Currently, there is a serious lack of stereoscopic observation data, which limits the comprehensive understanding of the complex temporal and spatial processes of typical atmospheric pollution. Due to its flexibility and controllability, UAVs can continuously monitor traffic pollutants in an all-round way, and reveal the law of vertical diffusion of traffic pollutants more truthfully. They have gained popularity in recent years.

Based on multi-rotor drones, the research team built portable air pollution monitoring equipment. Through wind tunnel tests and numerical simulations, they designed an air pollution monitoring platform based on drones to conduct integrated monitoring of urban air pollution. Up to now, the research team has carried out vertical observation experiments of air pollutants in more than ten cities including the North China Plain, the Yangtze River Delta and the Pearl River Delta using drone platforms, and has achieved good research results, which are the source of local air pollution. And governance provides vertical data support.

The main research cases include:

1.Research on Vertical Diffusion of Air Pollution Based on UAV

Based on drones and portable air pollution monitoring equipment, build a drone air pollution monitoring platform. Through the monitoring of the vertical distribution of air pollutants on urban roads, it is revealed that the built environment such as elevated and vegetation affects the vertical diffusion of urban air pollution.

2.Research on the Probable Source of Air Pollution Based on UAV

Based on the self-built drone air pollution monitoring platform, the three-dimensional diffusion law of air pollutants above a steel plant was evaluated, and the pollution sources in the plant area were analyzed.

3.UAV-based port and marine environment monitoring

Considering the safety of flying on the water, autonomously build an unmanned aerial vehicle platform suitable for flying on the water to conduct three-dimensional monitoring of the spread of oil spills on the sea and the spread of ship emissions.

4.UAV-based integrated ground-air monitoring of traffic pollution

Based on the UAV air pollution monitoring platform, combined with the on-board exhaust gas monitoring system (PEMS), a three-dimensional monitoring platform integrating ground and air is constructed. Through the integrated ground-air monitoring of urban air pollution, the three-dimensional diffusion process and regional transmission laws of urban air pollutants are discussed.

Transportation, environment, and health is one of the most active interdisciplinary research at present. Through joint research on traffic, environment, and health issues, common problems in traffic congestion, traffic pollution, and health can be revealed. Then realize the coordinated control of traffic congestion and traffic pollution, and provide theoretical guidance and decision consultation for the current urban disease management.

Based on the research of three-dimensional spatial and temporal distribution of urban traffic emissions, the research team revealed the exposure levels of different groups of people to urban traffic emissions in combination with the physiological characteristics. In addition, by evaluating the health risks of pedestrian traffic pollution exposure, recommendations for optimization and control of urban traffic emissions are made.

The main research cases include:

1.Mobile monitoring of traffic pollution based on the portable emission measurement system (PEMS)

The portable emission measurement system is an integrated onboard exhaust gas analysis system. It integrates a high-precision emission analyzer, flow meter, main control computer, and can measure and record vehicle exhaust, geographical location information, meteorological parameters, and vehicle speed in driving. Based on the PEMS equipment, the research team carried out mobile monitoring of traffic pollution in the urban road network.

2.Spatial and temporal distribution of urban road traffic pollution

The spatial and temporal distribution of traffic pollution in different scales (micro, mid-scope, and macro) of the urban road network was revealed by mobile monitoring based on PEMS equipment and portable equipment. On this basis, traffic simulation and air pollution numerical simulation evaluated the impact of traffic restriction strategies and traffic infrastructure. Thereby optimizing the strategies to realize the coordinated control of traffic congestion and traffic pollution, and to provide theoretical guidance and decision-making consultation for the current urban disease management.

3.Study on urban traffic pollution and health risk of residents

Residents near the street and pedestrians are close to traffic pollution sources and are inevitably exposed to air pollution for a long time, thus their physical and mental health is greatly threatened. Through mobile monitoring of traffic pollution, the research group revealed the spatial and temporal distribution characteristics of air pollutants on urban roads, assessed the health risks of residents and pedestrians exposed to air pollution, and constructed optimization strategies for the spatial layout of street neighborhoods.

4.Research on the temporal and spatial distribution of urban road traffic carbon emissions

The transportation industry is not only a major energy consumer but also a major carbon emitter. In the context of the carbon peak and carbon neutrality of China, it is urgent to explore carbon emissions in the transportation sector. This research is based on the PEMS to reveal the carbon emissions characteristics of different types of vehicles on urban roads, explore the temporal and spatial distribution of carbon emissions in urban road network traffic, and provide science and data support for achieving carbon neutrality and carbon peak in the transportation field.

5. Portable air pollution mobile monitoring backpack

Compared with fixed monitoring, mobile monitoring has strong flexibility and can monitor the multi-dimensional temporal and spatial distribution of pollutants on a smaller scale, which is a powerful supplement to fixed monitoring. The research group invented a portable smart backpack for mobile monitoring of air pollution, which can be used flexibly in a variety of built environments.

Intelligent transportation relies on Internet of things, big data, artificial intelligence and other information technologies to analyze, simulate and optimize the transportation system. The group is problem-oriented and data-driven to study a series of urban traffic problems.

Intelligent transportation relies on Internet of things, big data, artificial intelligence and other information technologies to analyze, simulate and optimize the transportation system. The group is problem-oriented and data-driven to study a series of urban traffic problems.

The main research cases include:

1.Evaluation and prediction of transportation resilience under extreme weather events

Based on the attention mechanism in deep learning and graph convolution neural network, the resilience and toughness characteristics of road traffic system under extreme weather are evaluated, and the evolution trend of traffic system recovery is finely predicted. The research results could assist researchers and policy makers clearly understand the real-world resilience of urban road networks in both theory and practice, and take effective responses under emergent disruptive events.

2.Fine prediction of urban air pollution based on deep learning

Considering the temporal and spatial correlation of air quality monitoring network, a model combining attention seq2seq model and depth neural network is constructed to adaptively fuse and learn the proximity factors and spatial information in the air quality monitoring network, and finely predict the ozone concentration near the ground in urban areas.

3.Study on regional transmission of air pollution based on complex network

Based on the complex network method, the community structure of air pollution transmission network is described, and the law of regional transmission of air pollution is discussed, so as to provide decision support for the joint prevention and control of regional air pollution and carbon emission.

4.Real-time spatiotemporal prediction and imputation of traffic status based on LSTM and Graph Laplacian regularized matrix factorization

Accurate prediction of traffic status in real time is critical for advanced traffic management and travel navigation guidance. There are many attempts to predict short-term traffic flows using various deep learning algorithms. Most existing prediction models are only tested on spatiotemporal data assuming no missing data entries. However, this ideal situation rarely exists in real world due to sensor or network transmission failure. Missing data is a nonnegligible problem. In this study, we propose an online framework that can make spatiotemporal predictions based on raw incomplete data and impute possible missing values at the same time. We design a novel spatial and temporal regularized matrix factorization model, namely LSTM-GL-ReMF. The experimental results show our approach has a robust and accurate performance in terms of prediction and imputation accuracy under various data missing scenarios.