Near-optimal real-time spacecraft guidance and control using harmonic potential functions and a modified RRT

Abstract

A primary requirement for any rendezvous and proximity (RPO) guidance algo- rithm is to ensure mission safety through the generation of collision-free trajec- tories in a fuel-efficient manner. This work presents a real-time hybrid guidance method which fuses the flexibility and robustness of Harmonic Potential Functions with the asymptotically-optimal Rapidly-expanding Random Tree Star method. The proposed method allows to plan trajectories on cluttered environments while producing near-fuel-optimal trajectories. To quantify and validate the performance of this method an experimental campaign is performed utilizing the Naval Post- graduate School POSEIDYN test bed. Lastly, implementation considerations and experimental results are discussed.