Federated Deep Reinforcement Learning-Driven O-RAN for Automatic Multirobot Reconfiguration

2025-06-01

AI Summary

To address the need for reliable and efficient communication among multi-robot systems in dynamic industrial 4.0 smart factories, this paper proposes a zero-touch O-RAN architecture embedded with Federated Deep Reinforcement Learning (FedDRL). By integrating FedDRL into the O-RAN control plane for the first time, the approach enables decentralized, distributed decision-making and dynamic reconfiguration of transmission parameters across edge nodes, jointly optimizing wireless resource allocation and energy efficiency while preserving data privacy. Compared to standalone DRL baselines, the proposed method achieves a 12% increase in system throughput, a 32% improvement in normalized average energy efficiency, and a 28% reduction in average transmission energy consumption. This work overcomes key limitations of conventional DRL in industrial settings—namely slow convergence, low energy efficiency, and weak privacy protection—and establishes a scalable, robust, autonomous optimization paradigm for multi-robot cooperative communication in smart factories.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Intelligent Robots: Learning & Optimization for ROB
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Federated Web and WoT systems, including distributed, federated and edge-based data processing
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

The rapid evolution of Industry 4.0 has led to the emergence of smart factories, where multirobot system autonomously operates to enhance productivity, reduce operational costs, and improve system adaptability. However, maintaining reliable and efficient network operations in these dynamic and complex environments requires advanced automation mechanisms. This study presents a zero-touch network platform that integrates a hierarchical Open Radio Access Network (O-RAN) architecture, enabling the seamless incorporation of advanced machine learning algorithms and dynamic management of communication and computational resources, while ensuring uninterrupted connectivity with multirobot system. Leveraging this adaptability, the platform utilizes federated deep reinforcement learning (FedDRL) to enable distributed decision-making across multiple learning agents, facilitating the adaptive parameter reconfiguration of transmitters (i.e., multirobot system) to optimize long-term system throughput and transmission energy efficiency. Simulation results demonstrate that within the proposed O-RAN-enabled zero-touch network platform, FedDRL achieves a 12% increase in system throughput, a 32% improvement in normalized average transmission energy efficiency, and a 28% reduction in average transmission energy consumption compared to baseline methods such as independent DRL.

Problem

• Optimizing multirobot system reconfiguration for smart factories
• Enhancing network reliability in dynamic industrial environments
• Improving throughput and energy efficiency via federated learning

Innovation

• Federated deep reinforcement learning for distributed decisions
• O-RAN architecture for dynamic resource management
• Zero-touch network platform for autonomous reconfiguration