Adaptive Particle Swarm Optimization for through-foliage target detection with drone swarms

Abstract: This work contributes to efforts on autonomously detecting a vegetation-occluded target by airborne observers. It investigates and enhances previous work on a Particle Swarm Optimization (PSO) strategy for Airborne Optical Sectioning (AOS) drone swarms. First, it identifies two issues with that method and proposes to resolve them by a leader stabilization for its scattering and projection-based line positions for its default scanning pattern. Second, it connects this method to other PSO variants and presents a new adaptive PSO strategy for AOS drone swarms that draws on the ideas of Adaptive PSO (APSO).