Non-signalized Intersections Research Articles

V2X Empowered Non-Signalized Intersection Management in the AI Era: Opportunities and Solutions

With significant advancements of vehicle-to-everything (V2X) communication technologies in recent years, intelligent collaboration between connected vehicles and roadside units provides new opportunities for better road safety and vehicle traffic management via innovative non-signalized intersections. In this article, to deal with sophisticated traffic information brought about by a dynamic non-signalized intersection environment, artificial intelligence (AI) solutions together with V2X communication technologies are proposed to provide data-driven intersection management strategies. We first present the emerging technologies and key challenges toward V2X empowered non-signalized intersection management. By introducing the applications of typical AI technologies, we pinpoint three main research issues on V2X empowered non-signalized intersection management, leveraging various AI approaches. Finally, we present a case study where a multi-agent learning approach is applied for intelligent multiple intersection management to demonstrate the effectiveness of our non-signalized intersection solution.

Characteristics of vehicular traffic flow at nonsignalized intersection: Simulation study by Lárraga and Alvarez-Icaza model

This paper develops Lárraga and Alvarez-Icaza (LAI) cellular automata model which is based on safe human reaction. Vehicles are moving at two single-lane streets with periodic boundary condition and pass from intersection point via yielding mechanism avoiding collision. The model characteristics and fundamental diagrams have been obtained by Monte Carlo simulations. Our results suggest that LAI model with yielding mechanism is able to reproduce the realistic acceleration and deceleration capabilities, which are desired parameters. Moreover, the plateau region in total current illustrates the hardness of the yielding mechanism and limited speed strategy, which is applied near the intersection point.

Spatial Model of Traffic Congestion by the Changes on City Transportation Route

Traffic congestion is a problem for every city in Java Island, Indonesia, including Bogor City. Factors causing congestion in Bogor City are thought to come from land use, the geometry and performance of road, and public transport routes (urban transport). The change of urban transportation (angkot) route carried out by Bogor City Government aims to reduce traffic density and congestion, but it is not a guarantee that the main problem will be solved. This study aims to determine the spatial patterns of traffic density and congestion with current angkot routes and to construct a model to predict traffic density and congestion when new angkot routes are used. Variables in this research are land use (number of schools and markets/malls), geometry and performance of road (vehicle volume, road capacity, average velocity, road type, number of lanes, number of signalled and non-signal intersection), and an angkot route passing a road. The method used in modelling is multiple regression using one dummy variable and a stepwise regression method. The result of modelling shows that the variables affecting traffic density are velocity, some signalled and non-signal intersection, and angkot route with R2 value 66.9%. At the same time, the influential variables in the traffic congestion model are vehicle volume, road capacity, number of the signalled intersection, and angkot route with R2 value 81.4%. To see accuracy in predicting model of traffic density and congestion, Mean Absolute Percentage Error (MAPE) validation is used. The results show a value of 12.46% for traffic density, which means that the model has good prediction accuracy and 5.62% for traffic congestion, which means the model has high prediction accuracy. Thus, in this study, the land use of school and market is not a factor causing traffic density and congestion, while the geometry and performance of roads and public transportation routes as a factor causing congestion.

Multicriteria Autonomous Vehicle Control at Non-Signalized Intersections

The aim of the paper is to describe a multicriteria model predictive control method for autonomous vehicles at non-signalized intersections. The centralized controller aims to describe control action for each autonomous vehicle to guarantee collision free passage. At the same time, performances are defined for the centralized Model Predictive Controller, namely the minimization of traveling time and energy consumption. Since these control goals are often conflicting, a scheduling variable is introduced to create a balance between them. Hence, the centralized controller can be tuned based on the importance of each control goal, which can be useful in urban environment where traffic densities may vary heavily depending on the period of the day. The effectiveness of the proposed centralized multicriteria controller is demonstrated through a complex simulation example in CarSim simulation environment using different tpye of autonomous vehicles.

A bi-level cooperative driving strategy allowing lane changes

This paper studies the cooperative driving of connected and automated vehicles (CAVs) at conflict areas (e.g., non-signalized intersections and ramping regions). Due to safety concerns, most existing studies prohibit lane change since this may cause lateral collisions when coordination is not appropriately performed. However, in many traffic scenarios (e.g., work zones), vehicles must change lanes. To solve this problem, we categorize the potential collision into two kinds and thus establish a bi-level planning problem. The right-of-way of vehicles for the critical conflict zone is considered in the upper-level, and the right-of-way of vehicles during lane changes is then resolved in the lower-level. The solutions of the upper-level problem are represented in tree space, and a near-optimal solution is searched for by combining Monte Carlo Tree Search (MCTS) with some heuristic rules within a very short planning time. The proposed strategy is suitable for not only the shortest delay objective but also other objectives (e.g., energy-saving). Numerical examples show that the proposed strategy leads to good traffic performance in real-time.

Proximal Policy Optimization Through a Deep Reinforcement Learning Framework for Multiple Autonomous Vehicles at a Non-Signalized Intersection

Advanced deep reinforcement learning shows promise as an approach to addressing continuous control tasks, especially in mixed-autonomy traffic. In this study, we present a deep reinforcement-learning-based model that considers the effectiveness of leading autonomous vehicles in mixed-autonomy traffic at a non-signalized intersection. This model integrates the Flow framework, the simulation of urban mobility simulator, and a reinforcement learning library. We also propose a set of proximal policy optimization hyperparameters to obtain reliable simulation performance. First, the leading autonomous vehicles at the non-signalized intersection are considered with varying autonomous vehicle penetration rates that range from 10% to 100% in 10% increments. Second, the proximal policy optimization hyperparameters are input into the multiple perceptron algorithm for the leading autonomous vehicle experiment. Finally, the superiority of the proposed model is evaluated using all human-driven vehicle and leading human-driven vehicle experiments. We demonstrate that full-autonomy traffic can improve the average speed and delay time by 1.38 times and 2.55 times, respectively, compared with all human-driven vehicle experiments. Our proposed method generates more positive effects when the autonomous vehicle penetration rate increases. Additionally, the leading autonomous vehicle experiment can be used to dissipate the stop-and-go waves at a non-signalized intersection.

Development of a Partial Proportional Odds Model for Pedestrian Injury Severity at Intersections

Pedestrian injury in crashes at intersections often results from complex interaction among various factors. The factor identification is a critical task for understanding the causes and improving the pedestrian safety. A total of 2,614 crash records at signalized and non-signalized intersections were applied. A Partial Proportional Odds (PPO) model was developed to examine the factors influencing Pedestrian Injury Severity (PIS) because it can accommodate the ordered response nature of injury severity. An elasticity analysis was conducted to quantify the marginal effects of contributing factors on the likelihood of PIS. For signalized intersections, seven explanatory variables significantly affect the likelihood of PIS, in which five explanatory variables violate the Proportional Odds Assumption (POA). Local driver, truck, holiday, clear weather, and hit-and-run lead to higher likelihood of severer PIS. For non-signalized intersections, six explanatory variables were found significant to the PIS, in which three explanatory variables violate the POA. Young and adult drivers, senior pedestrian, bus/van, divided road, holiday, and darkness tend to increase the likelihood of severer PIS. The vehicles of large size and heavy weight (e.g. truck, bus/van) are significant factors to the PIS at both signalized and non-signalized intersections. The proposed PPO model has demonstrated its effectiveness in identifying the effects of contributing factors on the PIS.

Investigating the Factors Influencing Pedestrian–Vehicle Crashes by Age Group in Seoul, South Korea: A Hierarchical Model

Walking is the most natural form of moving in everyday life. However, the urban environment is not always safe for walking. Hence, it is appropriate to create pedestrian safety-oriented neighborhood environments to promote walkability. This study identifies factors that can affect the severity of pedestrian injuries by considering and comparing the individual characteristics of pedestrian crashes and the built environment of neighborhoods using a hierarchical model. The main results of this study are as follows. Those under 18 and over 65 years of age, which are recognized as vulnerable road users, are more affected by the neighborhood environments than other age groups. Parks, convenience stores, and restaurants are likely to play a significant role in reducing the severity of pedestrian injuries. However, for pedestrians under 18 years of age, convenience stores that students often visit after school are likely to increase the severity of pedestrian injuries. The severity of pedestrian injuries is likely to be lower at non-signalized crosswalks and intersections than at signalized crosswalks and intersections. The study is expected to contribute to existing literature on the topic and guide policy makers and planners to arrive at consensus approaches when making decisions on how to build pedestrian safety-oriented neighborhoods.

State of the art and computational aspects of time-dependent waiting models for non-signalised intersections

Abstract Time-dependent models are of great importance in highway engineering as they are appropriate for evaluating waiting times and queue lengths at intersections, which are integral parts of various activities in planning, verification and decision support for infrastructure. After reviewing the literature of the main time-dependent models based on the coordinate transformation method and a discussion about some computational issues in time-evolution profiles for non-signalised intersections, the paper identifies the requirements these models have to satisfy in order to be used as “basic” cases for analysing complex evolutionary situations. Three “basic” cases are presented with their time-dependent equations for vehicle waiting times and vehicle number; they have been completed and dimensionally homogenised in this paper. As they are recursive, these formulas can be applied for sequential intervals in the time domain in both vehicles and passenger car units. The closed-form expressions for state variables show to be mutually equivalent in comparison with discrete event simulation models and imbedded Markov chain results. For all the three models the paper presents a common deterministic simplification for average waiting time, with good approximation results in the tested cases. The proposed time-dependent formulas will contribute to a better adherence to the real phenomena, compared to the extremely simplified and unrealistic methods suggested by the international manuals for level of service assessment. The proposed formulas will be useful for current applications and possible future development in order to meet the emerging needs of road and transport engineering.

Truck drivers’ interaction with cyclists in right-turn situations

One of the more hazardous situations for a bicyclist is to go straight on in an intersection where a motor vehicle is turning right, and especially so when heavy vehicles are involved. The aim of this study was to investigate truck drivers’ speed choice, gaze behaviour and interaction strategies in relation to vulnerable road users (VRU) when turning right in signalised and non-signalised intersections. Truck drivers experienced (n = 14) or inexperienced (n = 15) with urban traffic drove a 15 km long test route in an urban environment. To guarantee the presence of VRUs, a confederate cyclist with the task to cycle straight on was present in three intersections. Overall, the results suggest that the specific experience of driving a truck in the city has little effect on the strategies employed when interacting with cyclists in a right turn scenario. Neither gaze nor strategic placement or speed related variables differed significantly between the groups, though the drivers inexperienced with urban traffic tended to be more cautious. Glance and driving behaviour were more related to the preconditions afforded by the infrastructure and to interaction type, which is a combination of those infrastructural preconditions and the truck driver’s own choice of action. The likelihood of a favourable interaction should be increased, where the truck remains behind the VRUs on the approach to the intersection, something which eliminates the potential for a collision. Education of truck drivers, infrastructure design and improved traffic light sequences are potential ways to reduce the occurrence of more demanding and dangerous interaction types.

Non-Signalized Intersection Network Management With Connected and Automated Vehicles

Road network control is challenging but critical in enhancing urban traffic. This paper proposes a management method for a road network with non-signalized intersections in the connected-vehicle environment, which coordinates connected and automated vehicles' movements to make the non-signalized intersections conflict-free and efficient. Firstly, the proposed method projects vehicles in road networks into virtual platoons and builds the road-network-wide conflict-free geometric topology considering the vehicles' conflicting relationships, which describes the geometry car-following relationship in virtual platoons. Then, a distributed linear controller is designed considering vehicle dynamics and communication topology to organize vehicles' movements with the desired geometric topology. Finally, simulations are conducted to verify the proposed method with different traffic demands. Simulation results show the proposed method can significantly improve traffic efficiency, as well as traffic safety.

Model Predictive Control for the Coordination of Autonomous Vehicles at Intersections

Abstract The paper describes a method for defining the crossing order of autonomous vehicles at a non-signalized intersection. The safe passage of vehicles through the intersection is achieved by a centralized Model Predictive Control (MPC) method. The minimization of traveling time or energy consumption and the condition of collision-avoidance are incorporated in the control design. The goal of the research is to guarantee safe passage of the autonomous vehicles by taking uncertainties of position measurements into consideration. The operation of the proposed control method is demonstrated by simulation examples made in a high-accuracy simulation environment to present its efficiency.

Fuel economy benefit analysis of pass-at-green (PaG) V2I application on urban routes with STOP signs

In this paper, the fuel economy benefit of a vehicle-to-infrastructure (V2I) algorithm called Pass-at-Green (PaG) was analysed and compared with different speed profile characteristics. The fuel-optimal speed profile calculated offline using Dynamic Programming (DP) was used for benchmarking. Manual Driving Speed Profile was included in the analysis as the human driver. Intelligent Driver Model (IDM), whose parameters were modified to model cautious, normal and aggressive behaviour, was also used for comparison. The state-of-the-art connected and autonomous driving Hardware-in-the-loop (HIL) setup enabled the use of Electronic Horizon (eHorizon), which provided localisation and detailed map information to the simulated test vehicle. An eHorizon Informed Manual Driving speed profile, where the eHorizon HIL system provided upcoming STOP sign location information, was also modelled. The results showed that having V2I capability in the longitudinal speed control of a vehicle has potential for substantial fuel consumption reduction while navigating through signalised and non-signalised intersections.

Safety Evaluation of Pedestrians’ Walking Position Around Non-signalized Intersection of Residential Road

Safety Evaluation of Pedestrians’ Walking Position Around Non-signalized Intersection of Residential Road

Optimal traffic control at smart intersections: Automated network fundamental diagram

Recent advances in artificial intelligence and wireless communication technologies have created great potential to reduce congestion in urban networks. In this research, we develop a stochastic analytical model for optimal control of communicant autonomous vehicles (CAVs) at smart intersections. We present the automated network fundamental diagram (ANFD) as a macro-level modeling tool for urban networks with smart intersections. In the proposed cooperative control strategy, we make use of the headway between the CAV platoons in each direction for consecutive passage of the platoons in the crossing direction through non-signalized intersections with no delay. For this to happen, the arrival and departure of platoons in crossing directions need to be synchronized. To improve system robustness (synchronization success probability), we allow a marginal gap between arrival and departure of the consecutive platoons in crossing directions to make up for operational error in the synchronization process. We then develop a stochastic traffic model for the smart intersections. Our results show that the effects of increasing the platoon size and the marginal gap length on the network capacity are not always positive. In fact, the capacity can be maximized by optimizing these cooperative control variables. We analytically solve the traffic optimization problem for the platoon size and marginal gap length and derive a closed-form solution for a normal distribution of the operational error. The performance of the network with smart intersections is presented by a stochastic ANFD, derived analytically and verified numerically using the results of a simulation model. The simulation results show that optimizing the control variables increases the capacity by 138% when the error standard deviation is 0.1 s.

Solving Traffic Signal Scheduling Problems in Heterogeneous Traffic Network by Using Meta-Heuristics

This paper addresses a traffic signal scheduling (TSS) problem in a heterogeneous traffic network with signalized and non-signalized intersections. The objective is to minimize the total network-wise delay time of all vehicles within a given finite-time window. First, a novel model is proposed to describe a heterogeneous traffic network with signalized and non-signalized intersections. Second, five meta-heuristics are implemented to solve the TSS problem. Based on the problem characteristics, three local search operators and their ensemble are proposed. Then, five meta-heuristics with such an ensemble are proposed to solve the TSS problem. Third, experiments are carried out based on the real traffic data in the Jurong area of Singapore. The performance of the ensemble of local search operators is verified. Ten algorithms, including five meta-heuristics with and without the ensemble, are evaluated by solving 18 cases with different scales. Finally, the algorithm with the best performance is compared against the currently used traffic signal control strategies. The comparisons and discussions show the competitiveness of the proposed model and meta-heuristics.

Investigating secondary pedestrian-vehicle interactions at non-signalized intersections using vision-based trajectory data

Vision-based trajectory data provides great details for investigating microscopic behavior and road safety at the level of interactions. Most studies investigating pedestrian-vehicle interactions at non-signalized intersections have focused on interactions at the crosswalk on the same approach where the vehicle is coming from, which are referred to as primary interactions in this study. However, secondary interactions, defined as interactions between vehicles exiting the intersection and crossing pedestrians, have not been adequately studied. Second interactions can pose more dangers to pedestrians due to the driver’s unclear knowledge of right-of-way, acceleration attempts to recover the speed, and the complex situation the driver faces in the intersection. This paper aims at studying the safety issue of secondary pedestrian-vehicle interactions at non-signalized intersections. For that purpose, a methodology is proposed based on surrogate measures of safety and behavioral measures derived from vision-based trajectories. This methodology is implemented through a case study involving ten all-way stop intersections from Montreal, Canada. Road user trajectory data are automatically extracted from the videos. Different measures are used in the study: from the interaction analysis that determines vehicle-pedestrian interactions based on a Distance-Velocity (DV) model, average crossing speeds and vehicle approaching behaviors in terms of speed and acceleration. Computer tools are developed to extract these measures from the trajectory data. Based on these measures, a comparative analysis is carried out between primary and secondary interactions. Results show that secondary interactions are more dangerous than primary interactions. Among the three secondary interaction types, secondary through interactions are the most dangerous.

Evaluation of Performance Non-signal Intersection on Telekomunikasi Road Intersection - Terusan Buah Batu Road, Bandung - West Java

Evaluation of performance non-signal intersection on Telekomunikasi Road Intersection – Terusan Buah Batu Road is carried out by considering the geometric conditions of the road, traffic volume, side barriers, and environmental conditions around the intersection. Primary data and secondary data were analyzed by referring to the Indonesian Road Capacity Manual Method (MKJI 1997). From the results of the study, it was found that on Wednesday at 15:20 – 16:20, the maximum capacity value of 9,326/hours exceeded the base capacity of type 324 crossing (Co = 3,200 pcu / hour). The highest degree of saturation (DS) occurs on Monday at 6:20 - 7:20, which is 0.86. This means that the service level value is in class E (0.85 - 1.00). Namely, unstable traffic flow, solid vehicles, stagnant traffic and intersection volume higher are than its capacity, so that crossing conditions require handling. From the data above, it is known that the Telekomunikasi Road intersection - Terusan Buah Batu Road is not feasible to accommodate traffic flows and it needs to be handled by making stop lines, road signs as vehicle lane separators and warning signs, danger signs or by installing traffic signal lights.

Analisis Kinerja Simpang Tak Bersinyal Jalan Raya Dramaga-Bubulak Bogor, Jawa Barat

The potential of traffic problems at non-signal intersection relatively higher and more complex than signalized intersection. Intersection of Dramaga-Bubulak street has no signal and is one of the intersection with high level mobility. The research objectives were to analysis the performance of actual conditions and alternatives strategies to improve intersection performance as well as to know the influence of one way traffic system on intersection performance. The results showed that the volume of traffic passing the intersection were 3816 passenger car unit (pcu)/hour at peak hour with degree of saturation (DS) of 0.85 and delay on the intersection was 14.41 sec/pcu so the intersection service level was B. The service level of intersection was good, so the research analyzed three alternatives strategies to decrease DS value. Performance analysis of traffic showed that alternatives I was the best model alternative condition, because application of one way traffic system on outer ring Bubulak would decreased the service level of Dramaga-Bubulak street.Key words: intersection, non-signal, street

Exposure to pedestrian crash based on household survey data: Effect of trip purpose

Pedestrian are vulnerable to severe injury and mortality in the road crashes. Understanding the essence of the pedestrian crash is important to the development of effective safety countermeasures and improvement of social well-being. It is necessary to measure the exposure for the quantification of pedestrian crash risk. The primary goals of this study are to explore the efficient exposure measure for pedestrian crash, and identify the possible factors contributing to the incidence of pedestrian crash. In this study, amount of travel was estimated based on the Travel Characteristic Survey (TCS) data in 2011, and the crash data were obtained from the Transport Information System (TIS) of the Hong Kong Transport Department during the period from 2011 to 2015. Total population, walking frequency and walking time were adopted to represent the pedestrian exposure to road crash. The effect of trip purpose on pedestrian crash was evaluated by disaggregating the pedestrian exposure proxies by purpose. Three random-parameter negative binomial regression models were developed to compare the performances of the three pedestrian exposure proxies. It was found that the model in which walking frequency was used as the exposure proxy provided the best goodness-of-fit. Frequency of walking back home, among other trip purposes, was the most sensitive to the increase in pedestrian crash risk. Additionally, increase in the frequency of pedestrian crash was correlated to the increases in the proportions of children and elderly people. Furthermore, household size, median household income, road density, number of non-signalized intersection as well as number of zebra crossings also significantly affected the pedestrian crash frequency. Findings of this study should be indicative to the development and implementation of effective traffic control and management measures that can improve the pedestrian safety in the long run.