. We consider both kinds of adaptivity in the design of a MAS to control network packet routing. We demonstrate on the OPNET event-driven network simulator the perhaps surprising fact that simply changing the action space of the agents to be better suited to RL can result in very large improvements in their potential performance: at their best settings, our learning-amenable router agents achieve throughputs up to three and one half times better than that of the standard Bellman-Ford routing algorithm, even when the Bellman-Ford protocol traffic is maintained. We then demonstrate that much of that potential improvement can be realized by having the agents learn their settings when the agent interaction structure is itself adaptive.

Wolpert, David H.↗
NASA NTRS1999

A Survey of Collective Intelligence

This chapter presents the science of "COllective INtelligence" (COIN). A COIN is a large multi-agent systems where: i) the agents each run reinforcement learning (RL) algorithms; ii) there is little to no centralized communication or control; iii) there is a provided world utility function that, rates the possible histories of tile full system. Tile conventional approach to designing large distributed systems to optimize a world utility does not use agents running RL algorithms. Rather that approach begins with explicit modeling of the overall system's dynamics, followed by detailed hand-tuning of the interactions between the components to ensure that they "cooperate" as far as the world utility is concerned. This approach is labor-intensive, often results in highly non-robust systems, and usually results in design techniques that, have limited applicability. In contrast, with COINs we wish to solve the system design problems implicitly, via the 'adaptive' character of the RL algorithms of each of the agents. This COIN approach introduces an entirely new, profound design problem: Assuming the RL algorithms are able to achieve high rewards, what reward functions for the individual agents will, when pursued by those agents, result in high world utility? In other words, what reward functions will best ensure that we do not have phenomena like the tragedy of the commons, or Braess's paradox? Although still very young, the science of COINs has already resulted in successes in artificial domains, in particular in packet-routing, the leader-follower problem, and in variants of Arthur's "El Farol bar problem". It is expected that as it matures not only will COIN science expand greatly the range of tasks addressable by human engineers, but it will also provide much insight into already established scientific fields, such as economics, game theory, or population biology.

Wolpert, David H.↗
NASA NTRS2001

Learning Sequences of Actions in Collectives of Autonomous Agents

In this paper we focus on the problem of designing a collective of autonomous agents that individually learn sequences of actions such that the resultant sequence of joint actions achieves a predetermined global objective. We are particularly interested in instances of this problem where centralized control is either impossible or impractical. For single agent systems in similar domains, machine learning methods (e.g., reinforcement learners) have been successfully used. However, applying such solutions directly to multi-agent systems often proves problematic, as agents may work at cross-purposes, or have difficulty in evaluating their contribution to achievement of the global objective, or both. Accordingly, the crucial design step in multiagent systems centers on determining the private objectives of each agent so that as the agents strive for those objectives, the system reaches a good global solution. In this work we consider a version of this problem involving multiple autonomous agents in a grid world. We use concepts from collective intelligence to design goals for the agents that are 'aligned' with the global goal, and are 'learnable' in that agents can readily see how their behavior affects their utility. We show that reinforcement learning agents using those goals outperform both 'natural' extensions of single agent algorithms and global reinforcement, learning solutions based on 'team games'.

Turner, Kagan↗
NASA NTRS2003

Collectives for Multiple Resource Job Scheduling Across Heterogeneous Servers

Efficient management of large-scale, distributed data storage and processing systems is a major challenge for many computational applications. Many of these systems are characterized by multi-resource tasks processed across a heterogeneous network. Conventional approaches, such as load balancing, work well for centralized, single resource problems, but breakdown in the more general case. In addition, most approaches are often based on heuristics which do not directly attempt to optimize the world utility. In this paper, we propose an agent based control system using the theory of collectives. We configure the servers of our network with agents who make local job scheduling decisions. These decisions are based on local goals which are constructed to be aligned with the objective of optimizing the overall efficiency of the system. We demonstrate that multi-agent systems in which all the agents attempt to optimize the same global utility function (team game) only marginally outperform conventional load balancing. On the other hand, agents configured using collectives outperform both team games and load balancing (by up to four times for the latter), despite their distributed nature and their limited access to information.

Tumer, K.↗
NASA NTRS2001

Design and Control of Large Collections of Learning Agents

The intelligent control of multiple autonomous agents is an important yet difficult task. Previous methods used to address this problem have proved to be either too brittle, too hard to use, or not scalable to large systems. The 'Collective Intelligence' project at NASA/Ames provides an elegant, machine-learning approach to address these problems. This approach mathematically defines some essential properties that a reward system should have to promote coordinated behavior among reinforcement learners. This work has focused on creating additional key properties and algorithms within the mathematics of the Collective Intelligence framework. One of the additions will allow agents to learn more quickly, in a more coordinated manner. The other will let agents learn with less knowledge of their environment. These additions will allow the framework to be applied more easily, to a much larger domain of multi-agent problems.

Agogino, Adrian↗
NASA NTRS2000

Autonomous Agents and Intelligent Assistants for Exploration Operations

Human exploration of space will involve remote autonomous crew and systems in long missions. Data to earth will be delayed and limited. Earth control centers will not receive continuous real-time telemetry data, and there will be communication round trips of up to one hour. There will be reduced human monitoring on the planet and earth. When crews are present on the planet, they will be occupied with other activities, and system management will be a low priority task. Earth control centers will use multi-tasking "night shift" and on-call specialists. A new project at Johnson Space Center is developing software to support teamwork between distributed human and software agents in future interplanetary work environments. The Engineering and Mission Operations Directorates at Johnson Space Center (JSC) are combining laboratories and expertise to carry out this project, by establishing a testbed for hWl1an centered design, development and evaluation of intelligent autonomous and assistant systems. Intelligent autonomous systems for managing systems on planetary bases will commuicate their knowledge to support distributed multi-agent mixed-initiative operations. Intelligent assistant agents will respond to events by developing briefings and responses according to instructions from human agents on earth and in space.

Malin, Jane T.↗