Bin Ran

Pedestrian trajectory prediction is essential for various applications in active traffic management, urban planning, traffic control, crowd management, and autonomous driving, aiming to enhance traffic safety and efficiency. Accurately predicting pedestrian trajectories requires a deep understanding of individual behaviors, social interactions, and road environments. Existing studies have developed various models to capture the influence of social interactions and road conditions on pedestrian trajectories. However, these approaches are limited by the lack of a comprehensive view of social interactions and road environments. To address these limitations and enhance the accuracy of pedestrian trajectory prediction, we propose a novel approach incorporating trip information as a new modality into pedestrian trajectory models. We propose RNTransformer, a generic model that utilizes crowd trip information to capture global information on social interactions. We incorporated RNTransformer with various socially aware local pedestrian trajectory prediction models to demonstrate its performance. Specifically, by leveraging a pre-trained RNTransformer when training different pedestrian trajectory prediction models, we observed improvements in performance metrics: a 1.3/2.2% enhancement in ADE/FDE on Social-LSTM, a 6.5/28.4% improvement on Social-STGCNN, and an 8.6/4.3% improvement on S-Implicit. Evaluation results demonstrate that RNTransformer significantly enhances the accuracy of various pedestrian trajectory prediction models across multiple datasets. Further investigation reveals that the RNTransformer effectively guides local models to more accurate directions due to the consideration of global information. By exploring crowd behavior within the road network, our approach shows great promise in improving pedestrian safety through accurate trajectory predictions.

Transportation systems serve as a crucial foundation for maintaining the normal operation of cities and satisfying the requirements of public life. With the development of next-generation information technologies, automated driving technologies have brought new opportunities to improve the performance of traffic systems and the intelligence level of cities. Currently, significant research efforts have been conducted to develop automated driving systems in three major industries, i.e., automobile, roadway, and telecommunication. However, the collaboration and integration of automated driving systems among automobile, roadway, and telecommunications are still lacking, especially for collaborative development of system architecture and objectives. To address the need, this study first proposes a system architecture of vehicle–road-cloud collaborative automated driving system (VRC-CADS). Three levels of collaborative development, i.e., collaborative sensing, collaborative decision-making, and collaborative control, are designed for the VRC-CADS. Based on that, the typical scenarios of automated driving for each level of the system are further defined and interpreted. Moreover, feasible and systematic suggestions for the collaborative development of the VRC-CADS are provided, considering the cross-cutting collaboration among government agencies, academia, and industry. The proposed system architecture of the VRC-CADS will facilitate the optimization of urban lifelines and the evolution of intelligent cities.

Investigation shows that train travel has a lower pollution impact on the environment than flight travel or car travel. A stated preferences (SP) survey can effectively obtain the data of the commuter’s response to the hypothetical train price changes beyond the scope of previous observations. To this end, based on SP survey, we estimate the price elasticity of train travel demand and analyze its variation rules. It is shown that: (1) the own-price elasticities of demand are −1.049028 during peak period and −1.090438 during off-peak period, respectively; (2) the cross-price elasticities of demand are 0.001280 for train and air and 0.001156 for train and car during peak period; and 0.001350 for train and air and 0.001230 for train and car during off-peak period; (3) the own and cross-price elasticities of demand during off-peak period are bigger than the ones during peak period; (4) when the influence factors’ influence degree is 3 or 5, the own and cross-price elasticities of demand are largest; when the influence degree is 1, the own and cross-price elasticities of demand are smallest. A result application example shows that the elasticities obtained from this paper could be used to reduce energy used and CO₂ emissions effectively.

The increasing vehicle usage has brought about a sharp increase in greenhouse gas (GHG) emissions of vehicles, which brings severe challenges to the sustainable development of road transportation in Chinese counties. Low-carbon transportation planning is an essential strategy for carbon control from the source of carbon emissions and is crucial to the full transition to a low-carbon future. For transportation planning designers, a quick and accurate estimation of carbon emissions under different transportation planning schemes is a prerequisite to determine the optimal low-carbon transportation development plan. To address this issue, a novel prediction method of hourly GHG emissions over the urban roads network was constructed in this paper. A case study was conducted in Changxing county, and the results indicate the effectiveness of our proposed method. Furthermore, we applied the same approach to 30 other counties in China to analyze the influencing factors of emissions from urban road networks in Chinese counties. The analysis results indicate that the urban road mileage and arterial road ratio are the two most important factors affecting road network GHG emissions in road traffic planning process. Moreover, the method was employed to derive peak hour emission coefficients that can be used to quickly estimate daily or annual GHG emissions. The peak hour emission of CO₂, CH₄, and N₂O accounts for approximately 9–10%, 8.5–10.5%, 5.5–7.5% of daily emissions, respectively. It is expected that the findings from this study would be helpful for establishing effective carbon control strategies in the transportation planning stage to reduce road traffic GHG emissions in counties.

Cooperative intelligent freeway traffic control is an important application in intelligent transportation systems, which is expected to improve the mobility of freeway networks. In this paper, we propose a deep neuroevolution model, called ES-CTC, to achieve a cooperative control scheme of ramp metering, differential variable speed limits and lane change control agents for improving freeway traffic. In this model, the graph convolutional networks are used to learn more meaningful spatial pattern from traffic sensors, a knowledge sharing layer is designed for communication between different agents. The proposed neural networks structure allows different agents share knowledge with each other and execute action asynchronously. In order to address the delayed reward and action asynchronism issues, the evolutionary strategy is utilized to train the agents under stochastic traffic demands. The experimental results on a simulated freeway section indicate that ES-CTC is a viable approach and outperforms several existing methods

Autonomous vehicle (AV) is an innovative transport option that has the potential to disrupt all industries tied to transportation systems. The advent of AV technology will bring a novel on-demand mobility pattern such as shared autonomous vehicle (SAV). To promote AV technology, it is important to understand which factors influence travelers’ intention to use AVs and SAVs. This paper collected literature from databases such as Scopus, Web of Science and ScienceDirect, and made a systematic review. The study aims to explore the determinants that influence travelers’ behavioral intentions towards use AVs and SAVs based on an extended version of the theory of planned behavior, which incorporates knowledge and perceived risk. This study was tested empirically using a valid survey sample collected from 906 respondents in China. Structural equation model was conducted to investigate the predictors of intentions to use AVs and SAVs. Results showed that knowledge about AV technology and perceived risk are the two main potential obstacles for travelers to use AVs and SAVs. Attitude significantly affects AVs and SAV choice intentions. Subjective norm is the most critical factor affecting the travelers’ intention to use AVs. Perceived behavioral control potentially stymie the travelers’ intention to use SAVs. The findings will enhance the understanding of travelers’ choice motivation from psychological and service perspectives, and provide data support for governments and companies in improving travel management strategies and product services.

Variable speed limits (VSL) control is a flexible way to improve traffic condition,increase safety and reduce emission. There is an emerging trend of using reinforcement learning technique for VSL control and recent studies have shown promising results. Currently, deep learning is enabling reinforcement learning to develope autonomous control agents for problems that were previously intractable. In this paper, we propose a more effective deep reinforcement learning (DRL) model for differential variable speed limits (DVSL) control, in which the dynamic and different speed limits among lanes can be imposed. The proposed DRL models use a novel actor-critic architecture which can learn a large number of discrete speed limits in a continues action space. Different reward signals, e.g. total travel time, bottleneck speed, emergency braking, and vehicular emission are used to train the DVSL controller, and comparison between these reward signals are conducted. We test proposed DRL baased DVSL controllers on a simulated freeway recurrent bottleneck. Results show that the efficiency, safety and emissions can be improved by the proposed method. We also show some interesting findings through the visulization of the control policies generated from DRL models.

This paper proposes an equilibrium bus boarding model to investigate optimal pricing and service for peak-period bus commuting inefficiency of boarding queuing congestion. Commuters are assumed to choose their optimal time-of-use decision from home or the workplace to the bus. We found that: (1) when the earliest commuter boards the bus as soon as the bus arrives at the bus station, the dynamic boarding queuing congestion toll that eliminates the boarding queuing congestion creates social optimal equilibrium and the optimal bus departure interval during the peak period; (2) the optimal bus departure interval during the peak period is the time that the preceding bus riders spend on boarding, which means the relationship between service frequency and ridership does not conform to the square root principle: the optimal bus frequency is proportional to the square root of the number of commuters.

The ability to identify hazardous traffic events is already considered as one of the most effective solutions for reducing the occurrence of crashes. Only certain particular hazardous traffic events have been studied in previous studies, which were mainly based on dedicated video stream data and GPS data. The objective of this study is twofold: (1) the Markov blanket (MB) algorithm is employed to extract the main factors associated with hazardous traffic events; (2) a model is developed to identify hazardous traffic event using driving characteristics, vehicle trajectory, and vehicle position data. Twenty-two licensed drivers were recruited to carry out a natural driving experiment in Wuhan, China, and multi-sensor information data were collected for different types of traffic events. The results indicated that a vehicle’s speed, the standard deviation of speed, the standard deviation of skin conductance, the standard deviation of brake pressure, turn signal, the acceleration of steering, the standard deviation of acceleration, and the acceleration in Z (G) have significant influences on hazardous traffic events. The sequential minimal optimization (SMO) algorithm was adopted to build the identification model, and the accuracy of prediction was higher than 86%. Moreover, compared with other detection algorithms, the MB-SMO algorithm was ranked best in terms of the prediction accuracy. The conclusions can provide reference evidence for the development of dangerous situation warning products and the design of intelligent vehicles.

Computer algorithms used to identify recurrent freeway bottlenecks have been studied since the deployment of loop detecting systems. Such algorithms automatically analyze the archived loop detector data and identify potential recurrent bottlenecks and their characteristics, such as location, time of day, and activation rate, for further investigation. In a highway congestion mitigation project, such algorithms can save time and resources for the initial screening of bottlenecks over a large freeway network. These algorithms include rule-based, contour-map-based, and simulation-based methods. However, existing methods require loop detector data with high accuracy and consistency, which is difficult to achieve in prevailing loop detecting systems. This paper proposes a new bottleneck identification algorithm with strong error and noise tolerance. Several simple denoising methods to improve the error resistance of existing algorithms are also proposed. Using statistical error analysis methods, the pr...