Modeling Teamwork in the CAST Multi-Agent System

1
Modeling Teamwork in theCAST Multi-Agent System

• Thomas R. Ioerger
• Department of Computer Science
• Texas AM University

2
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

3
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

4
CAST Collaborative AgentArchitecture for
Simulating Teamwork

• developed at Texas AM 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.

5
CAST Architecture
expand team tasks into Petri nets
keep track of who is doing each step
agent teammates
MALLET knowledge base (definition of
roles, tasks, etc.)
messages
human teammates
events, actions state data
JARE knowledge base (domain rules)
simulation
make queries to evaluate conditions, assert/retrac
t information
models of other agents beliefs
Agent
6
MALLET
descriptions of team structure
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))
• (if (not (hit ?target))
• (do ?s (report-accuracy-of-aim FSO))
• (do FSO (adjust-coordinates ?target))))))

descriptions of team process
7
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))
• gt(query (threat ?x TF-122))
• solution 1 ?x Reg-52
• solution 2 ?x Reg-54

8
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 (pre-conds
  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

9
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
• (bel S2 (attacking reg-54 TF-122))
• (bel S2 (location Reg-60 unknown))
• (bel S6 (cutoff supply-route TF-122))

S3-agents database
10
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

11
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
  sub-tasks 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

12
Applications of CAST

• Battalion Tactical Operations Centers
• model of staff operations (agentsS2, 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