Leveraging digital twin for autonomous docking of a container truck with stabilization control

This paper describes the design, development, and implementation of a high-precision autonomous docking control system for a container truck based on a digital twin approach. The digital twin is used to simulate the dynamic behavior of the physical truck and to design the controller by providing a virtual platform to test, validate, and optimize control strategies and algorithms before their deployment in the actual system. To this end, a cascade of a nonlinear observer and an unscented Kalman filter is used to estimate the state variables of the physical truck for point-stabilization and orientation controls during the autonomous docking process. The docking motion involves two stabilization problems: point stabilization for smooth motion from the initial configuration to the docking slot, and orientation control to deliver the container truck to the final docking position with a margin of error of 5 cm for position and 0.0087 rad for orientation. The stability of both controllers is investigated, and simulations and experiments are conducted to demonstrate the accuracy of the proposed method in a container terminal environment.