Bio-Inspired Swarm Intelligence for Decentralized UAV Fleet Coordination

Abstract

Coordinating fleets of unmanned aerial vehicles without a central controller is increasingly framed as a bio-inspired
swarm-intelligence problem, drawing on decades of research into how flocks, ant colonies, and bee colonies achieve robust,
scalable collective behavior from purely local interaction rules. This article synthesizes that literature as it applies specifically
to decentralized UAV fleet coordination, tracing the line from Reynolds's boids model and stigmergic self-organization through
population-based metaheuristics such as particle swarm optimization, ant colony optimization, grey wolf optimizer, and
several more recent animal-inspired variants, to the consensus, potential-field, and distributed-auction theory that turns a
bio-inspired optimizer into a genuinely decentralized fleet behavior. We then examine how these two threads combine in recent
applied architectures for multi-UAV path planning, cooperative coverage search, and heterogeneous task allocation, and in the
communication and routing substrate any of this coordination depends on. Read together, the evidence supports bio-inspired,
layered swarm architectures as a coherent and increasingly well-evidenced design pattern for decentralized UAV fleets, while
leaving field validation under contested communication, adversarial interference, and genuinely heterogeneous vehicle
capability as substantially open research questions rather than solved engineering problems