Projection-Based Inter-Agent Collision Avoidance in Dual Agent Systems
No abstract provided
Bayard, David S.↗
SEARCH · Engineering Papers
Results for “Agents”
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.
Petri Nets as Modeling Tool for Emergent Agents
Emergent agents, those agents whose local interactions can cause unexpected global results, require a method of modeling that is both dynamic and structured Petri Nets, a modeling tool developed for dynamic discrete event system of mainly functional agents, provide this, and have the benefit of being an established tool. We present here the details of the modeling method here and discuss how to implement its use for modeling agent-based systems. Petri Nets have been used extensively in the modeling of functional agents, those agents who have defined purposes and whose actions should result in a know outcome. However, emergent agents, those agents who have a defined structure but whose interaction causes outcomes that are unpredictable, have not yet found a modeling style that suits them. A problem with formally modeling emergent agents that any formal modeling style usually expects to show the results of a problem and the results of problems studied using emergent agents are not apparent from the initial construction. However, the study of emergent agents still requires a method to analyze the agents themselves, and have sensible conversation about the differences and similarities between types of emergent agents. We attempt to correct this problem by applying Petri Nets to the characterization of emergent agents. In doing so, the emergent properties of these agents can be highlighted, and conversation about the nature and compatibility of the differing methods of agent creation can begin.
Bergman, Marto↗
The role of information structures in game-theoretic multi-agent learning
Multi-agent learning (MAL) studies how agents learn to behave optimally and adaptively from their experience when interacting with other agents in dynamic environments. The outcome of a MAL process is jointly determined by all agents’ decision-making. Hence, each agent needs to think strategically about others’ sequential moves, when planning future actions. The strategic interactions among agents makes MAL go beyond the direct extension of single-agent learning to multiple agents. With the strategic thinking, each agent aims to build a subjective model of others decision-making using its observations. Such modeling is directly influenced by agents’ perception during the learning process, which is called the information structure of the agent’s learning. As it determines the input to MAL processes, information structures play a significant role in the learning mechanisms of the agents. This review creates a taxonomy of MAL and establishes a unified and systematic way to understand MAL from the perspective of information structures. Here we define three fundamental components of MAL: the information structure (i.e., what the agent can observe), the belief generation (i.e., how the agent forms a belief about others based on the observations), as well as the policy generation (i.e., how the agent generates its policy based on its belief). In addition, this taxonomy enables the classification of a wide range of state-of-the-art algorithms into four categories based on the belief-generation mechanisms of the opponents, including stationary, conjectured, calibrated, and sophisticated opponents. We introduce Value of Information (VoI) as a metric to quantify the impact of different information structures on MAL. Finally, we discuss the strengths and limitations of algorithms from different categories and point to promising avenues of future research.
97 MATHEMATICS AND COMPUTING↗
Application of multi-criteria decision analysis techniques and decision support framework for informing select agent designation for agricultural animal pathogens
The United States Department of Agriculture (USDA), Division of Agricultural Select Agents and Toxins (DASAT) established a list of biological agents and toxins (Select Agent List) that potentially threaten agricultural health and safety, the procedures governing the transfer of those agents, and training requirements for entities working with them. Every 2 years the USDA DASAT reviews the Select Agent List, using subject matter experts (SMEs) to perform an assessment and rank the agents. To assist the USDA DASAT biennial review process, we explored the applicability of multi-criteria decision analysis (MCDA) techniques and a Decision Support Framework (DSF) in a logic tree format to identify pathogens for consideration as select agents, applying the approach broadly to include non-select agents to evaluate its robustness and generality. We conducted a literature review of 41 pathogens against 21 criteria for assessing agricultural threat, economic impact, and bioterrorism risk and documented the findings to support this assessment. The most prominent data gaps were those for aerosol stability and animal infectious dose by inhalation and ingestion routes. Technical review of published data and associated scoring recommendations by pathogen-specific SMEs was found to be critical for accuracy, particularly for pathogens with very few known cases, or where proxy data (e.g., from animal models or similar organisms) were used to address data gaps. The MCDA analysis supported the intuitive sense that select agents should rank high on the relative risk scale when considering agricultural health consequences of a bioterrorism attack. However, comparing select agents with non-select agents indicated that there was not a clean break in scores to suggest thresholds for designating select agents, requiring subject matter expertise collectively to establish which analytical results were in good agreement to support the intended purpose in designating select agents. The DSF utilized a logic tree approach to identify pathogens that are of sufficiently low concern that they can be ruled out from consideration as a select agent. In contrast to the MCDA approach, the DSF rules out a pathogen if it fails to meet even one criteria threshold. Both the MCDA and DSF approaches arrived at similar conclusions, suggesting the value of employing the two analytical approaches to add robustness for decision making.
60 APPLIED LIFE SCIENCES↗
Application of Multi-Criteria Decision Analysis Techniques for Informing Select Agent Designation and Decision Making
The Centers for Disease Control and Prevention (CDC) Select Agent Program establishes a list of biological agents and toxins that potentially threaten public health and safety, the procedures governing the possession, utilization, and transfer of those agents, and training requirements for entities working with them. Every 2 years the Program reviews the select agent list, utilizing subject matter expert (SME) assessments to rank the agents. In this study, we explore the applicability of multi-criteria decision analysis (MCDA) techniques and logic tree analysis to support the CDC Select Agent Program biennial review process, applying the approach broadly to include non-select agents to evaluate its generality. We conducted a literature search for over 70 pathogens against 15 criteria for assessing public health and bioterrorism risk and documented the findings for archiving. The most prominent data gaps were found for aerosol stability and human infectious dose by inhalation and ingestion routes. Technical review of published data and associated scoring recommendations by pathogen-specific SMEs was found to be critical for accuracy, particularly for pathogens with very few known cases, or where proxy data (e.g., from animal models or similar organisms) were used to address data gaps. Analysis of results obtained from a two-dimensional plot of weighted scores for difficulty of attack (i.e., exposure and production criteria) vs. consequences of an attack (i.e., consequence and mitigation criteria) provided greater fidelity for understanding agent placement compared to a 1-to-n ranking and was used to define a region in the upper right-hand quadrant for identifying pathogens for consideration as select agents. A sensitivity analysis varied the numerical weights attributed to various properties of the pathogens to identify potential quantitative (x and y) thresholds for classifying select agents. The results indicate while there is some clustering of agent scores to suggest thresholds, there are still pathogens that score close to any threshold, suggesting that thresholding “by eye” may not be sufficient. The sensitivity analysis indicates quantitative thresholds are plausible, and there is good agreement of the analytical results with select agent designations. A second analytical approach that applied the data using a logic tree format to rule out pathogens for consideration as select agents arrived at similar conclusions.
60 APPLIED LIFE SCIENCES↗
Discerning Deception: An Empirically-Driven Agent-Based Model of Expert Evaluation of Scientific Content
Both human subject experiments and computational, modeling and simulations have been used to study detection of deception. This work aims to combine these two methods by integrating empirically-derived information (from human subject experiments) into agent-based models to generate novel insights into the complex problems of detection of disinformation content. Computational experiments are used to simulate across multiple scenarios for evaluation and decision-making regarding the validity of potentially deceptive scientific documents. Factors influencing the human agent behaviors in the model were identified through a human subject experiment that was conducted to evaluate and characterize decision making related to disinformation discernment. Correlation and regression analyses were used to translate insights from the human subjects experiment to inform the parameterization of agent features and scenario development. Three scenarios were evaluated with the agent-based models to help evaluate the replicability of the simulations (validation analysis) and assess the influence of human agent and document features (sensitivity analyses). A replication of the human participant experiment demonstrated that the agent-based simulations compare favorably to empirical findings. The agent-based modeling was then used to conduct sensitivity analysis on the accuracy of deception detection as a function of document proportions and human agent features. Results indicate that precision values are adversely impacted when the proportion of deceptive documents is lower in the overall sample, whereas recall values are more sensitive to changes in human agent features. These findings indicate important nuances in accuracy evaluations that should be further considered (including consideration of potential alternate metrics) in future agent-based models of disinformation. Additional areas for future exploration include extension of simulations to consider other ways to align the agent-based model design with psychological theory and inclusion of agent-agent interactions, especially as it pertains to sharing of scientific information within an organizational context.
99 GENERAL AND MISCELLANEOUS↗
Multi-agent planning and scheduling, execution monitoring and incremental rescheduling: Application to motorway traffic
This article describes a planning method applicable to agents with great perception and decision-making capabilities and the ability to communicate with other agents. Each agent has a task to fulfill allowing for the actions of other agents in its vicinity. Certain simultaneous actions may cause conflicts because they require the same resource. The agent plans each of its actions and simultaneously transmits these to its neighbors. In a similar way, it receives plans from the other agents and must take account of these plans. The planning method allows us to build a distributed scheduling system. Here, these agents are robot vehicles on a highway communicating by radio. In this environment, conflicts between agents concern the allocation of space in time and are connected with the inertia of the vehicles. Each vehicle made a temporal, spatial, and situated reasoning in order to drive without collision. The flexibility and reactivity of the method presented here allows the agent to generate its plan based on assumptions concerning the other agents and then check these assumptions progressively as plans are received from the other agents. A multi-agent execution monitoring of these plans can be done, using data generated during planning and the multi-agent decision-making algorithm described here. A selective backtrack allows us to perform incremental rescheduling.
Mourou, Pascal↗
A user-system interface agent
Agent-based technologies answer to several challenges posed by additional information processing requirements in today's computing environments. In particular, (1) users desire interaction with computing devices in a mode which is similar to that used between people, (2) the efficiency and successful completion of information processing tasks often require a high-level of expertise in complex and multiple domains, (3) information processing tasks often require handling of large volumes of data and, therefore, continuous and endless processing activities. The concept of an agent is an attempt to address these new challenges by introducing information processing environments in which (1) users can communicate with a system in a natural way, (2) an agent is a specialist and a self-learner and, therefore, it qualifies to be trusted to perform tasks independent of the human user, and (3) an agent is an entity that is continuously active performing tasks that are either delegated to it or self-imposed. The work described in this paper focuses on the development of an interface agent for users of a complex information processing environment (IPE). This activity is part of an on-going effort to build a model for developing agent-based information systems. Such systems will be highly applicable to environments which require a high degree of automation, such as, flight control operations and/or processing of large volumes of data in complex domains, such as the EOSDIS environment and other multidisciplinary, scientific data systems. The concept of an agent as an information processing entity is fully described with emphasis on characteristics of special interest to the User-System Interface Agent (USIA). Issues such as agent 'existence' and 'qualification' are discussed in this paper. Based on a definition of an agent and its main characteristics, we propose an architecture for the development of interface agents for users of an IPE that is agent-oriented and whose resources are likely to be distributed and heterogeneous in nature. The architecture of USIA is outlined in two main components: (1) the user interface which is concerned with issues as user dialog and interaction, user modeling, and adaptation to user profile, and (2) the system interface part which deals with identification of IPE capabilities, task understanding and feasibility assessment, and task delegation and coordination of assistant agents.
Wakim, Nagi T.↗
The Deontic Transaction Model in Multi-Agent Normative Systems
In the area of multi-agent systems, much research is devoted to the coordination of the agents. There exist several issues, two of which are summarized. The first is that, although agents are said to be autonomous, they always react in a predictable way to each message, and they cannot decide to violate the conventions that are hard-wired into the protocol. In fact, there might be circumstances in which the agent violates a convention in order to realize a private goal that it considers to be more important. Another issue is that, if the protocols that agents use to react to the environment are fixed, they have no way to respond to changes. However, an important characteristic of agents is that they eon react to a changing environment. Although transaction models ([BOH92], [GR93]) evolved from the database domain, they establish a general execution paradigm that ideally covers all the subsystems invoked in a sequence of transactions. So transaction models apply to multi-agent systems. Recently, some research has been devoted to overcoming the limitations of the traditional transaction models which are suitable for conventional systems and focus on system integrity, e,g., [SJ97]. Here we solve above issues by turning to a deontic concept: obligation. In multi-agent systems, agents interact with each other according to norms, We use deontic logic ([And58], [Aqv84], [B C96], [JS94], [MW93]) to model norms. Here the norms prescribe how the agents ought to behave, but-- and this is essential-- they do not exclude the possibility of "bad" behavior (i.e., the actual behavior may deviate from the ideal), and so they also prescribe what should be done in circumstances of norm violation. Thus, we propose a new approach --- a deontic transaction model for multi-agent normative systems. Our approach improves the protocol of "abort/commit" of traditional transaction models to a protocol of "abort/exception/commit". In multi-agent normative systems, we can see the violation of a norm as an exception and how the other agents react to the violation as exception handling. This paper concentrates on our formal model.
Huang, Yonghua↗
Agent Reward Shaping for Alleviating Traffic Congestion
Traffic congestion problems provide a unique environment to study how multi-agent systems promote desired system level behavior. What is particularly interesting in this class of problems is that no individual action is intrinsically "bad" for the system but that combinations of actions among agents lead to undesirable outcomes, As a consequence, agents need to learn how to coordinate their actions with those of other agents, rather than learn a particular set of "good" actions. This problem is ubiquitous in various traffic problems, including selecting departure times for commuters, routes for airlines, and paths for data routers. In this paper we present a multi-agent approach to two traffic problems, where far each driver, an agent selects the most suitable action using reinforcement learning. The agent rewards are based on concepts from collectives and aim to provide the agents with rewards that are both easy to learn and that if learned, lead to good system level behavior. In the first problem, we study how agents learn the best departure times of drivers in a daily commuting environment and how following those departure times alleviates congestion. In the second problem, we study how agents learn to select desirable routes to improve traffic flow and minimize delays for. all drivers.. In both sets of experiments,. agents using collective-based rewards produced near optimal performance (93-96% of optimal) whereas agents using system rewards (63-68%) barely outperformed random action selection (62-64%) and agents using local rewards (48-72%) performed worse than random in some instances.
Tumer, Kagan↗
Chemical Action of Halogenated Agents in Fire Extinguishing
The action of halogenated agents in preventing flame propagation in fuel-air mixtures in laboratory tests is discussed in terms of a possible chemical mechanism. The mechanism chosen is that of chain-breaking reactions between agent and active particles (hydrogen and oxygen atoms and hydroxyl radicsls). Data from the literature on the flammability peaks of n-heptane agent-air mixtures are treated. Ratings of agent effectiveness in terms of the fuel equivalent of the agent, based on both fuel and agent concentrations at the peak, are proposed as preferable to ratings in terms of agent concentration alone. These fuel-equivalent ratings are roughly correlated by reactivities assigned to halogen and hydrogen atoms in the agent molecules. It is concluded that the presence of hydrogen in agent need not reduce its fire-fighting ability, provided there is enough halogen to make the agent nonflammable. A method is presented for estimating from quenching-distance data a rate constant for the reaction of agent with active particles. A quantitative result is obtained for methyl bromide. This rate constant predicts the observed peak concentration of methyl bromide quite well. However, more data are needed to prove the validity of the method. The assumption that hal.ogenatedagents act mainly by chain-bresking reactions with active particles is consistent with the experimental facts and should help guide the selection of agents for further tests.
Belles, Frank E.↗
Formal Modeling of Multi-Agent Systems using the Pi-Calculus and Epistemic Logic
Multi-agent systems have become important recently in computer science, especially in artificial intelligence (AI). We allow a broad sense of agent, but require at least that an agent has some measure of autonomy and interacts with other agents via some kind of agent communication language. We are concerned in this paper with formal modeling of multi-agent systems, with emphasis on communication. We propose for this purpose to use the pi-calculus, an extension of the process algebra CCS. Although the literature on the pi-calculus refers to agents, the term is used there in the sense of a process in general. It is our contention, however, that viewing agents in the AI sense as agents in the pi-calculus sense affords significant formal insight. One formalism that has been applied to agents in the AI sense is epistemic logic, the logic of knowledge. The success of epistemic logic in computer science in general has come in large part from its ability to handle concepts of knowledge that apply to groups. We maintain that the pi-calculus affords a natural yet rigorous means by which groups that are significant to epistemic logic may be identified, encapsulated, structured into hierarchies, and restructured in a principled way. This paper is organized as follows: Section 2 introduces the pi-calculus; Section 3 takes a scenario from the classical paper on agent-oriented programming [Sh93] and translates it into a very simple subset of the n-calculus; Section 4 then shows how more sophisticated features of the pi-calculus may bc brought into play; Section 5 discusses how the pi-calculus may be used to define groups for epistemic logic; and Section 6 is the conclusion.
Rorie, Toinette↗
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.↗
Cooperation and Coordination Between Fuzzy Reinforcement Learning Agents in Continuous State Partially Observable Markov Decision Processes
Successful operations of future multi-agent intelligent systems require efficient cooperation schemes between agents sharing learning experiences. We consider a pseudo-realistic world in which one or more opportunities appear and disappear in random locations. Agents use fuzzy reinforcement learning to learn which opportunities are most worthy of pursuing based on their promise rewards, expected lifetimes, path lengths and expected path costs. We show that this world is partially observable because the history of an agent influences the distribution of its future states. We consider a cooperation mechanism in which agents share experience by using and-updating one joint behavior policy. We also implement a coordination mechanism for allocating opportunities to different agents in the same world. Our results demonstrate that K cooperative agents each learning in a separate world over N time steps outperform K independent agents each learning in a separate world over K*N time steps, with this result becoming more pronounced as the degree of partial observability in the environment increases. We also show that cooperation between agents learning in the same world decreases performance with respect to independent agents. Since cooperation reduces diversity between agents, we conclude that diversity is a key parameter in the trade off between maximizing utility from cooperation when diversity is low and maximizing utility from competitive coordination when diversity is high.
Berenji, Hamid R.↗
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↗
CATS-based Air Traffic Controller Agents
This report describes intelligent agents that function as air traffic controllers. Each agent controls traffic in a single sector in real time; agents controlling traffic in adjoining sectors can coordinate to manage an arrival flow across a given meter fix. The purpose of this research is threefold. First, it seeks to study the design of agents for controlling complex systems. In particular, it investigates agent planning and reactive control functionality in a dynamic environment in which a variety perceptual and decision making skills play a central role. It examines how heuristic rules can be applied to model planning and decision making skills, rather than attempting to apply optimization methods. Thus, the research attempts to develop intelligent agents that provide an approximation of human air traffic controller behavior that, while not based on an explicit cognitive model, does produce task performance consistent with the way human air traffic controllers operate. Second, this research sought to extend previous research on using the Crew Activity Tracking System (CATS) as the basis for intelligent agents. The agents use a high-level model of air traffic controller activities to structure the control task. To execute an activity in the CATS model, according to the current task context, the agents reference a 'skill library' and 'control rules' that in turn execute the pattern recognition, planning, and decision-making required to perform the activity. Applying the skills enables the agents to modify their representation of the current control situation (i.e., the 'flick' or 'picture'). The updated representation supports the next activity in a cycle of action that, taken as a whole, simulates air traffic controller behavior. A third, practical motivation for this research is to use intelligent agents to support evaluation of new air traffic control (ATC) methods to support new Air Traffic Management (ATM) concepts. Current approaches that use large, human-in-the-loop simulations are unquestionably valuable for this purpose, but pose considerable logistical, fiscal, and experimental control problems. First, data analysis is extremely complicated, owing simply to the large number of participants and data sources in such simulations. In addition, experienced human air traffic controllers working adjacent sectors tend to flexibly adapt to the evolving control problem - potentially shifting to other strategies than those under investigation. In addition, their performance is tightly coupled to the control interface, which in the development phase may support some concepts and supporting strategies better than others. A simple shift in strategy by one controller can change the character of a particular traffic scenario dramatically, which makes experimental comparison of ATC performance under different traffic scenarios difficult. Training a given team of controllers on operations under a new ATM concept for a sufficient period of time could avert such difficulties, but instituting an adequate training program is expensive and logistically difficult.
Callantine, Todd J.↗
Application of multi-criteria decision analysis techniques and decision support framework for informing plant select agent designation and decision making
The United States Department of Agriculture (USDA) Division of Agricultural Select Agents and Toxins (DASAT) established a list of biological agents (Select Agents List) that threaten crops of economic importance to the United States and regulates the procedures governing containment, incident response, and the security of entities working with them. Every 2 years the USDA DASAT reviews their select agent list, utilizing assessments by subject matter experts (SMEs) to rank the agents. We explored the applicability of multi-criteria decision analysis (MCDA) techniques and a decision support framework (DSF) to support the USDA DASAT biennial review process. The evaluation includes both current and non-select agents to provide a robust assessment. We initially conducted a literature review of 16 pathogens against 9 criteria for assessing plant health and bioterrorism risk and documented the findings to support this analysis. Technical review of published data and associated scoring recommendations by pathogen-specific SMEs was found to be critical for ensuring accuracy. Scoring criteria were adopted to ensure consistency. The MCDA supported the expectation that select agents would rank high on the relative risk scale when considering the agricultural consequences of a bioterrorism attack; however, application of analytical thresholds as a basis for designating select agents led to some exceptions to current designations. A second analytical approach used agent-specific data to designate key criteria in a DSF logic tree format to identify pathogens of low concern that can be ruled out for further consideration as select agents. Both the MCDA and DSF approaches arrived at similar conclusions, suggesting the value of employing the two analytical approaches to add robustness for decision making.
59 BASIC BIOLOGICAL SCIENCES↗
MADA: Multi-Agent Design Assistant
MADA (Multi-Agent Design Assistant) is a Large Language Model (LLM) powered multi-agent framework that coordinates specialized agents for complex design workflows. The system was designed for HPC workflows with the following agents in mind: 1) A Job Management Agent (JMA) launches and manages ensemble simulations on HPC systems, 2) a Geometry Agent (GA) generates meshes, and 3) an Inverse Design Agent (IDA) proposes new designs informed by simulation outcomes. Our framework reduces cumbersome manual workflow setup, and enables automated design exploration at scale. However, the software also enables users to rapidly create new multi-agent systems. Simply define new agents in a configuration file, giving each their own set of tools (via MCP), and then chat and prompt your new multi-agent system. Is
Gunnarson, BrianS [Lawrence Livermore National Lab↗