Utilizing ML Algorithms to Track and Identify UAS Threats

The objective of this feasibility study is to assess the concept of using LiDAR to detect, track, and identify sUAS threats assisted by Artificial Intelligence (AI) agents. Inherent in this objective is concept development and feasibility assessment of Machine Learning (ML) and AI algorithms for creation and use of LiDAR target profiles for sUAS surveillance and identification. This objective will be satisfied with fact-based recommendations for an overall system design to improve sUAS detection, provide real-time alerts and geolocate sUAS in flight. During this phase, the team will investigate the availability and suitability of LiDAR sensors to address the following issues: a) Detection and tracking of sUAS in flight; b) Discrimination and identification of sUAS; c) Capability to integrate with other C-UAS sensors; d) Integration of AI/ML fusion applications; and e) Contribution to a 3D object mapping library. We will assess the performance of LiDAR sensors using the following criteria: a) Detection – range, accuracy, and completeness; b) Geospatial accuracy; c) Track accuracy and completeness; d) Capability of producing target profiles supporting rapid identification and object discrimination; and e) Considerations of Size, Weight, and Power (SWaP) and durability of the LiDAR system.