UTOPIACOMPRESSION,CORPORATION — Department of Defense SBIR Phase I: A17-102

Description

UtopiaCompression proposes novel machine learning based multi-scale detection algorithm coupled with a novel tracking approach using a Track Before Detect (TBD) paradigm to address the problem of problem of detecting and tracking small Unmanned Aircraft System and Mortars in a low Signal-to-Noise Ratio (SNR).Small UAS platforms are smaller in size, fly at lower altitudes and are slower in speed when compared to other larger air traffic. Mortars, on the other hand, move in a predictable manner, but have low SNR and small size that makes them harder to detect and track, especially against cluttered background. During Phase I, we will conduct requirements analysis to identify an appropriate Medium Wave Infrared imaging system and a suitable compute platform to support the requested data rates.We will leverage our experience in developing systems for UAS detect and avoid applications to provide a computationally feasible solution to the problem at hand. We will demonstrate the feasibility using modeling and simulation methods and by conducting our own data collection experiments. Finally, we will analyze the computational performance of our algorithm and direct our efforts in identifying suitable platforms to enable real-time detection and tracking of multiple targets in low SNR and cluttered environments.