Pheromone-Based Approach for Scalable Task Allocation

The application and evolution of multi-agent algorithms to the domain of unmanned aerial vehicles (UAVs) or drones will enable swarms of heterogeneous agents with emergent behaviour that dynamically and optimally respond to changes in the environment and resources to complete heterogeneous tasks in multi-objective missions. A key enabler of this capability is task allocation, but the state of the art generally focus on optimality and do not scale well. We propose a general, bio-inspired pheromone-based approach that can be applied to existing task allocation algorithms to enable scalable, emergent behaviour and swarm intelligence.