A 5-Dimension Digital Twin Framework for Collision Detection in Open PLC-Based Collaborative Robots

Digital Twin (DT) technology has evolved from passive visualization into a proactive 5-dimensional (5D) framework for enhancing the safety and autonomy of robotic systems. This paper implements a 5D DT to expand the capabilities of a collaborative 6-DOF manipulator controlled by an open PLC-based controller. While such open architectures offer high flexibility, they often lack the integrated high-level safety layers found in proprietary systems. To address this, we developed a bidirectional connection using the TwinCAT ADS protocol, achieving a stable 2.5 ms latency for real-time synchronization between the physical and virtual entities. To ensure the real-time feasibility of the safety services, the virtual model’s collision geometry is optimized by replacing detailed STL meshes with simplified geometric primitives, such as cylinders and boxes. This optimization significantly reduces computational load, enabling high-frequency collision checks for both self-collision and environmental obstacles. Furthermore, a trajectory planning service based on an optimized Rapidly-exploring Random Tree (RRT) algorithm is implemented to generate efficient, collision-free paths by reducing redundant waypoints. The results validate the effectiveness of the 5D framework in acting as a proactive safety supervisor, enhancing the operational capabilities of open-controller robotic systems in industrial environments.