Making adversary decision modeling tractable with intent inference and information fusion

Dept of Computer Science & Engineering UTEB, University of Connecticut, U-155, 06269-3155, Storrs, CT

ABSTRACT Military and domestic security analysts and planners are facing threats whose asymmetric nature will sharply increase the challenges of establishing an adversary's intent. This complex environment will severely limit the capabilities of the classic doctrinal approach to diagnose adversary activity. Instead, a more dynamic approach is required -adversary decision modeling (ADM) -that, while a critical capability, poses a range of daunting technological challenges. We are developing methodologies and tools that represent a tractable approach to ADM using intelligent software-based analysis of adversarial intent. In this paper we present work being performed by our team (University of Connecticut, Lockheed Martin Advanced Technology Laboratories, and the Air Force Research Laboratory Human Effectiveness Directorate) toward a preliminary composite theory of adversary intent and its descriptive models, to provide a coherent conceptual foundation for addressing adversary decision processes, tasks, and functions. We then introduce notional computational models that, given own system-of-systems actions (movements and activities) and observations of an adversary's actions and reactions, automatically generate hypotheses about the adversary's intent. We present a preliminary software architecture that implements the model with: (1) intelligent mobile agents to rapidly and autonomously collect information, (2) information fusion technologies to generate higher-level evidence, and (3) our Intent Inference engine that models interests, preferences, and context.

  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Intent inference involves the analysis of actions and activities of a target of interest to deduce its purpose. This paper proposes an ap-proach for intent inference based on aircraft flight profile analysis. Simulation tests are carried out on flight profiles generated using different combinations of flight parameters. In each simulation test, Interacting Multiple Model-based state estimation is carried out to update the state vectors of the aircraft being monitored. Relevant variables of the filtered flight trajectory are subsequently used as inputs for a Mamdani-type fuzzy inference system. Research on two applications is reported. The first application involves the determi-nation of the likelihood of weapon delivery by an attack aircraft under military surveillance. Test results verify that the method is feasible and is able to provide timely inference. By extending the method to take the environmental context of the tracked aircraft into consideration when executing the inference process, it is likely that the military defenders would be able to raise their alert earlier against potential adversaries. This would provide them with more time to react and devise pre-emptive counteraction. The second ap-plication concerns conformance monitoring in air traffic control sys-tems. Experimental results show that the proposed solution can be used to assist air traffic control system operators in determining if aircraft navigate according to planned trajectories. Consequently, corrective action can be taken on detection of anomalous behavior. A brief discussion on extending the proposed method to deal with multiple aircraft is also presented.
  • [Show abstract] [Hide abstract]
    ABSTRACT: We examine an adversary model that captures goals, intentions, biases, beliefs, and perceptions based on a dynamic cognitive architecture that evolves over time. The model manages the uncertainty surrounding the adversary using probabilistic networks. In particular, we consider the challenges of constructing such adversaries and provide solutions towards more effective and efficient engineering of such adversaries. We present the AII Template Generator tool which enables the rapid deployment of adversary models as well as on-demand construction of new adversary components.
    Proceedings of SPIE - The International Society for Optical Engineering 08/2004; DOI:10.1117/12.546771 · 0.20 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Intelligence Preparation of the Battlespace (IPB) is a pre- dominantly "gray matter-based" fusion and information syn- thesis process conducted to predict possible future adversary courses of action. The purpose is to understand where the enemy is in the battlespace, and to infer what we believe he will do next. From that understanding, military commanders plan their own course of action. As the state of the art im- proves, we are in a position to begin applying technologies to move the labor-intensive parts of IPB to the computer, al- lowing the planner to perform those tasks that are more suited to human capabilities. This is a primary focus of our research effort. This paper presents the approaches that we are adopting to acquire the knowledge necessary to build models to assist decision makers determine adversary intent. We discuss how the IPB process can assist with knowledge acquisition and we present a detailed discussion of our AII system and the techniques we have developed to collect and process the data necessary to map observations of the adver- sary into evidence to support reasoning about the adver- sary's intent.

Full-text (4 Sources)

Available from
May 27, 2014