Swarm Intelligence Algorithms in Drone Swarm Task Allocation: Comprehensive Survey and Future Directions

Drone Swarm applications have become increasingly diverse in military reconnaissance, material delivery, and disaster rescue. Task allocation, a key part of swarm execution, has been widely studied. Traditional exact or rule-based methods often struggle to balance solution quality and response time in large or uncertain drone swarm missions. Swarm intelligence (SI) algorithms offer a flexible population-based alternative through distributed search, parallel evaluation, and self-organized adaptation, although their advantages depend on mission structure and reporting conditions. This paper reviews recent literature on SI algorithms for drone swarm task allocation. It summarizes common task allocation models and algorithm families, distills a survey-level five-stage paradigm, and categorizes applications by environmental dynamics and platform homogeneity. The paper also discusses challenges related to dynamic replanning, heterogeneous collaboration, and resource communication constraints, while outlining future research directions for drone swarm task allocation.