Energy Entropy Weighted Field-Based Collective Intelligence (EEW-FBCI) for Robust Decentralized Multi-Robot Task Allocation

Abstract

The problem of robust multi-robot task allocation under uncertain conditions is addressed in this paper. Uncertainty arises from unknown robot failures, stochastic motion, and dynamic environmental interactions. To address these challenges, we propose the EEW-FBCI method, which integrates field energy, entropy, and task-driven attraction to coordinate swarm robots. The method is evaluated in simulations with 50 robots performing coordinated tasks, including a 30% robot failure occurring at timestep 400. Validation metrics entropy, field energy, cluster consistency, coverage, and task reach demonstrate that the swarm maintains high coordination, stability, and adaptive behavior. In particular, the evolution of energy and entropy highlights the swarm’s ability to dynamically reorganize and recover after failures. The swarm achieves a final task reach of 0.94 and a high proportion of active robots (97.1%), demonstrating strong resilience. Comparative analysis shows that the proposed method outperforms conventional conservative and balanced strategies in task completion, coordination, and robustness to failures. These results highlight the potential of the method for real-world applications requiring failure-tolerant, adaptive, and self-organized swarm systems, where energy and entropy serve as key performance indicators.