Effective allocation of tasks in heterogeneousmulti-robot systems (MRS) is a pivotal challenge as real-world deployments increasinglyfeature diverse robotic agents with varied capabilities, resources, and mobilityconstraints. Achieving scalable, robust, and efficient collaboration in such environmentsis impeded by the combinatorial complexity of task allocation, the need for real-timeadaptability, and the possibility of dynamic changes such as agent failures or shiftingteam compositions. While centralized approaches can solve small-scale problems,they falter in large teams due to high computational and communication overhead,lack of resilience, and suboptimal adaptability to environmental changes or agentfaults. Recent work in coalition game theory, deep reinforcement learning (RL),and distributed architectures has made significant progress in these areas—offeringdecentralized, learning-driven methodologies for task allocation and scheduling.However, a fully unified framework that reliably addresses agent heterogeneity,asynchrony, coalition formation, and failure handling in a modular and generalizableway remains an open challenge.
In this work, we synthesize the state-of-the-art,drawing from coalition-game utility modeling, RL-based decentralized scheduling,asynchronous multi-agent RL, and ROS-enabled distributed architectures, to proposea unified decentralized framework for robust task allocation in heterogeneous multi-robotteams. Our architecture integrates decentralized policy engines, dynamic coalitionnegotiation, macro-action–based asynchrony, motivation-driven task reallocation,and communication middleware, allowing teams to self-organize, adapt, and recoverfrom failures without central control. This design enables scalable deployment inrealistic environments, supports online learning and adaptation to unforeseen tasksor conditions, and minimizes both resource contention and deadlocks through predictivecommitment and masking mechanisms. Planned simulation and physical experiments willvalidate our approach on metrics such as global task completion, resource efficiency,makespan, and fault tolerance. Our unified framework not only harmonizes the majoradvances in multi-robot task allocation literature but also positions itself asa practical blueprint for real-world heterogeneous MRS deployments in domains suchas disaster response, industrial automation, and exploration.