An Artificial Intelligence Platform for Network-wide Congestion Detection and Prediction Using Multi-Source Data
-
2019-06-01
-
Details:
-
Creators:
-
Corporate Creators:
-
Corporate Contributors:
-
Subject/TRT Terms:
-
Resource Type:
-
Geographical Coverage:
-
Corporate Publisher:
-
Abstract:The advancement of new smart traffic sensing, mobile communication, and artificial intelligence technologies has greatly stimulated the growth of transportation data. A great deal of generated transportation data is playing an important role in modern smart transportation and smart city research and applications. The increase of computation power enabled by advanced hardware and the rise of artificial intelligence (AI) technologies, especially in the deep learning field, provides great opportunities to comprehensively utilize the transportation big data. When applying AI in the transportation area, transportation domain knowledge is beneficial for designing AI models and solving transportation problems in a smarter way. However, because most AI algorithms were not originally designed for transportation problems, using big data and AI technologies to solve transportation problems is facing challenges. Since key hyper-parameters are missed in some proposed AI models and the software and hardware adopted by various studies are different, many proposed AI-based methods can hardly be accurately re-implemented. Further, inmost of the AI-based transportation research studies, there is no uniform dataset to evaluate the proposed models. Thus, to overcome the challenges mentioned earlier, this project seeks to build a transportation AI platform with widely accepted datasets, provide well-established models, and use uniform training and testing procedures to assist the evaluation of emerging novel methodologies. Traffic forecasting involving high-dimensional spatiotemporal data is a good applicable scenario to utilize novel deep learning models to solve complicated transportation problems. Thus, in this project, the prototype platform mainly focuses on solving the traffic prediction problem. More specifically, the major contributions of this project include: Develop a prototypical artificial intelligence platform for solving challenging transportation problems, which require large-volume high-dimensional transportation data and complex models. This AI platform is capable of providing standardized datasets and novel deep learning-based models for specific problems; Design a novel architecture for the transportation AI platform to enhance the efficiency of the transportation data processing, management, and communication and increase the computational power of the platform; Design a data storage and management schema to manage multiple network-wide traffic datasets for supporting the traffic prediction task and simplify the whole training and testing process; Develop multiple deep learning-based models for solving the network-wide traffic prediction problem, which can be the baseline models to evaluate novel deep learning-based models. The developed transportation AI platform is capable of evaluating the traffic prediction performance of various implemented models by comparing and visualizing the prediction results tested on multiple real-world network-wide traffic state datasets. Future work will focus on increasing the computation capability of the platform and broadening the topics that the platform can deal with.
-
Format:
-
Funding:
-
Collection(s):
-
Main Document Checksum:
-
Download URL:
-
File Type: