3차원 하중 맵을 이용한 지능형 화물적재 시스템의 설계

Abstract

As we enter the era of the Fourth Industrial Revolution, international trade and the quantity of goods transported between nations are on a significant upward trend. Among the existing transportation methods, mass maritime shipping utilizing containers, the most efficient means, is experiencing the most growth. Consequently, maritime transportation accidents are also on the rise. Cargo falling incidents, the most frequent type of maritime accidents, occur due to improper weight declarations for containers, resulting from overloading, imbalanced loading, and other factors. In this thesis, we propose an Intelligent Container Loading System using 3-D Weight Map, which employs load cells to monitor real-time cargo weight and weight distribution. This system not only displays cargo weight and weight distribution to the user but also implements an optimal loading and stability solution for the vessel. Presently, container loading and stowage on ships are conducted through simulations using software. However, in the case of improperly declared containers, it need to physically remove all cargo from the container for re-weighing incurs significant time and cost expenses. Moreover, simulations rely on weight data from documents rather than real-time weight data. The proposed approach in this thesis involves attaching weight sensors to the bottom of each container, allowing for the monitoring of weight distribution and real-time weight. This information is stored in a server database through an embedded system, and a rule-based system generates weight bias codes through a rule-based engine. As containers are stowed on the vessel, the weight bias codes assigned to each container are systematized into a matrix and undergo a compaction process to achieve weight distribution uniformity, resulting in a 3-D Weight Map. The embedded system is built using Python on Arduino-Mega and communication is facilitated through serial communication to demonstrate its practicality and efficiency. Additionally, we have developed a prototype of this system to demonstrate its practicality and efficiency.