Enhancing FAA Counter UAS Detection and Mitigation Efforts

Published by Melissa Rodriguez on

Take Our Content to Go

Introduction 

Mosaic ATM’s research on UAS integration into the National Airspace System (NAS) has been at the leading edge since our founding in 2004. In this case study, Mosaic supports the Federal Aviation Administration’s (FAA) Counter UAS Detection and Mitigation project, a critical initiative aimed at enhancing the safety and security of airspace around sensitive locations such as launch sites and airports. The project evaluates the performance of diverse technologies submitted by vendors to develop specifications for future solutions. Through meticulous data processing and analysis, Mosaic ATM enables the FAA to assess the effectiveness of these technologies in detecting and tracking unauthorized drone activity. 

Challenge 

The FAA’s Counter UAS initiative faces complex challenges in standardizing, processing, and analyzing data from multiple vendors. The data collected from various drone detection technologies is often noisy, lacks uniformity, and requires substantial processing to yield actionable insights. Key challenges included: 

  • Standardization: Diverse data formats and inconsistent reporting fields required normalization into a single format for analysis. 
  • Accuracy Assessment: Vendor data had to be rigorously compared with truth data from controlled drone flights. 
  • Complex Analyses: Evaluations needed to determine drone compliance with regulations such as altitude limits (Above Ground Level), line-of-sight constraints, and daytime operation requirements. 
  • Frequent Changes: Requirements and project scope evolved as the team uncovered new insights during the data exploration process. 

Solution 

Mosaic ATM’s team of analysts provided robust data processing and analysis support to address these challenges, ensuring reliable and insightful evaluations for the FAA. This process included: 

Data Standardization and Cleaning 

  • Mosaic analysts normalized data across vendors into a single format with consistent fields suitable for analysis. 
  • Noise was filtered out, and data without positional information was excluded to maintain integrity. 

Altitude Conversion and Validation 

  • Altitude data reported in Above Mean Sea Level (AMSL) or Above Ellipsoid (AE) was converted to Above Ground Level (AGL) using NASA’s TERRA dataset. 
  • AGL data was compared with FAA’s Low Altitude Authorization and Notification Capability (LAANC) tiles to identify over-ceiling operations. 

Custom Software Development 

  • Custom algorithms were built to evaluate drone detections within airport runway obstruction surfaces. 
  • Data in spherical coordinates was converted to linear distances to determine if drones exceeded the visual line-of-sight limit (approximately 2,000 ft). 
  • Local sunrise and sunset data was incorporated to identify nighttime drone operations. 

Experimental Data Validation 

  • Truth data for experimental flights was established using on-board GPS or UAS decoding technologies. 
  • Cross-track, along-track, and vertical errors were analyzed to assess the accuracy of vendor detections. 
  • A custom hit rate analysis based on velocity and altitude metrics evaluated technology coverage. 

Visualization and Reporting 

  • Results were summarized in user-friendly tables and charts to support decision-making by non-technical stakeholders. 
  • Mosaic analysts maintained agility, adapting analyses to evolving requirements while anticipating customer needs. 

Results 

Mosaic ATM’s contributions enabled the FAA to: 

  • Establish a robust baseline for evaluating drone detection technologies. 
  • Quantify vendor performance in detecting unauthorized drone activity under various scenarios. 
  • Develop comprehensive specifications and recommendations for future detection and mitigation solutions. 
  • Present findings in a clear and actionable format, facilitating effective communication with stakeholders. 

Conclusion 

Through expert data analysis and software development, Mosaic ATM played a pivotal role in advancing the FAA’s Counter UAS Detection and Mitigation initiative. By normalizing data, validating vendor performance, and delivering actionable insights, Mosaic supports the FAA’s mission to safeguard sensitive airspace against unauthorized drone activity. This partnership highlights Mosaic’s ability to navigate complex data challenges and deliver innovative solutions that enhance airspace safety and security. 

Categories: Case Studies

Related Posts

Case Studies

Explanations in Lunar Surface Exploration (ELSE) for Machine Learning Models to Assist in NASA Rover Exploration

Mosaic ATM is supporting NASA by developing explainable machine learning models to assist with lunar rover exploration.

Case Studies

ATD-2 Machine Learning Model Validation Framework Design

To realize the full power of machine learning, organizations need to focus on operations and delivery of the predictions as much as the development of the algorithm itself. In the case of NASA and ATD-2, building trust in the machine learning predictions is essential to expanding their use to improve inefficient operations and unused capacity. Air traffic controllers must trust the recommendations presented by them, and validation is essential towards building trust.

Case Studies

Taxi Time Predictions

The taxi-out time predictions help pilots decide when to "single-engine taxi": taxi most of the way to the runway with only a single-engine turned on, turning on the second engine just a few minutes before take-off. The single-engine taxi decision is typically made by pilots within an hour of push back. Still, our customer asked for predictions up to four hours in advance of the expected push back time.