Inhi Keeps closely working with Monash University as an adjunct senior lecturer until 2023. Kongju National University and Monash University will collaborate in teaching and research continuously.
NEWS (page 8 of 14)
Congrats Bo! The mid review has been successfully completed.
Title
Short-term traffic state estimation and prediction based on spatiotemporal neural networks
Summary
Spatiotemporal neural networks (NN) models have recently achieved competitive results for short-term traffic prediction and achieved outstanding outcomes. However, two problems still require further study in terms of model performance and inner mechanism understanding: (1). The forecasting model affected by many aspects like model inputs, model structure, external factors, and optimisation function. How to design an appropriate framework for short-term traffic state prediction? (2). The existing related studies mainly use the knowledge and advantage from the neural network field, but how to incorporate transport domain kn owledge with the above framework? Therefore, th e aims of this study are: (1) Presenting an overall framework from data management to model training for traffic network state estimation and prediction, which provides better forecasting results and APIs for other applications. (2) Understanding the relations of components inside the framework and improving it by integrating transport domain knowledge. The outcomes of this study provide a general framework of the NN-based traffic forecasting model for practice and a better understanding, which will benefit the short-term traffic forecasting related research and industry.
https://www.youtube.com/watch?v=ptKB_hsGMEY&feature=youtu.be
Research Group
Transport
Research Theme
Monitoring, Prediction and Protection
Summary
Traffic incident is one of the factors causing traffic congestion on the roads. The traffic management centres start accumulating incident data to manage the congestion. However, the incidents are not properly managed because many minor accidents happen to be unreported. Therefore, systematic ways for detecting and accumulating events data are necessary. This research aim is to introduce a novel way of obtaining incident-related data and to develop a robust detection algorithm to recognize incidents and its type simultaneously by machine learning.
Bio
I received the M.S degree in transportation engineer ing from Kongju National University, South Korea in 2017. I commenced Joint Ph.D. degree with Southeast University, China in 2018 under the supervision of Dr. Inhi Kim, Prof. Graham Currie and Prof. Zhibin Li. Research interests include intelligent transport systems, traffic simulation, virtual reality, driving simulator and deep learning analysis.
The presentation can be watched below;
Research Group
Transport
Research Theme
Resilience, Infrastructure and Society
Summary
Although many strategies and policies have been advocated, either from the demand or supply aspect, to mitigate the traffic congestion, still, the problem remains to exist. A new perspective to visit the causes and to tackle this problem is urged. Extensive studies have revealed that the built environment can highly be associated with determining human spatial activities, especially automobile travel behaviors. Through travel behaviors, lead to the generation of traffic, which has an impact on traffic operation and control action. However, how the built environment directly influences the traffic performance (congestions and delays) have been rarely studied and lack of evidence.
In this context, this research aims to establish an in-depth understanding of the impacts of the built environment on traffic congestion at different spatial scales. Trying to answer the questions as 1) What built environment features are most relevant to the change of traffic congestion? 2) To what extent the change of the built environment features (by considering geographical scale) causes a difference in the traffic congestion level. 3) How can urban mobility and accessibility be optimized and enhanced via manipulating the design of the built environment? To address the above issues, the objectives of this research are 1) To synthesize the critical built environment indicators that hold accountable for the differences in traffic performance between areas; 2) To quantify the relationship of the built environment in raising the traffic congestion at different spatial scales; 3) To inform the built environment policy framework towards optimized transport planning and management .
By knowing this relationship, the outcome of this research can help guide the design of the urban/transportation planning policy for urban sprawl controls and mobility enhancement from the beginning design of the land use and the infrastructure implementation plaN.
For his presentation the video is below
Big congrats. Lilian has successfully passed the progress review with outstanding marks.
Please join to watch her seminar.
I am pleased that TUPA has made good outcomes at the 99th TRB conference.
- Inferring The Optimal Number Of Dockless Shared Bike In A New Area By Applying The Gradient Boosting Decision Tree Model by Dong Xiao, Tianqi Gu, Yuanqiu Bao, and Inhi Kim
- The Use of Emerging Virtual Reality Technology in Road Safety Analysis: The Hook-Turn Case by Taeho Oh, Yanping Xu, Zhibin Li, and Inhi Kim
- Short-Term Traffic Prediction Using A Spatial-Temporal CNN Model With Transfer Learning by Wang Bo, Hai, Vu, and Inhi Kim
Congratulation on Bo’s confirmation. All the panels and the chair impressed Bo’s research progress. The presentation was also very comprehensive. If anyone needs the seminar in a video file please contact me. Well done Bo!!
|
Title: Short-term traffic state estimation and prediction based on spatiotemporal neural networks Research Theme: Monitoring, Prediction and Protection Group: Transport Summary: Spatiotemporal neural networks (NN) models have recently achieved competitive results for short-term traffic prediction and achieved outstanding outcomes. However, two problems still require further study in terms of model performance and inner mechanism understanding: (1). The forecasting model affected by many aspects like model inputs, model structure, external factors, and optimisation function. How to design an appropriate framework for short-term traffic state prediction? (2). The existing related studies mainly use the knowledge and advantage from the neural network field, but how to incorporate transport domain knowledge with the above framework? Therefore, th e aims of this study are: (1) Presenting an overall framework from data management to model training for traffic network state estimation and prediction, which provides better forecasting results and APIs for other applications. (2) Understanding the relations of components inside the framework and improving it by integrating transport domain knowledge. The preliminary works of this study focus on the inputs and structure of NN-based model, which used the datasets of bike-sharing traffic network (New York and Suzhou cities) and highway traffic network (PeMS – Caltrans Performance Measurement System). The main outcomes are (1). The impact and inner relation of the external factors (discrete variables like weather, POI, and holidays) are studied. (2). A more accurate forecasting model based on 3D residual NN is presented, which learning the spatial-temporal features and being trained with traffic correlated input data with temporal autocorrelation. (3). The advantage of model fine-tun ing (a technique of transfer learning in the NN field) is studied with PeMS dataset, which has improved the model with limited training data. In the future work, more complex input and output data (network level) will be considered. Since traffic flow, density, and speed are essential factors in traffic flow theory; therefore, further studies will aim to integrate more domain knowledge to guide the model training. Transportation is fundamental to a thriving society. The results of short-term traffic forecasting affect both decision support of city planning and traffic management. The outcomes of this study provide a general framework of the NN-based traffic forecasting model for practice and a better understanding, which will benefit the short-term traffic forecasting related research and industry.
Bio: Bo Wang received the double M.S degree in transportation engineering from Monash University and Southeast University in 2018. He commenced his PhD at Monash University in 2018 under the supervision of Dr. Inhi Kim, Prof. Hai Vu and Dr. Chen Cai. His research interests include intelligent transport systems, deep neural networks and big data mining in transportation. |
I am pleased to inform you that my phd student, Tianqi Gu won the best phd research award from the 1st Joint Workshop held in Xian-Liverpool University, Suzhou. Also our phd candidates (Taeho and Dong) made a wonderful presentation today.
Dr. Mike Ma gave an invited talk to the audience about smart operation of public transportation.
