Xiqun | Searcharxiv

arXiv2018

Spatial-Temporal Inference of Urban Traffic Emissions Based on Taxi Trajectories and Multi-Source Urban Data

Vehicle trajectory data collected via GPS-enabled devices have played increasingly important roles in estimating network-wide traffic, given their broad spatial-temporal coverage and representativeness of traffic dynamics. This paper exploits taxi GPS data, license plate recognition (LPR) data and geographical information for reconstructing the spatial and temporal patterns of urban traffic emissions. Vehicle emission factor models are employed to estimate emissions based on taxi trajectories. The estimated emissions are then mapped to spatial grids of urban areas to account for spatial heterogeneity. To extrapolate emissions from the taxi fleet to the whole vehicle population, we use Gaussian process regression models supported by geographical features to estimate the spatially heterogeneous traffic volume and fleet composition. Unlike previous studies, this paper utilizes the taxi GPS data and LPR data to disaggregate vehicle and emission characteristics through space and time in a large-scale urban network. The results of a case study in Hangzhou, China, reveal high-resolution spatiotemporal patterns of traffic flows and emissions and identify emission hotspots at different locations. This study provides an accessible means of inferring the environmental impact of urban traffic with multi-source data that are now widely available in urban areas.

physics.soc-ph↗

arXiv2017

Short-Term Forecasting of Passenger Demand under On-Demand Ride Services: A Spatio-Temporal Deep Learning Approach

Short-term passenger demand forecasting is of great importance to the on-demand ride service platform, which can incentivize vacant cars moving from over-supply regions to over-demand regions. The spatial dependences, temporal dependences, and exogenous dependences need to be considered simultaneously, however, which makes short-term passenger demand forecasting challenging. We propose a novel deep learning (DL) approach, named the fusion convolutional long short-term memory network (FCL-Net), to address these three dependences within one end-to-end learning architecture. The model is stacked and fused by multiple convolutional long short-term memory (LSTM) layers, standard LSTM layers, and convolutional layers. The fusion of convolutional techniques and the LSTM network enables the proposed DL approach to better capture the spatio-temporal characteristics and correlations of explanatory variables. A tailored spatially aggregated random forest is employed to rank the importance of the explanatory variables. The ranking is then used for feature selection. The proposed DL approach is applied to the short-term forecasting of passenger demand under an on-demand ride service platform in Hangzhou, China. Experimental results, validated on real-world data provided by DiDi Chuxing, show that the FCL-Net achieves better predictive performance than traditional approaches including both classical time-series prediction models and neural network based algorithms (e.g., artificial neural network and LSTM). This paper is one of the first DL studies to forecast the short-term passenger demand of an on-demand ride service platform by examining the spatio-temporal correlations.

cs.LG↗