Cooperative Learning and Optimization in Multi-Robot Systems

Multi-robot systems are increasingly deployed in complex, large-scale, and safety-critical environments such as environmental monitoring, infrastructure inspection, warehouse automation, and disaster response. In these settings, centralized control and learning paradigms often become impractical due to scalability limits, communication constraints, single points of failure, and privacy or robustness concerns. This has driven growing interest in distributed learning and optimization as foundational methodologies for enabling scalable, resilient, and adaptive multi-robot autonomy. The integration of learning with distributed decision-making presents unique challenges in robotic systems. Multi-robot environments are inherently non-stationary, partially observable, and subject to communication delays, packet loss, and topology changes. Robots may have heterogeneous sensing, actuation, and computational capabilities, further complicating coordination. Addressing these challenges requires new algorithmic frameworks that combine learning, optimization, and control while ensuring robustness, safety, and real-time performance.The goal of this Research Topic is to advance the theoretical foundations, algorithmic methods, and practical deployment of distributed learning and optimization in multi-robot systems, enabling teams of robots to operate autonomously, safely, and efficiently in large-scale, uncertain, and communication-constrained environments. Specifically, this Research Topic aims to bring together contributions that bridge the gap between principled distributed optimization and data-driven multi-agent learning, addressing the challenges posed by decentralization, heterogeneity, limited and unreliable communication, and dynamic interaction topologies. By unifying perspectives from robotics, control, optimization, and artificial intelligence, the Topic seeks to promote methods that are not only adaptive and scalable, but also interpretable, robust, and amenable to performance and safety guarantees.This Research Topic focuses on distributed learning and optimization methodologies for multi-robot systems, with emphasis on approaches that are theoretically grounded, communication-aware, and validated under realistic robotic constraints. We welcome contributions that advance both foundational theory and practical deployment, and that explicitly address the challenges arising from decentralization, uncertainty, and system-level complexity. Contributors are encouraged to address (but are not limited to) the following themes: • Cooperative Robotics • Distributed and Decentralized Optimization • Distributed and Multi-Agent Learning • Learning–Optimization–Control Integration • Robustness, Safety, and Resilience • Decentralized and distributed reinforcement learning • Coordination among heterogeneous robots (sensing, actuation, computation) • Applications including exploration, mapping, inspection, manipulation, logistics, and swarm robotics We welcome the following manuscript types: • Original Research Articles presenting novel theoretical results, algorithms, or system-level implementations • Methods Papers introducing new distributed learning or optimization frameworks with clear experimental validation • Review and Survey Articles that synthesize recent advances, identify open challenges, and outline future research directions • Perspective or Conceptual Articles proposing emerging paradigms, unifying viewpoints, or research roadmaps • Brief Research Reports describing timely, focused contributions or early-stage but impactful results Topic editors: Zhongguo Li, Yunda Yan, Zuo Wang, Xue-Fang Wang Source: https://www.frontiersin.org/research-topics/78862/cooperative-learning-and-optimization-in-multi-robot-systems