Local ant system for allocating robot swarms to time-constrained tasks

We propose a novel application of the Ant Colony Optimization algorithm to efficiently allocate a swarm of homogeneous robots to a set of tasks that need to be accomplished by specific deadlines. We exploit the local communication between robots to periodically evaluate the quality of the allocation solutions, and agents select independently among the high-quality alternatives. The evaluation is performed using pheromone trails to favor allocations which minimize the execution time of the tasks. Our approach is validated in both static and dynamic environments (i.e. the task availability changes over time) using different sets of physics-based simulations.