Zheng Zhao

Bio: Zheng Zhao is an academic researcher from Beihang University. The author has contributed to research in topics: Exoskeleton & Traffic flow. The author has an hindex of 3, co-authored 12 publications receiving 766 citations.

LSTM network: a deep learning approach for short-term traffic forecast

Abstract: Short-term traffic forecast is one of the essential issues in intelligent transportation system. Accurate forecast result enables commuters make appropriate travel modes, travel routes, and departure time, which is meaningful in traffic management. To promote the forecast accuracy, a feasible way is to develop a more effective approach for traffic data analysis. The availability of abundant traffic data and computation power emerge in recent years, which motivates us to improve the accuracy of short-term traffic forecast via deep learning approaches. A novel traffic forecast model based on long short-term memory (LSTM) network is proposed. Different from conventional forecast models, the proposed LSTM network considers temporal-spatial correlation in traffic system via a two-dimensional network which is composed of many memory units. A comparison with other representative forecast models validates that the proposed LSTM network can achieve a better performance.

A novel assistive on-ramp merging control system for dense traffic management

Abstract: On-ramps are area of frequent traffic congestion Proper traffic guidance around on-ramp merging areas exerts a positive effect on the relief of traffic congestion The objective of this paper is to design an assistive ramp-merging control (ARMCON) system It utilizes knowledge about professional driver behavior and the dynamical relationship among the on-ramp vehicles, to produce timely information so as to guide the on-ramp drivers when merging with the main traffic flow Under the guidance of ARMCON, disruption of the main traffic on the express way is minimized while a certain merging rate is maintained

The Concept of Stimuli-Induced Equilibrium Point and Its Application in Ramp-Merging Control

Abstract: In this paper, we propose the novel concept of stimuli-induced equilibrium point to synthesize the speed and position references for automatic on-ramp merging systems. Based on the psychological field theory, the intensity of the stimuli that act upon a driver between two vehicles in a three-vehicle platooning configuration is mathematically modeled to calculate a point at which the stimuli resultant becomes zero. This approach intends to mimic drivers’ decision process when certain distance separation with respect to the leader and follower vehicles is attained for safety. The location of this point is continuously updated according to the speed of the middle vehicle and the current traffic scenario. Such stimuli-induced equilibrium point has shown to improve the performance of existing automatic merging control schemes while increasing safety conditions by providing enough reaction time for drivers to avoid an eventual collision.

Experimental Evaluation of the Stimuli-Induced Equilibrium Point Concept for Automatic Ramp Merging Systems

Abstract: The concept of stimuli-induced equilibrium point (SIEP) has been recently introduced to characterize the psychological interaction of a ramp driver with its putative leader and follower on the expressway during a merging maneuver. It enables the computation of the reference target gap speed and position for the on-ramp merging vehicle based on current traffic conditions and ramp vehicle response. The SIEP has been shown to improve the performance of existing automatic ramp merging control strategies while increasing the level of safety during the merging maneuver. The performance and advantages of the SIEP-based approach have been assessed only through numerical simulations. In this paper, we conducted a comprehensive experimental evaluation of the performance and safety of this SIEP-based approach to ramp merging control using a lab-based test-bed. We employed a novel Pc metric, which is based on the concept of probability of collision, to perform a systematic validation of the SIEP safety. Such a metric serves as a standardized methodology to quantitatively compare the SIEP-based approach with those in this paper.

Stimuli-induced equilibrium point: A psychological field theory application in ramp merging systems

Abstract: Based on the psychological field theory, the stimuli-induced equilibrium point (SIEP) model is formulated to characterize the interaction of a driver between two vehicles (leader and follower) in a three-vehicle platooning configuration. Considering the intensity of the stimuli that act upon a driver between these vehicles, the point at which the stimuli resultant becomes zero (i.e. the equilibrium point) is obtained. Consequently, the location of such a point within the leader-follower gap changes continuously according to the speed of the middle vehicle and the current traffic scenario. This methodology enables the systematic computation of speed and position references for automatic ramp merging systems. Such stimuli-induced equilibrium point has shown to improve the performance of existing merging control schemes while increasing safety conditions by providing enough reaction time for drivers to avoid an eventual collision.

Traffic Graph Convolutional Recurrent Neural Network: A Deep Learning Framework for Network-Scale Traffic Learning and Forecasting

Abstract: Traffic forecasting is a particularly challenging application of spatiotemporal forecasting, due to the time-varying traffic patterns and the complicated spatial dependencies on road networks. To address this challenge, we learn the traffic network as a graph and propose a novel deep learning framework, Traffic Graph Convolutional Long Short-Term Memory Neural Network (TGC-LSTM), to learn the interactions between roadways in the traffic network and forecast the network-wide traffic state. We define the traffic graph convolution based on the physical network topology. The relationship between the proposed traffic graph convolution and the spectral graph convolution is also discussed. An L1-norm on graph convolution weights and an L2-norm on graph convolution features are added to the model’s loss function to enhance the interpretability of the proposed model. Experimental results show that the proposed model outperforms baseline methods on two real-world traffic state datasets. The visualization of the graph convolution weights indicates that the proposed framework can recognize the most influential road segments in real-world traffic networks.

Short-term forecasting of passenger demand under on-demand ride services: A spatio-temporal deep learning approach

Abstract: 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 dependencies, temporal dependencies, and exogenous dependencies 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 dependencies 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. The experimental results, validated on the real-world data provided by DiDi Chuxing, show that the FCL-Net achieves the better predictive performance than traditional approaches including both classical time-series prediction models and state-of-art machine learning algorithms (e.g., artificial neural network, XGBoost, LSTM and CNN). Furthermore, the consideration of exogenous variables in addition to the passenger demand itself, such as the travel time rate, time-of-day, day-of-week, and weather conditions, is proven to be promising, since they reduce the root mean squared error (RMSE) by 48.3%. It is also interesting to find that the feature selection reduces 24.4% in the training time and leads to only the 1.8% loss in the forecasting accuracy measured by RMSE in the proposed model. 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.

Spatiotemporal Recurrent Convolutional Networks for Traffic Prediction in Transportation Networks.

Abstract: Predicting large-scale transportation network traffic has become an important and challenging topic in recent decades. Inspired by the domain knowledge of motion prediction, in which the future motion of an object can be predicted based on previous scenes, we propose a network grid representation method that can retain the fine-scale structure of a transportation network. Network-wide traffic speeds are converted into a series of static images and input into a novel deep architecture, namely, spatiotemporal recurrent convolutional networks (SRCNs), for traffic forecasting. The proposed SRCNs inherit the advantages of deep convolutional neural networks (DCNNs) and long short-term memory (LSTM) neural networks. The spatial dependencies of network-wide traffic can be captured by DCNNs, and the temporal dynamics can be learned by LSTMs. An experiment on a Beijing transportation network with 278 links demonstrates that SRCNs outperform other deep learning-based algorithms in both short-term and long-term traffic prediction.

Spatiotemporal Recurrent Convolutional Networks for Traffic Prediction in Transportation Networks

Abstract: Predicting large-scale transportation network traffic has become an important and challenging topic in recent decades. Inspired by the domain knowledge of motion prediction, in which the future motion of an object can be predicted based on previous scenes, we propose a network grid representation method that can retain the fine-scale structure of a transportation network. Network-wide traffic speeds are converted into a series of static images and input into a novel deep architecture, namely, spatiotemporal recurrent convolutional networks (SRCNs), for traffic forecasting. The proposed SRCNs inherit the advantages of deep convolutional neural networks (DCNNs) and long short-term memory (LSTM) neural networks. The spatial dependencies of network-wide traffic can be captured by DCNNs, and the temporal dynamics can be learned by LSTMs. An experiment on a Beijing transportation network with 278 links demonstrates that SRCNs outperform other deep learning-based algorithms in both short-term and long-term traffic prediction.