The finals of the "Huazhan Logistics Cup" 17^th National University Student Transportation Technology Competition was held at Shanghai Maritime University on 10 July, 2022, with the theme of "digital intelligent transportation and low-carbon transportation". Aiming at specific problems in the transportation system, the participants are requested to utilize relevant professional knowledge, and create the optimal methods or solutions that are novel, feasible, practical with completion and certain degree of difficulty. Our FACTE students majoring in transportation participated in the competition and won two third prizes.

Project 1: A Model-Driven Intelligent Connected Platoon Generation Method and Visualization System

Instructors: Li Haijian, Wu Yiping

Team members: Chang Yujie, Zhang Yue, Yuan Jingyu, Wang Yunpeng

Project Overview: Based on the PERMIT-Plexe-SUMO co-simulation, this research proposes a vehicle platoon generation method and visualization system that can flexibly generate any form of platoon according to the set parameters in various road scenarios. The parameters can be set through visual interface, allowing the system to display platoon operation status in the visualization system, and output any vehicle data curve, as well statically save some or all of the data. For researchers, the system is conducive to testing the effect of parameter setting, and laying a technical basis for platoon generation in the future research on intelligent connected platoons.

Project 2: Identification and Early Warning Platform of Inland Water Navigation Risk Behaviour Based on AIS Data and Machine Learning Method

Instructors: Gan Shaojun, Wang Yanxia

Team members: Zhang Ziqi, Nie Wenchen, Tang Jiayi, Li Jiale, Jin Xi

Project Overview: This project builds a risk behaviour identification model to predict ships' behaviours in inland waterways based on the AIS data analysis. First, a single-chip microcomputer is adopted to build a remote AIS data acquisition platform to read the AIS data and send it to the cloud server through MQTT. Second, a data enhancement algorithm is proposed based on the generative adversarial network. Then, the time convolutional neural network algorithm is further introduced to construct the ship anomaly identification model. Last, a ship navigation status monitoring platform is built based on the B/S structure and the aforementioned model.

The excellent results are inseparable from the hard work of all students and the thoughtful instructions from teachers. Congratulations again to the winning students and teachers! The FACTE will continue to build a platform for practical innovation and talent education, cultivate the innovative competence of students, and enhance the construction of talent teams.