Orbital debris selection by means of genetic algorithm optimization

Abstract

This thesis details development methodologies and code implementation for Obsolete Spacecraft Capture and Removal (OSCaR) cube satellite target selection algorithm. The algorithm selects a set of objects ideal for OSCaR to capture and deorbit. This thesis describes the formation of the algorithm itself and its related functions: population member fitness evaluation, nonlinear constraint function, and initial data set generation. The optimization method, MATLAB's genetic algorithm, maximizes the fitness function and returns a specified number of objects in the order OSCaR captures the set. The algorithm also returns the remaining propellant budget and other data that may be interesting to the user. The algorithm consistently converges to the best fitness value for up to three targeted objects.