Modeling Teamwork in the CAST Multi-Agent System

Modeling Teamwork in the CAST Multi-Agent System

Modeling Teamwork in the CAST Multi-Agent System Thomas R. Ioerger Department of Computer Science Texas A&M University The Nature of Teamwork Team Psychology Research: Salas, Cannon-Bowers, Serfaty, Ilgen, Hollenbeck, Koslowski, etc. 2 or more individuals working together... members often distinct roles types of control: centralized (hierarchical) vs. distributed (consensus-oriented) process measures vs. outcome measures communication, adaptiveness shared mental models

Computational Models of Teamwork Commitment to shared goals Joint Intentions (Cohen & Levesque; Tambe) Cooperation, non-interference Backup roles, helping behavior Mutual awareness goals of teammates; achievement status information needs Coordination, synchronization Distributed decision making consensus formation (voting), conflict resolution CAST: Collaborative Agent Architecture for Simulating Teamwork

developed at Texas A&M; part of MURI grant from DoD/AFOSR, plus support from ARL multi-agent system implemented in Java components: MALLET: a high-level language for describing team structure and processes JARE: logical inference, knowledge base Petri Net representation of team plan special algorithms for: belief reasoning, situation assessment, information exchange, etc. CAST Architecture MALLET knowledge base (definition of roles, tasks, etc.) JARE knowledge

base (domain rules) expand team tasks into Petri nets keep track of agent teammates who is doing each step messages human teammates events, actions state data make queries to evaluate conditions,

assert/retract information models of other agents beliefs Agent simulation descriptions of team structure MALLET evaluated by queries to JARE

knowledge base (role sam scout) (role bill S2) (role joe FSO) (responsibility S2 monitor-threats) (capability UAV-operator maneuver-UAV) (team-plan indirect-fire (?target) (select-role (scout ?s) (in-visibility-range ?s ?target)) (process (do S3 (verify-no-friendly-units-in-area ?target)) (while (not (destroyed ?target)) (do FSO (enter-CFF ?target)) (do ?s (perform-BDA ?target)) descriptions of (if (not (hit ?target)) (do ?s (report-accuracy-of-aim FSO)) team process (do FSO (adjust-coordinates ?target))))))

JARE Knowledge Base First-order Horn-clauses (rules with variables) Similar to PROLOG Make inferences by back-chaining consequent antecedents ((threat ?a ?b)(enemy ?a)(friendly ?b) (in-contact ?a ?b)(larger ?a ?b) (intent ?a aggression)) >(query (threat ?x TF-122)) solution 1: ?x = Reg-52 solution 2: ?x = Reg-54 Proactive Information Exchange

Information sharing is a key to efficient teamwork Want to capture information flow in team, including proactive distribution of information Agent A should send message I to Agent B iff: A believes I is true A believes B does not already believe I (non-redundant) I is relevant to one of Bs goals, i.e. pre-condition of current goal that B is responsible for in plan DIARG Algorithm (built into CAST): 1. check for transitions which other agents are responsible for that can fire (preconds satisfied) 2. infer whether other agent might not believe pre-conds are true (currently, beliefs based on post-conditions of executed steps, i.e. tokens in output places) 3. send proactive message with information

Belief Reasoning Modeling teammates beliefs - an important component of a shared mental model Knowledge of pre-conds and effects of actions Mutual observability - common events in env. e.g. messages on radio net; symbols shared on radar Interaction with other justifications of beliefs: inferences, assumptions, persistence Prioritized logic program impacts team interactions and communications S3-agents database (bel S2 (attacking reg-54 TF-122)) (bel S2 (location Reg-60 unknown))

(bel S6 (cutoff supply-route TF-122)) Situation Awareness Central part of Command-and-Control information gathering, group decision-making, Drives communication within teams in tactical environments Our approach: based on RPD Recognition-Primed Decision-Making Finite set of anticipated situations: e.g. flank attack, deception, bypass, pincer Look for features associated with each situation Trigger when sufficient features found Implementation of SA in CAST

Generic plan for RPD in MALLET Domain-specific situations and features encoded in JARE predicates Plan executes loop: While no situation has sufficient features detected: For each feature F whose value is unknown but potentially relevant to situation, try to find out about status of F Domain-specific find-out procedures encoded as subtasks: ask scouts, ask radar operator, ask S2, use UAV, check JSTARs, take probing actions... info. management: what questions to ask when? current work: extend to multi-agent team plan Applications of CAST Battalion Tactical Operations Centers

model of staff operations (agents=S2, S3, FSO...) hooked-up to OTB (OneSAF Testbed Baseline) focus on modeling information flow interact with human brigade staff trainees via report/request forms AWACS weapons directors model coordination, load-balancing hooked-up to DDD simulation (Aptima) studying effects of workload on helping behavior

Recently Viewed Presentations

  • Forensic Biology by Richard Li - Fayetteville State University

    Forensic Biology by Richard Li - Fayetteville State University

    Antigen polymorphisms that are present on the erythrocyte surface 29 blood group systems E.g. Rh, MNSs, Kell, ABO ABO system is the most commonly used in forensics Forensic Biology by Richard Li * ABO System A, B, AB, O Type...
  • Master Asl! Important Things to Know….

    Master Asl! Important Things to Know….

    MASTER ASL! IMPORTANT THINGS TO KNOW…. GLOSSING pgs. 129 - 130 in your workbook Developed by ASL teachers for HEARING students. Deaf people do NOT write in GLOSS. Glossing is a system of using one language to write another. GLOSSING...
  • LINGUISTICS 200: Introduction to Linguistic Thought

    LINGUISTICS 200: Introduction to Linguistic Thought

    Consonants Speech sounds produced with a narrowing somewhere in the vocal tract Major structures Active vs. Passive Articulators Articulatory Description For consonants, three-part classification system: 1) Voicing 2) Place (of articulation) 3) Manner (of articulation) e.g., voiced labiodental fricative =...
  • Modèles hiérarchiques probabilistes et bio-inspirés de la ...

    Modèles hiérarchiques probabilistes et bio-inspirés de la ...

    Par fonction analytique f E F x | f(x) Par extension Ensemble de points (pas pratique pour un ensemble infini) * Quelle méthode pour la devinette ? Passage de points à une fonction Utilisation de la fonction pour prédire le...
  • Parameters of ASL

    Parameters of ASL

    Handshape. There are over 45 different hand shapes in ASL . A - aunt, sweetheart open A - girl, which, remember, tomorrow. B - son, blue, brown, door. C - cousin ...
  • What were the differences between the lives of the rich and ...

    What were the differences between the lives of the rich and ...

    What jobs do you think most people did in Tudor times? How was life in Tudor times different from life today? Homework: Try to find some information about Tudor houses. What were the differences between the lives of the rich...
  • Receiving, Distributing, and Dispensing SNS Assets

    Receiving, Distributing, and Dispensing SNS Assets

    Document Overview:Receiving, Distributing, and Dispensing SNS Assets . A Guide to Preparedness . Version 11. Jim Sowards, WV SNS Coordinator. Humbert Zappia, CDC Public Health Advisor
  • Where did the social planner go? Authoritarian rulers ...

    Where did the social planner go? Authoritarian rulers ...

    Certainly not the social planner that dictates the optimal allocation of resources at will! (Nobody really believes this anyway, after thinking over it. The problem is that many analysts forget thinking about the issue.) Institutions and constraints matter also in...