AGATE generates a set of ranked strategies that enables an autonomous vehicle to track/trail another vehicle that is trying to break the contact using evasive tactics. The software is efficient (can be run on a laptop), scales well with environmental complexity, and is suitable for use onboard an autonomous vehicle. The software will run in near-real-time (2 Hz) on most commercial laptops. Existing software is usually run offline in a planning mode, and is not used to control an unmanned vehicle actively.
JPL has developed a system for AGATE that uses adversarial game theory (AGT) methods (in particular, leader-follower and pursuit-evasion) to enable an autonomous vehicle (AV) to maintain tracking/trailing operations on a target that is employing evasive tactics. The AV trailing, tracking, and reacquisition operations are characterized by imperfect information, and are an example of a non-zero sum game (a positive payoff for the AV is not necessarily an equal loss for the target being tracked and, potentially, additional adversarial boats). Previously, JPL successfully applied the Nash equilibrium method for onboard control of an autonomous ground vehicle (AGV) travelling over hazardous terrain.
This work was done by Terrance L. Huntsberger of Caltech for NASA’s Jet Propulsion Laboratory.
This software is available for commercial licensing. Please contact Daniel Broderick of the California Institute of Technology at
This Brief includes a Technical Support Package (TSP).
AGATE: Adversarial Game Analysis for Tactical Evaluation
(reference NPO-48697) is currently available for download from the TSP library.
Don't have an account?
Overview
The document titled "AGATE: Adversarial Game Analysis for Tactical Evaluation" is a technical support package from NASA's Jet Propulsion Laboratory (JPL) that discusses the application of adversarial game theory techniques in the context of autonomous vehicles (AVs) operating in complex maritime environments. The focus is on the challenges faced by AVs, particularly when trailing submarines that employ evasive tactics and are potentially supported by adversarial surface boats.
The document outlines two primary game theoretical models relevant to tracking and trailing operations: the pursuit-evasion model and the leader-follower model. The pursuit-evasion model is best suited for reacquiring a submarine when contact is lost, while the leader-follower model is more appropriate for continuous tracking and trailing operations. The choice of these models is influenced by the difficulties associated with applying traditional minimax or alpha-beta pruning methods in continuous state situations, which can lead to accuracy issues due to the exponential branching of game trees.
The document highlights the successful application of the leader-follower game theory approach in real-world scenarios, such as the ARMOR security system at Los Angeles International Airport, which utilizes game theory for strategic resource allocation in security operations. This approach is also relevant for the IRIS Federal Air Marshals Service in flight scheduling to enhance security against potential threats.
A significant portion of the document discusses the AGATE algorithm, which was tested in a simulation scenario set in the Luzon Strait. The simulation involved an AV trailing a submarine while avoiding adversarial surface boats. The AGATE algorithm demonstrated promising results, accurately predicting the submarine's maneuvers and allowing the AV to maintain tracking without loss of contact. The algorithm's near real-time processing capability (operating at 2Hz and 5Hz) indicates its potential for onboard implementation in AVs.
Overall, the document emphasizes the importance of formal adversarial game theory techniques in addressing the dynamic and asymmetric challenges of modern maritime operations. It serves as a resource for researchers and practitioners interested in the intersection of game theory, autonomous systems, and tactical evaluation in complex environments.
