Lidar/ ML cUAS detection and identification

Atolla Tech is pioneering an advanced lidar-based Counter-Unmanned Aerial System (C-UAS) that rapidly identifies unmanned aerial systems (UAS) using their propeller electronic signatures. This process is enhanced by a machine learning (ML) algorithm that performs waveform analysis in both time and frequency domains, necessitating minimal data points for identification. This approach ensures quick processing and high fidelity in target identification, overcoming radar limitations in resolution and cost while surpassing the range and resolution of electro-optical/infrared (EO/IR) systems. Unlike traditional mapping lidars that require 3D point clouds and image processing, Atolla's system meets real-time identification demands, functioning effectively day and night with smaller optics. Originally designed for agricultural use in monitoring insect swarms, the SWAP-optimized lidar sensor and ML algorithm are being adapted for defense applications to extend range and improve precision in detecting and identifying small-UAS threats. This innovative method leverages frequency signature mapping to minimize data processing demands.Atolla Tech's system offers several technical objectives and innovative approaches. The Phase I objective is to design and build a field-ready prototype to determine its limitations and optimize it for real-time UAS detection and identification. Key tasks include specifying field prototype requirements, component sourcing, prototype design updates, building and calibrating the lidar sensor, configuring the system for optimal performance, and refining preprocessing algorithms. The goal is to achieve a working prototype that demonstrates the feasibility of using this technology for C-UAS applications. The prototype will address critical military needs by providing a reliable method for rapid UAS identification, enhancing situational awareness, and enabling timely responses to potential threats.This project also includes the development of an identification algorithm for drones, leveraging existing algorithms used for identifying flying insects. Modifications will be made to accommodate hardware changes and new waveform features specific to drones. The algorithm will analyze preprocessed data, extract key features such as propeller and resonant frequencies, and classify different objects (e.g., birds, drones) using ML or rule-based approaches. Experimental trials with various drones will validate system capabilities, ensuring accurate identification and detection range.By integrating these advancements, Atolla Tech's lidar-based C-UAS system will provide a robust solution for real-time UAS detection, classification, and identification, with potential applications in both military and civilian sectors. The project will contribute to the Army's "Defend the Force" C-UAS initiative by enhancing real-time detection, ensuring constant monitoring, and offering a cost-efficient alternative to traditional radar systems.