Developing a Digital Twin Framework for Monitoring the Trajectory of the UR10 Robot Using an Extended Kalman Filter

In the era of modern robotics and Industry 4.0, integrating advanced state estimation techniques with digital twin technology is crucial for enhancing robotic system performance and safety. This study presents a pioneering Digital Twin framework to monitor the trajectory of Universal Robots' UR10 robotic arm. We address challenges such as system uncertainties, sensor noise, and environmental factors by combining a Kalman Filter-based state estimation with a Digital Twin approach. By creating a virtual replica that accurately mimics the UR10's physical properties and dynamics, our method enables realtime tracking and prediction of the robot's states, improving control and operational efficiency. The Digital Twin incorporates the robot's kinematics and dynamics to compute the Jacobian matrix for EKF estimation. Through extensive simulations, we demonstrate a significant increase in accuracy in position (approximately 90.14%) and velocity (approximately 85.10%) estimations after introducing noise to the UR10 robot simulation. This approach paves the way for advanced control strategies, fault detection systems, and virtual testing platforms in industries like manufacturing, healthcare, and research, where precision and performance are critical.