Topology-based Launch Collision Avoidance augmented with Differential Evolution

Abstract

This work enhances the Launch COLlision Avoidance (LCOLA) tool ASSET (Ascent SafETy) developed at DLR/GSOC by introducing a continuous, topology-based methodology for conjunction detection across full launch windows. A two-dimensional time of closest approach search, formulated in the domain of liftoff time and mission elapsed time, and solved via differential evolution optimization, enables continuous liftoff coverage, and accurate launch window closures identification based on probability of collision and miss distance. Validation on a representative mission demonstrates full identification of risky conjunctions while reducing computational cost by two orders of magnitude compared to the discrete approach, significantly improving operational responsiveness and launch safety.