Electro-Optical and Infrared Design for Uncrewed Aerial System Collision Avoidance

Abstract

In recent decades, uncrewed aerial vehicle (UAV) technology has undergone repeated rapid advancements, enabling novel applications in civil, commercial, and scientific projects. Implementing UAV operations in each of these contexts requires the development of thorough safety systems for traffic management and collision avoidance. This is particularly true for beyond visual line-of-sight (BVLOS) operations, which require automated Detect and Avoid (DAA) Systems to recognize collision threats and direct avoidance maneuvers when necessary. EO/IR sensors are commonly employed by DAA systems, providing a low-weight and low-power means for capturing high resolution imagery of surrounding airspace. This thesis begins with a review of the history of the aircraft collision avoidance problem followed by an introduction to EO/IR system design. The historical review provides relevant regulatory and technological background for optical engineers who may be unaware of the context surrounding DAA technology. The technical review, then, provides DAA researchers who may be more familiar with other system components or sensing modalities with the necessary optical background to appreciate the contribution of the EO/IR study. After this introduction, plane-to-sky contrast is compared in the VIS (0.4 to 0.7 µm), SWIR (1 to 1.7 µm), MWIR (3 to 5 µm), and LWIR (8 to 14 µm) to determine which band is most sensitive to a C182 general aviation aircraft signal against a clear sky background. Contrast in the two reflective bands (VIS and SWIR) is determined in terms of equivalent reflectivity, and contrast in the two emissive bands (MWIR and LWIR) is determined in terms of equivalent blackbody temperature. Sensitivity data is then used alongside resolution specifications to estimate detection range performance using the Night Vision Integrated Performance Model (NVIPM). Results are extrapolated to a maritime atmosphere using the MODerate resolution atmospheric TRANsmission model (MODTRAN). The analysis of contrast, detection range-performance, and recommendations on band selection are provided for reference in the design of EO/IR systems for aircraft collision avoidance. Future research may study band performance at night, against other backgrounds (e.g. clouds, ocean, ground terrain), and over different elevation angles (±15° degrees).