Application of Multi-Agent Systems to Free Flight and ATM

1
Application of Multi-Agent Systems to Free Flight
and ATM

• What are agents?
• What role can they play in aviation?
• What are the most relevant technologies?
• What are future directions for theoretical
  research?
• What are future directions for applied research?

2
Roles for Agents in Aviation

• Agents are software processes that can simulate
  decision-making and be adaptive
• In cockpit planning flight path, managing fuel,
  maintaining stability of flight, monitoring
  traffic or weather conflicts
• On ground (TRACON, ARTCC) planning trajectories,
  resolving conflicts, approach metering, handling
  emergencies, coordination with ground ops,
  airlines, etc.

3
Agent Interactions Between Aircraft

• The Vision automatically formulate and negotiate
  En route and Terminal trajectories
• En route de-crowd trans-oceanic routes, permit
  airlines wider choices on climb/descent for
  efficiency, routing around weather cells
• Terminal Area efficient sequencing handle speed
  variability better
• Handle by on-board computers
• Use datalink for Air/Air Air/Ground comm.

4

• ATC only has to monitor, or occasionally
  arbitrate
• Decentralized computing reduce bottlenecks and
  decrease sensitivity to failures/attacks
• Off-load approval of minor deviations
• Still maintains ultimate authority

5
What are agents?

• Essential Characteristics
• Situated (can sense and take actions in dynamic
  environment)
• Goal-oriented
• Autonomous
• Social (collaborative)
• Types of Agents (abstract architectures)
• Reactive (trigger rules)
• Deliberative (reasoning, planning)
• Cognitive (Mentalistic, BDI beliefs, desires,
  intentions)
• Utility-based, decision theoretic

6
Collaboration Models

• Negotiation protocols
• Contract networks
• Bids based on marginal utility
• Share justifications and beliefs to compromise
• Teamwork
• Command hierarchies (with delegation) vs.
  distributed structure (load-balancing, consensus)
• Key concepts roles and responsibilities
• Shared plans implicit coordination,
  synchronization
• Theory joint intentions (for robust backup
  behavior)

7
Concepts for Development of Multi-Agents for Free
Flight

• Strategic (trajectory planning/management) vs.
  Tactical (avoidance maneuvers)
• Actionable decisions
• Alter flight path heading, altitude, speed
• Factors weather, terrain, traffic
• Constraints fuel, speed/alt range
• Preferences time, fuel cost, comfort

8
Distributed Constraint Satisfaction

• Conflict detection projected interferences
• No designated leader with universal authority
• Dynamic coalition formation
• Initiating agent proposes a solution others
  refine it
• Negotiation by argumentation
• State what is wrong with proposed solution and
  why
• Communicate preferences as well as constraints
• make up when behind schedule
• minimize fuel consumption
• maneuver limitations (safety, comfort)
• Shared responsibility (with ground too)

9
Role of Simulated Mental Attitudes

• Intent transmit more than position/vector
• Desire to avoid weather, flight plan, will be
  turning north, descending due to turbulence,
  reason for deviation
• Beliefs
• shared info (weather, congestion, aircraft
  emergencies)
• common picture of situation
• common knowledge STARs, fixes, active runways,
  traffic patterns
• manage uncertainty

10
Prior Work on Multi-Agent Systems at Texas AM

• 1999-2001 University XXI program
• Coordinated through CTSF at Ft. Hood
• Designed TaskableAgents architecture to train
  brigade TOC staff officers by simulating
  interactions with battalion TOCs
• Interoperated with OneSAF Testbed distributed
  combat simulation (STRICOM DIS protocol)

11

• 2001-? Army Research Lab
• Funding to continue work on TaskableAgents
• Add HLA-interoperability
• Augment command-and-control (C2) reasoning
• Add cognitive models of situation assessment

12

• 2000-2005 MURI (DOD/AFOSR), 4.3M
• Multi-disciplinary University Research Initiative
• Intelligent team and group training (e.g. DMT)
• Merge agent technology and cognitive theories to
  find principled methods for using agents to
  improve human and team performance
• Developing CAST Architecture Collaborative
  Agents for Simulating Teamwork
• Collaborators
• Richard Volz (TAMU, CS), John Yen (PSU, CS)
• Wayne Shebilske, Pamela Tsang (Wright State
  Dept. of Psychology)