Description:
OBJECTIVE: Develop innovative intelligent agents for the prediction of swarming and other nonlinear, dispersed tactics against Naval surface forces utilizing airborne radar and electro-optic sensor information. DESCRIPTION: Naval forces conducting transits through straits and other congested littoral operational areas are presented with a challenging force protection requirement. Surface traffic density is often high, with many ferries, fishing and pleasure boats, and large cargo ships maneuvering in a small area. Existing rules of engagement designate query and warning ranges, but international law and freedom of navigation allow vessels to operate in very close proximity to our combatants. These small vessels are often difficult to regulate and many lack basic equipment such as bridge to bridge radios. With a host of stationary and seemingly randomly moving boats, determining a hostile action in a timely manner is difficult at best. These conditions make the identification of and defense against hostile small craft extremely difficult. Of particular concern when operating in these areas is countering pre-planned or opportunistic hostile swarm attacks. A variety of swarm and counter-swarm tactics exist, but the single most important enabler for success is having superior situational awareness. This is in fact a fundamental advantage that all types of military forces seek. At present airborne surveillance radar and electro-optics are of limited value to the surface combatant as they are seen as unable to provide the ship watch standers actionable information to support the determination of hostile action and/or hostile intent. A contact's speed, angle of approach, and location are not sufficient to determine intent in such high density environments. In the absence of additional insight regarding intent, surface combatants rely on a last minute determination of intent which is based on based direct visual observation by the ship's watch standers. However, radar systems can now provide the ability to observe the behavior of specific targets over extended periods of time utilizing maritime classification aids and target fingerprinting. The missing component and the focus of this topic is the development intelligent agent to predict impending threats, much as is done with crime forecasting tools, based on a knowledge of target classification and behaviors of all contacts within the surveillance field of view. These analysis tools are necessary for general pattern analysis including the detection of preparations for swarm encirclement actions. PHASE I: Investigate technical approaches that provide determination of, or give indications of, hostile intent against Naval surface forces operating in congested littoral operational areas. Radar and electro-optic maritime classification aids along with high range resolution radar fingerprinting and tracking are to be utilized in the technical approach. Identify the specific nature of the analysis algorithms and the human-machine interfaces to be used and develop a detailed implementation plan. PHASE II: Develop a robust intelligent agent analysis tool for the candidate sensor suite. Demonstrate non real time processing using government-provided data with sufficient fidelity to enable assessment of operator workload reduction, algorithm tuning, and sensor utilization. Demonstrate the functionality, performance, and correctness of all components. Prepare an integration plan to complete the development and transition of the toolset into the candidate sensor suite. PHASE III: Transition the intelligent agent to appropriate platforms and interested commercial entities. PRIVATE SECTOR COMMERCIAL POTENTIAL/DUAL-USE APPLICATIONS: There is a growing need for real-time automated sensor information analysis and planning tools. The ability to analyze and assess sensor information intelligently in time-critical scenarios is crucial for many commercial endeavors.
