Digital Twin-Based Autonomous Navigation and Control of Omnidirectional Mobile Robots

Abstract

In this paper, an autonomous navigation and control method is proposed for an omnidirectional mobile robot (OMR) based on digital twin (DT). An OMR-based DT system is designed from four dimensions, i.e., physical space, virtual space, application service and data processing. By the DT system, the autonomous navigation and control method is implemented from two steps: Autonomous navigation in virtual space and physical-virtual motion synchronization. The autonomous navigation of a virtual OMR is completed by fusion of an Astar algorithm and a dynamic windows approach in virtual environment. After obtaining an optimal navigation trajectory in virtual space, a physical OMR is driven by a physical-virtual motion synchronization method. Integral sliding mode controllers are designed to enhance robustness of the physical-virtual motion synchronization such that velocity tracking errors between the physical OMR and the virtual OMR are convergent. Experiment results show effectiveness of the DT-based autonomous navigation and control method by a physical-virtual synchronization tracking error 0.061 m and capability to handle various tasks.