Autonomous Driving on Container Terminals

Terminals play an important role in the process of transporting containers from origins to destinations all over the world. Recent developments have led many terminals to explore the opportunities of automation, specifically those offered by Automated Guided Vehicles (AGV’s). So far, AGV systems have primarily used central routing and driving methods to control movements of these vehicles. The aim of the system’s strategy is to prevent both collisions and deadlocks and to maximize container handling speed and efficiency. This thesis investigates the possibilities of autonomous AGV’s, which are responsible for making their own driving decisions and for coordinating with other vehicles around to prevent collisions and deadlocks. A decentralised strategy is developed and tested extensively by simulation experiments, the results of which show a performance that is roughly equal to that of a centralised strategy adopted by TBA’s TIMESquare software.