Distributed multi-robot task dynamic allocation in digital-twin factory towards industry 5.0

Task allocation is a critical component in multi-robot manufacturing systems (MRMS), significantly affecting operational efficiency and system resilience. Within the framework of industry 5.0, which emphasises decentralised autonomous manufacturing and sustainability, traditional static and centralised task allocation methods prove inadequate for the dynamic and complex demands of contemporary manufacturing environments. To overcome these limitations, this study introduces a distributed multi-robot task dynamic allocation method for digital twin factories. We develop a digital twin-driven robot model, coupled with the task dynamic allocation process to facilitate real-time monitoring and anomaly resolution. To derive high-quality and scalable solutions within a decentralised framework, the Alternating Direction Method of Multipliers (ADMM) and Augmented Lagrangian Coordination (ALC) methods are employed to analyze and resolve task allocation challenges. The proposed methodology enhances system configurability and resilience, aligning with the sustainability objectives of industry 5.0. The efficacy and efficiency of the proposed method are demonstrated through a practical case study.