Development of a cloud-based service framework for energy conservation in a sustainable intelligent transportation system

The research aims to develop a cloud-based service framework for reducing carbon dioxide emission and fuel consumption in intelligent transportation system. It collects traffic condition, driving behavior, and video through telematics and digital tachygraphy and road-side cameras to facilitate advanced data analytics for the reduction of fuel consumption. There are three specific features regarding this framework. First, a transportation cloud is built for the storage of massive data and video. This cloud-based system not only avoids the use of hard disks at client-site for energy conservation and reliability improvement, but also allows the back-end data analytics at both server and client sites. Second, a real-time traffic condition analytic was developed by mobile machine vision techniques based on video and data collected from road-side cameras to analyze and recognize traffic conditions, such as traffic flow, braking events, traffic lights, and count-down timers. Then, a fuel-efficient route navigation technology is also developed for eco-driving based on real time traffic information and a dynamic shortest path algorithm for saving time and fuel consumption. Third, a sequential pattern mining model was proposed to diagnose misguided driving behavior for eco-driving based on the real-time data collected from digital tachygraphy and on-board diagnostics system. Furthermore, an e-Learning visualization system was developed to provide advice and instruction for correction of misguided driving behavior. Indeed, the fuel consumption and power consumption can be reduced simultaneously based on the proposed framework regarding cloud-based system and eco-driving.