7 October 2020 in IEEE Transactions on Intelligent Transportation Systems
Accurate and timely traffic flow prediction is crucial for intelligent transportation systems (ITS). Recent advances in graph-based neural networks have achieved promising prediction results. However, some challenges remain, especially regarding graph construction and the time complexity of models. In this paper, we propose a multi-stream feature fusion approach to extract and integrate rich features from traffic data and leverage a data-driven adjacent matrix instead of the distance-based matrix to construct graphs. We calculate the Spearman rank correlation coefficient between monitor stations to obtain the initial adjacent matrix and fine-tune it while training. As to the model, we construct a multi-stream feature fusion block (MFFB) module, which includes a three-channel network and the soft-attention mechanism. The three-channel networks are graph convolutional neural network (GCN), gated recurrent unit (GRU) and fully connected neural network (FNN), which are used to extract spatial, temporal and other features, respectively. The soft-attention mechanism is utilized to integrate the obtained features. The MFFB modules are stacked, and a fully connected layer and a convolutional layer are used to make predictions. We conduct experiments on two real-world traffic prediction tasks and verify that our proposed approach outperforms the state-of-the-art methods within an acceptable time complexity.
24 April 2023 in IEEE Transactions on Intelligent Vehicles
Reducing vehicle exhaust pollution and energy consumption is of great significance for improving the sustainability of social development. Currently, most energy-efficient and regenerative energy recovery methods are from a vehicle control perspective, ignoring the impact on the overall traffic environment. An important reason is that the transportation system's large scale, complexity and social nature restrict its energy-efficient development. Hence, this letter proposes energy-efficient and regenerative energy recovery schemes for sustainable intelligent transportation system using the Artificial societies, Computational experiments, Parallel execution (ACP) framework. The framework includes three parts: energy-efficient oriented intelligent road infrastructure design, transportation traffic flow and vehicle velocity profile planning co-design, and cloud-based vehicle engine parameter calibration. This letter is the second part of Distributed/Decentralized Hybrid Workshop on Sustainability for Transportation and Logistics (DHW-STL) and aims to enhance the sustainability of transportation system from the energy-efficient perspective.
6 April 2023 in IEEE Transactions on Intelligent Vehicles
As economic and social activities continue to increase, transportation is increasingly contributing to climate change, air pollution and other environmental damage. The growing concerns about the sustainability of transportation are forcing everyone in this field to think about solutions to keep our mobility environmentally, economically and socially sustainable.To provide a forum for the exchange of ideas and experiences from industry, academia and the public sector, we have recently held a series of seminars and the first Distributed/Decentralized Hybrid Workshop on Sustainability for Transportation and Logistics (DHW-STL), part of Distributed/Decentralized Hybrid Symposia on Sustainability for Transportation and Logistics (DHS-STL), and Distributed/Decentralized Hybrid Conferences on Sustainability for Transportation and Logistics (DHC-STL). This letter provides a brief report of the First DHW-STL and discusses the potentials, possibilities and perspectives driven by Intelligent Vehicles (IV) technologies to achieve sustainable intelligent systems for transportation and logistics.
10 November 2020 in IEEE Transactions on Intelligent Transportation Systems
Accurate traffic prediction under various conditions is an important but challenging task. Due to the complicated non-stationary temporal dynamics in traffic flow time series and spatial dependencies on roadway networks, there is no particular method that is clearly superior to all others. Here, we focus on investigating ensemble learning that benefits from multiple base models, and propose a traffic-condition-aware ensemble approach that acts as a decision maker by stacking multiple predictions based on dynamic traffic conditions. To sense traffic conditions, we apply the Convolutional Neural Network (CNN) model to capture the spatiotemporal patterns embedded in traffic flow. Then, the high-level features extracted by CNN are used to generate weights to ensemble multiple predictions of different models. Extensive experiments are performed with a real traffic dataset from the Caltrans Performance Measurement System. We compare the proposed approach with competitive models, including Gradient Boosting Regression Tree (GBRT) model, Weight Regression model, Support Vector Regression (SVR) model, Long Short-term Memory (LSTM) model, Historical Average (HA) model and CNN model. Experimental results demonstrate that our method can effectively improve the performances of traffic flow prediction.
16 June 2023 in IEEE Transactions on Intelligent Vehicles
The Paris Agreement set global energy conservation and emission reduction standards in 2016, and the ambitious goal of carbon neutrality must be achieved by 2050. However, the conflicts between environmental protection, economic development and social fairness constrain the development of sustainable transportation systems. At the same time, the automotive industry believes that large-scale application of autonomous driving technology is expected by 2050. The development and application of autonomous driving technology can provide a new solution to improve the sustainability of the transportation system. Meanwhile, autonomous vehicles can significantly enhance the mobility of people with disabilities and seniors and reduce the transportation cost for poverty people, improving the transportation fairness of the entire society. This letter discusses the impact of autonomous driving technology on sustainable transportation systems from the perspectives of social fairness and economic responsibility, as a brief summary of the third part of Distributed/Decentralized Hybrid Workshop on Sustainability for Transportation and Logistics (DHW-STL).
21 December 2023 in IEEE Transactions on Intelligent Vehicles
The advent of Connected Autonomous Vehicles (CAVs) paves the way to a new era of urban traffic control and management, driven by Artificial Intelligence (AI)-enabled strategies. This advancement promises significant improvements in infrastructure use optimization, traffic delay reduction, and overall sustainability. The autonomous driving capabilities of CAVs, coupled with the communication technology, allow vehicles to play an active role in urban traffic control: they can follow tailored instructions and can act as highly accurate moving sensors for traffic authorities. However, such improved capabilities come at the cost of unprecedented vulnerabilities to cyber exploitation, and with the concrete potential to increase social and economic disparities. As an extension of TIV-DHW (Distributed/Decentralized Hybrid Workshop) on ERS (Ethics, Responsibility, and Sustainability), this letter explores how the AI-enabled routing methodology can enhance urban transportation sustainability while also discussing its ethical implications and challenges it presents.