A novel model for real-time risk evaluation of vehicle-pedestrian interactions at intersections.

Safety decisions for vehicles at an intersection rely on real-time, objective and continuous assessment of risks in vehicle-pedestrian interactions. Existing surrogate safety models, constrained by ideal assumptions of constant current speed and reliant on interaction points, often misjudge risks, and show inefficiency, inaccuracy and discontinuity. This work proposes a novel model for evaluation of those risks in vehicle-pedestrian interactions at intersections, which abstracts the pedestrian distribution density around a vehicle into a generalized model of driver-pedestrian interaction preferences. The introduction of two conceptions: 'driving risk index' and 'driving risk gradient,' facilitates the delineation of driving spaces for identifying safety-critical events. By means of the trajectory data from three intersections, model parameters are calibrated and a multidimensional vehicle-pedestrian interaction risk (VPIR) model is proposed to adapt the complex and dynamic characteristics of vehicle-pedestrian interactions at intersections. Commonly used surrogate safety models, such as Time to Collision (TTC), are selected as benchmark models. Results show that the proposed model overcomes the limitations of the existing interaction-point-based models, and offers a ideal assessment of driving risks at intersections. Finally, the model is illustrated with a case study that assesses the risks in vehicle-pedestrian interactions in varied scenarios and the case study indicates that the VPIR model works well in evaluating vehicle-pedestrian interaction risks. This work can facilitate humanoid learning in the autonomous driving domain, and achieve an ideal evaluation of vehicle-pedestrian interaction risks for safe and efficient vehicle navigation through an intersection.

A spatio-temporal deep learning approach to simulating conflict risk propagation on freeways with trajectory data.

On freeways, sudden deceleration or lane-changing by vehicles can trigger conflict risk that propagates backward in a specific pattern. Simulating this pattern of conflict risk propagation can not only help prevent crashes but is also vital for the deployment of advanced vehicle technologies. However, conflict risk propagation simulation (CRPS) on freeways is challenging due to the nuanced nature of the pattern, intricate spatio-temporal interdependencies among sequences and the high-resolution requirements. In this work, we introduce a conflict risk index to delineate potential conflict risk by aggregating various surrogate safety measures (SSMs) over time and space, and then propose a Spatio-Temporal Transformer Network (STTN) to simulate its propagation patterns. Multi-head attention mechanism and stacking layers enable the transformer to learn dynamic and hierarchical features in conflict risk sequences globally and locally. Two components, spatial and temporal learning transformers, are innovatively incorporated to extract and fuse these features, culminating in a fine-grained conflict risk inference. Comprehensive tests in real-world datasets verified the effectiveness of the STTN. Specifically, we employ three widely-recognized SSMs: Modified Time-To-Collision (MTTC), Proportion of Stopping Distance (PSD), and Deceleration Rate to Avoid a Collision (DRAC). These SSMs, gleaned from vehicle trajectories, are employed to delineate the conflict risk. Then, we conduct three comparative simulation tasks: MTTC-based model, PSD-based model, and DRAC-based model. Experimental results show that the PSD-based model exhibits a robust performance on all tasks, and is minimally affected by the durations of the simulation time, while the DRAC-based model more distinctly delineates the spatio-temporal conflict risk heterogeneity. Furthermore, we benchmark the STTN against three common state-of-the-art machine learning models across all simulation tasks. Results reveal that the STTN consistently surpassed these benchmark models (LSTM, CNN and ConvLSTM), suggesting the potential of the attention mechanism on the CRPS tasks. Our investigation offers crucial insights beneficial for traffic safety warning, advanced freeway management systems, and driver assistance systems, among others.

Modeling Limited-Stop Bus Corridor Services with Fare Payment Mode Choice and Trip Purpose Consideration.

This paper proposes a novel model for optimizing limited-stop bus corridor services with consideration of varied payment modes and different trip purposes. In the proposed model, the bus dwell time at a stop is dependent on the fare payment modes and the number of passengers getting off and waiting at the stop while those with the similar trip purpose are grouped into one user class. Given an origin-destination (OD) passenger trip matrix and a set of candidate bus lines serving a corridor, the proposed model is to minimize the total social cost that consists of the cost to the bus operator and the cost to the passengers. In the formulation of the optimization problem, a weighting parameter is adopted to balance the operator cost and the passenger cost. Numerical examples are presented to illustrate the importance of considering passenger flow impacts on bus (and passenger) travel times in the proposed model. We also investigate effects on the optimal limited-stop services (e.g., short-turn, skip-stop, and express) taking into account the choice of fare payment modes (e.g., on-board fare collection including payment by cash, magnetic strip or smart card, off-board fare collection) and different values of travel time due to passenger trip purposes. It is shown that the off-board payment mode would be more efficient in a high-demand corridor, that more passengers prefer to express and skip-stop services rather than normal regular services in the four collection systems, and that different limited-stop service plans should be used for different periods of the day in response to temporal variation in OD passenger travel patterns. The intellectual merit of this paper is not the seemingly obvious conclusions but that the proposed model can handle the problem of limited-stop bus corridor service design with the consideration of fare payment mode choice and trip purposes.

Optimal public-transport operational strategies to reduce cost and vehicle's emission.

Public transport passenger demand is inevitably made non-uniform because of spatial and temporal land use planning. This non-uniformity warrants the use of public transport operational strategies to attain operating efficiency. The optimization of these strategies is commonly being done from the operator perspective, and indirectly from the user perspective. However, the environmental perspective of these strategies, in terms of vehicle's emission, has not been investigated. This study proposed a methodology to analyze the benefits of using transit operational strategies to reduce operating cost and eventually also to reduce undesirable emissions. First, a strategy-based optimization model is established to minimize the number of transit vehicles required. Four candidate operational strategies are considered in this model, including full route operation (FRO), short turn, limited stop, and a combination of limited stop and short turn. Second, the pollutant emissions of transit vehicles are estimated by the MOVES emission model. The developed methodology is applied to a real life case study in Dalian, China. Results show that the use of operational strategies can not only significantly save the number of vehicles by 12.5%, but also reduce emissions of pollutants (i.e., CO2, HC, CO, NOx, PM2.5) by approximately 13%, compared with applying FRO strategy exclusively. In addition, both benefits can be further enhanced through the use of an efficient payment mode (e.g., off-board or contactless card) or improving bus performance in deceleration/acceleration as well as doors opening and closing at a stop.

Optimizing a desirable fare structure for a bus-subway corridor.

This paper aims to optimize a desirable fare structure for the public transit service along a bus-subway corridor with the consideration of those factors related to equity in trip, including travel distance and comfort level. The travel distance factor is represented by the distance-based fare strategy, which is an existing differential strategy. The comfort level one is considered in the area-based fare strategy which is a new differential strategy defined in this paper. Both factors are referred to by the combined fare strategy which is composed of distance-based and area-based fare strategies. The flat fare strategy is applied to determine a reference level of social welfare and obtain the general passenger flow along transit lines, which is used to divide areas or zones along the corridor. This problem is formulated as a bi-level program, of which the upper level maximizes the social welfare and the lower level capturing traveler choice behavior is a variable-demand stochastic user equilibrium assignment model. A genetic algorithm is applied to solve the bi-level program while the method of successive averages is adopted to solve the lower-level model. A series of numerical experiments are carried out to illustrate the performance of the models and solution methods. Numerical results indicate that all three differential fare strategies play a better role in enhancing the social welfare than the flat fare strategy and that the fare structure under the combined fare strategy generates the highest social welfare and the largest resulting passenger demand, which implies that the more equity factors a differential fare strategy involves the more desirable fare structure the strategy has.