Near Real-Time Optimal Trajectory Design for Active Debris Removal With Verification and Validation Using KOMPSAT-1

Geospace is getting busier with an increasing amount of debris, and their removal of such debris is becoming important for the safety, efficiency, and sustainability of geospace. The need for safe and sustainable removal has drawn significant attention to Active Debris Removal (ADR). When a chaser performs a fly-around maneuver to approach a target, it should solve the following optimization problem (OPT): (i) Minimize collision risk against other flying space objects, (ii) minimize fuel consumption, and (iii) maximize mission accomplishment. However, solving OPT is known to be very challenging, as it is at least NP-hard. This is because the trajectories of chasers are usually helical with their spines following targets’ Keplerian orbits to collect information about morphology, mass, attitude, etc. using optical sensors. In addition, the proximity among the O(10^4) fast-moving objects in the current space catalogue makes the solution process more challenging. Even worse, it is necessary to iteratively solve OPT because the chaser's contact strategy with the target can change as more information is collected. This implies that an efficient solution process of OPT is required despite its NP-hardness. This study introduces a near real-time algorithm for generating a reasonably good solution for OPT, i.e., a near-optimal trajectory of a chaser. Our algorithm is Generate-&-Test (G&T), i.e., we generate thousands of candidate trajectories and evaluate them as quickly as possible using two measures of collision risk and fuel consumption. We use a heuristic method to find optimality in the vast solution space. In this entire process, the efficiency and accuracy of the key computation are achieved by using the dynamic Voronoi diagram (DVD). As verification and validation, we present a case study of the ADR mission for the South Korean satellite KOMPSAT-1 that ended its operations in 2008.