Innovations in multi-agent systems

In the aftermath of major disasters, first responders are typically overwhelmed with large numbers of, spatially distributed, search and rescue tasks, each with their own requirements. Moreover, responders have to operate in highly uncertain and dynamic environments where new tasks may appear and hazards may be spreading across the disaster space. Hence, rescue missions may need to be re-planned as new information comes in, tasks are completed, or new hazards are discovered. Finding an optimal allocation of resources to complete all the tasks is a major computational challenge. In this paper, we use decision theoretic techniques to solve the task allocation problem posed by emergency response planning and then deploy our solution as part of an agent-based planning tool in real-world field trials. By so doing, we are able to study the interactional issues that arise when humans are guided by an agent. Specifically, we develop an algorithm, based on a multi-agent Markov decision process representation of the task allocation problem and show that it outperforms standard baseline solutions. We then integrate the algorithm into a planning agent that responds to requests for tasks from participants in a mixed-reality location-based game, called AtomicOrchid, that simulates disaster response settings in the real-world. We then run a number of trials of our planning agent and compare it against a purely human driven system. Our analysis of these trials show that human commanders adapt to the planning agent by taking on a more supervisory role and that, by providing humans with the flexibility of requesting plans from the agent, allows them to perform more tasks more efficiently than using purely human interactions to allocate tasks. We also discuss how such flexibility could lead to poor performance if left unchecked.

/pdf/human-agent-collaboration-for-disaster-response-53guiwf6vu.pdf

Human---agent collaboration for disaster response

Motivation - To investigate ways to support human-automation teams with real-world, imperfect automation where many system failures are the result of systematic failure.Research approach - An experimental approach was used to investigate how variance in agent reliability may influence human's trust and subsequent reliance on agent's decision aids. Sixty command and control (C2) teams, each consisting of a human operator and two cognitive agents, were asked to detect and respond to battlefield threats in six ten-minute scenarios. At the end of each scenario, participants completed the SAGAT queries, followed by the NASA TLX queries.Findings/Design - Results revealed that teams with experienced human operators accepted significantly less inappropriate recommendations from agents than teams with inexperienced operators. More importantly, the knowledge of agent's reliability and the ratio of unreliable tasks have significant effects on human's trust, as manifested in both team performance and human operators' rectification of inappropriate recommendations from agents.Originality/Value - It represents an important step toward uncovering the nature of human trust in human-agent collaboration.Take away message - This research has shown that given even minimal basis for understanding when the operator should and should not trust the agent recommendations allows operators to make better AUDs, to have better situation awareness on the critical issues associated with automation error, and to establish better trust in intelligent agents.

/pdf/the-influence-of-agent-reliability-on-trust-in-human-agent-228qs74igw.pdf

The influence of agent reliability on trust in human-agent collaboration

Team decision making under stress involving multiple contexts is an extremely challenging issue faced by various real world application domains. This research is targeted at coupling cognitive agent technology and human-centered teamwork to address the informational challenges associated with Command and Control (C2) teams in contemporary military operations. Two sets of experiments, each with various settings of context switching frequencies and tasking complexities, were conducted. To ensure that the human subjects were familiar with the C2 context, they were selected from US Army ROTC (Reserve Officer Training Corps) students. Experiments on C2 teams that involve human subjects only were conducted first. We observed the decision making behavior of human subjects and incorporated expert behaviors into R-CAST---an agent architecture built upon a naturalistic decision making model that captures how domain experts make decisions based on experiences and situational similarity recognition. We then conducted another set of experiments with R-CAST agents as teammates and decision aids for human subjects. The results show that RPD-enabled agents can significantly improve the tasking capacity of C2 teams in multi-context decision making under stress. It also suggests that higher demand situations require more competent teammates.

/pdf/rpd-enabled-agents-teaming-with-humans-for-multi-context-5e4inh26l5.pdf

RPD-enabled agents teaming with humans for multi-context decision making

An integrated approach to high-level information fusion

Intelligent aiding strategies were evaluated for a team planning task. The MokSAF interface agent links an Artificial Intelligence (AI) route planning agent to a Geographic Information System (GIS). Through this agent, the user specifies a start and an end point, and describes the composition and characteristics of a military platoon. Two aided conditions and one non-aided condition were examined. In the first aided condition, a route-planning agent determines a minimum cost path between the specified end points. The user is allowed to define additional “intangible” constraints that describe situational or social information. In the second aided condition the same knowledge of the terrain is used to plot the best route within bounds specified by the user. In the control condition the user draws a route without help. The reported study found across the board advantages for agent-based aiding when routes were merged in a team-planning task.

/pdf/agent-based-aiding-for-individual-and-team-planning-tasks-46zdou6tg4.pdf

Agent-Based Aiding for Individual and Team Planning Tasks

Evaluates the effectiveness of agent-based aids in support of a time-critical team-planning task for teams of both humans and heterogeneous software agents. The team task consists of human subjects playing the role of military commanders and cooperatively planning to move their respective units to a common rendezvous point, given time and resource constraints. The objective of the experiment was to compare the effectiveness of agent-based aids for individual and team tasks, as opposed to the baseline condition of manual route planning. There were two experimental conditions: the aided condition, where a route-planning agent (RPA) finds a least-cost plan between the start and rendezvous points for a given composition of force units, and the baseline condition, where the commanders determine initial routes manually and receive basic feedback about the route. We demonstrate that the aided condition provides significantly better assistance for individual route planning and team-based re-planning.

/pdf/agent-based-team-aiding-in-a-time-critical-task-3gmwkkos7h.pdf

Agent-based team aiding in a time critical task

In this paper, we describe the interactions between task characteristics and human agent interfaces in a team redezvous route-planning task. The agents include an interface agent and two different task agents that perform similar tasks. The MokSAF interface agent links an Artificial Intelligence (AI) route planning agent to a Geographic Information System (GIS). Through this agent, the user specifies a start and an end point, and describes the composition and characteristics of a military platoon. Two aided conditions and one non-aided condition were examined. In the first aided condition, a route-planning agent (known as the Autonomous RPA) determines a minimum cost path between the specified end points. The user is allowed to define additional "intangible" constraints that describe situational or social information that should be considered when determining the route. In the second aided condition, a different agent, the Cooperative RPA, uses the same knowledge of the terrain and cost functions available to the Autonomous RPA, but restricts its search to paths within regions drawn by the user. In the unaided condition, Naive RPA, the user draws the route manually, then submits it to be tested against the terrain and cost functions for feasibility. Both aided conditions are superior to the control but differ in their relative effectiveness by scenario. In this paper we examine the varieties of challenges faced by commanders in two scenarios and relate them to the differential effectiveness of the agents.

Task Characteristics and Intelligent Aiding

In this paper, we describe an interface agent, two different route planning agents and a pilot study which examined whether these agents could support a team planning task. The MokSAF interface agent links an Artificial Intelligence (AI) route-planning agent to a Geographic Information System (GIS). The user specifies a start and an end point and the route-planning agent finds a minimum cost path between the points. The user is allowed to define additional “intangible” constraints (not due to terrain characteristics) corresponding to geographic regions, which can be used to steer the agent’s behavior in a desired direction. A second agent (the naive route planning agent, or Naive RPA) has access to the same knowledge of the terrain and cost functions available to the Autonomous RPA, but uses this knowledge to critique paths specified by the user. We hypothesize that as the complexity of intangible aspects of a planning problem increase, the Naive RPA will improve in relative performance. The reported study found advantages across the board for the Autonomous RPA in a team-planning task.

/pdf/moksaf-how-should-we-support-teamwork-in-human-agent-teams-rusxpbf343.pdf

MokSAF: How should we support teamwork in human-agent teams?

As part of our research program on integration of intelligent agents with human teams, we analyzed the performance of 34 teams (102 subjects) in a target identification task Teams were assigned to one of four conditions The control group received no aiding; each agent condition was focused on an alternative strategy: 1) support individual team members by tracking known information; 2) support team communication by automatically passing information to the relevant member; or 3) support team coordination by identifying responsibility for specific data We found performance improvements for teams across trials under all conditions Teams using communication and coordination agents performed better on hard targets, although subjects perceived the coordination agent to be the least valuable for teamwork support Our findings suggest that direct support of teamwork is more beneficial than aiding individual tasks

Improving Performance: Should We Support Individuals or Teams?

Handle everyday research tasks with reliable, citation-backed results

Your personal Research Agent to handle research tasks with citation-backed results

Popular Tasks used by Researchers

How can I help with your research?

Meet SciSpace

Get more enhanced response by uploading the PDFs you want me to reference.

No relevant PDFs in your library

SciSpace is the AI research assistant for academics. Run systematic literature reviews on 280M+ papers, and write papers with cited sources. Trusted by 1M+ students, PhDs & researchers.

SciSpace | AI for Research

Analyze PDFs

Code & Manuscripts

Funding & Grants

Literature & Patents

Medical & Clinical Data

Systematic Review

Visualize & Present

Web & Data

Build a Google Scholar-like website for your research.

Build a website

Create charts and images for your research

Create a Chart

Write a paper for submission to a journal

Draft a manuscript

Patent Search

Design eye-catching scientific posters in minutes.

Scientific Poster Generation

Systematic Literature Review

One task is running at the moment. Your messages will be shown right after.

Drag and drop or click here to browse

Loved by <highlight>1 million+</highlight> researchers

Extract a list of specific topics and their sources from unstructured text

Topics

Compare and analyze relevant papers that matches with your search

Papers

Get insights from PDFs and bookmarked papers from your library

My library

Recent searches

Try searching for:

Catch AI-generated content in scholarly and non-scholarly content

{ai} Detector

Ai Writer

Get PDF Summaries, highlighted text explanations 

Chat with PDF

Effortlessly create in-text citations and bibliographies in APA and 2,500 other formats

Citation generator

Get explanations, summaries, and answers on academic papers

Ease up your research workflow with {scispace}'s cohort of exciting AI tools

Elevate your academic writing skills and convey your ideas the way you want

Paraphraser

Explore our range of reading and writing tools

Your file is being prepared and should be ready in a few minutes. If it's a large file, it might take a bit longer. You can close this window, and we'll email you the file when it's done.

You have reached a maximum limit of <strong>{limit}</strong> columns in the table. Remove at least <strong>1</strong> column to add or create another one.

Susan Hahn

Author Tools

Chat about Author