| dc.description.abstract | Abstract. Human–machine collaboration is an effective means to perform complex tasks in manufacturing. However, in distributed control systems that require temporal certainty, the temporal uncertainty in human–machine collaboration presents significant challenges for its practical implementation. Existing methods for uncertainty mitigation usually neglect the interdependencies between human–machine collaboration and other automated processes, leading to inaccurate estimation and poor handling of temporal uncertainty. To address this issue, a digital twin-enabled method for mitigating the temporal uncertainty is proposed. First, a digital twin model of a distributed control system is established, considering both human–machine collaboration and automated processes. On this basis, a digital twin-enhanced optimization module is then proposed to improve the iterative process of task allocation algorithms in human–machine collaboration assembly. Finally, a digital twin-driven supervisory control system is developed, capable of system-level mitigation of temporal uncertainties through holistic production coordination. The proposed method is validated through comparative experiments conducted in an experimental gearbox assembly system, demonstrating its capability to handle temporal uncertainty in human–machine collaboration. | |
