Author: Jiajie HuPublisher:
ISBN:
Category : Automated vehicles
Languages : en
Pages : 143
Book Description
The connected vehicle is an emerging technology aimed at deploying and developing a fully connected transportation system which allows the vehicles to dynamically transmit messages between the vehicles (V2V), infrastructure (V2I), Cloud (V2C) and everything (V2X). The connected vehicles can provide an unprecedented amount of data even in the traffic network with a low market penetration rate, which can provide new solutions to transportation issues. This study focuses on micromodeling and quantitatively assessing the potential benefits of the connected vehicles on safety, mobility, and energy efficiency perspectives. In this dissertation, we proposed deep-learning based systems to solve different transportation problems under the environment of connected vehicles. The crash risk prediction system can identify crash-prone intersections and guide the deployment of safety measures to prevent potential crashes. The pothole detection system provides a cost-effective strategy to map the road conditions, which will be beneficial to road maintenance especially when municipal budgets are limited. The slippery condition surveillance system achieves real-time monitoring of pavement slippery conditions impacted by adverse weather and promotes cautious driving behaviors. The adaptive traffic signal control system provides an adaptive, efficient and optimized traffic signal control agent, which can reduce vehicle delay and emissions, improve mobility and energy efficiency. Overall, connected vehicle technology shows great potential in the field of transportation. The safety, mobility and energy efficiency will be further improved with the widespread deployment of connected vehicles and increase of market penetration rate, which is achievable in the near future.
