Development of an Automated Container Depalletization System Using AMRs, Industrial Robots, and Vision Technologies

This thesis presents the development and evaluation of an automated solution for depalletizing boxes from shipping containers, carried out at SIR Robotics S.p.A. The proposed system integrates an Autonomous Mobile Robot (AMR), an industrial robotic manipulator, and an AI-based vision system to enable efficient and safe container unloading. The work was divided into three main phases. First, the AMR platform was tested and a strategy was developed to align the robot with the loading ramp and navigate safely inside the container. The second phase focused on evaluating different vision systems for box detection and localization. Finally, the picking procedure was validated by testing the grasping of boxes and their transfer onto a conveyor belt. The results demonstrate the feasibility of the proposed approach and provide a basis for the future development of fully automated container unloading systems.