Dynamic Multi-Robot Task Allocation under Uncertainty and Communication Constraints: A Game-Theoretic Approach

A research team from UC Berkeley and other institutions has published a breakthrough paper on arXiv titled "Dynamic Multi-Robot Task Allocation under Uncertainty and Communication Constraints: A Game-Theoretic Approach." The paper introduces Iterative Best Response (IBR), a decentralized algorithm that enables fleets of up to 100 drones to efficiently allocate tasks in real-time despite communication limitations and uncertain task completion times. The system models incomplete information through hub-based sensing regions and communication graphs, allowing each agent to select tasks that maximize local welfare contributions.

In extensive simulations of city-scale package delivery scenarios, IBR outperformed three established baselines: Earliest Due Date first (EDD), the Hungarian algorithm, and Stochastic Conflict-Based Allocation (SCoBA). The algorithm maintained competitive task completion performance while reducing computation time by approximately 30% compared to centralized approaches. This efficiency gain persisted even under sparse communication conditions where traditional methods typically degrade.

The researchers tested their approach across varying task arrival scenarios, demonstrating robustness against uncertainty in task durations and deadlines. The game-theoretic framework allows drones to make independent decisions based on locally available information, eliminating the need for constant central coordination. This makes the system particularly valuable for applications like emergency response, logistics, and infrastructure inspection where communication networks may be unreliable or intentionally limited.