Tactical Autonomous Look Optimization Network (TALON)

SSCI is proposing to use its flight-proven autonomy stack with information-driven path planning ATR algorithms using innovative Autonomous Closed-Loop Sensing (ACLS) technology to accelerate information gain for target classification across a set of autonomous assets. The system is referred to as TALON (Tactical Autonomous Look Optimization Network) and employs SSCI’s TRL 7 modular open system architecture autonomy stack implemented in C++ and designed for deployment on edge processing platforms like NVIDIA Jetson boards. The autonomy stack referred to as RAPTOR (Robotic Autonomy Platform for Tactical Operations and Reconnaissance) has integral swapable ATR libraries, adversarial target trackers, decision engines with behaviors trees to evaluate information, and route planners/executors tightly coupled in an operational loop. RAPTOR will act as the development testbed for ‘next best look’ optimization algorithms providing all the controls and analysis to gather formation and execute platform and sensor adjustments in an encapsulated piece of software.  Under the project we will also extend SSCI’s simulation framework with scalable complexity and fidelity for various scenarios to develop and evaluate directed information fusion algorithms using Bayesian Estimation to calculate ‘next best look’ information gain. The simulation environment uses a combination of Airsim for platform dynamics and sensor simulation, and Unreal engine for complex environment rendering to test and evaluate ATR algorithms, both compatible with the Air Force Colosseum environment.