Date of Award

6-2026

Degree Name

MS in Aerospace Engineering

Department/Program

Aerospace Engineering

College

College of Engineering

Advisor

Madhusudan Vijayakumar

Advisor Department

Aerospace Engineering

Advisor College

College of Engineering

Abstract

During celestial body observation mission planning, predicted sensor coverage is a key analysis tool used by designers to inform concept of operations. It enables trade studies by applying a metric to assess how well different sensor, satellite, and constellation configurations can observe a mission's region of interest. The effectiveness of the studied components is then paramount in determining whether they are justified in the system architecture. Coverage computation becomes more difficult as sensor types and regions of interest (RoIs) deviate from simple shapes, and as problem formulations seek higher-fidelity results. These simplifications include: conical camera sensors, RoIs represented as simple, low-vertex geometry in lat/lon coordinates, and analytical representations of satellite access times for a ground station. As formulation parameters become more complex, traditional techniques struggle or compensate for this complexity by increasing processing time. These techniques often rely on point grids, cells, envelopes, or otherwise simplifying assumptions; either pushing the technique towards quick, less accurate results, or slow, more accurate ones. This thesis develops a polygon-based coverage analysis method, where the swath of a sensor and the regions of interest are represented using polygons. This allows the mission design engineer to represent the polygonal shapes to any desired level of accuracy. The area of coverage is represented as the physical overlap of these surface polygons, which are projected onto Euclidean planes using an octant-based gnomonic projection to simplify the intersection process. An area-recovery method is proposed, which estimates ellipsoidal surface area from projected polygons, tying the polygon method together as a viable option for area-recovery metrics for sensors. The method supports continuous, discontinuous, and global regions of interest and can account for redundant coverage among sensors, satellites, and satellite constellations. The proposed method was compared to commercially available coverage computation software for performance evaluation. The comparison yielded similar coverage trends, while the proposed method offered significant computational improvement for global coverage computation. Additionally, the proposed polygon approach is combined with a genetic algorithm solver to optimize sensor pointing, enhancing coverage metrics for a given combination of sensor/satellite/constellation and a region of interest.

Share

COinS