International Conference on Transportation and Development 2020: Emerging Technologies and Their Impact

The study of autonomous vehicles (AVs) has become a hot topic in recent years. Understanding consumers’ interests and priorities relating to AVs has motived researchers to conduct market analysis studies. This study aims to analyze the willingness to pay (WTP) of adult residents in the U.S. for AVs. Implementing adaptive choice-based conjoint analysis (ACBC), which is a special market analysis technique, and using comprehensible and realistic attributes in the survey, are the main innovations of this study. The results of this study show that vehicle fuel type is the most important attribute for respondents in their next vehicle, and after that level of automation of the vehicle was the second most important factor. Respondents distinctively preferred fully automated vehicles over partially automated vehicles. Moreover, respondents indicated positive intentions to use exemption or maximum discounts on taxes and registration fees and any possible incentives and amenities that help their mobility. The findings of this study could be utilized by transportation authorities, transportation economists, advanced transport system investment agencies, and collaborators in emerging transportation systems.

Overnight idling by diesel-fueled long-haul trucks is a contributor to mobile-source emissions in the Dallas–Fort Worth (DFW) region and a detriment to the region’s overall air quality. Truck stop electrification (TSE) systems, also known as electrified parking spaces (EPS), are commercial products that provide an alternative for truck drivers to idling their engines during rest periods. Anecdotal evidence indicates that truck drivers often park in the spaces designated for TSE/EPS systems without using the service and idle their engines overnight in the DFW region. The objectives of this study are to: 1) collect field data to establish the TSE/EPS usage rates at three stations in the DFW area, 2) conduct a survey of drivers, truck stop/terminal owners, TSE/EPS system providers, and on-site TSE/EPS system representatives at these stations to identify common reasons for idling engines during rest periods while having access to the TSE/EPS systems, and 3) recommend proper incentives for drivers to use TSE/EPS systems and truck stop/terminal owners to encourage truck drivers to use TSE/EPS systems. The results of the data collection show that the rate of engine idling and TSE/EPS usage rate vary significantly across the stations; the private terminal had the lowest rate of idling compared to the public service stations since the terminal is fully controlled by a manager. The results of the survey show that drivers mostly prefer direct and monetary incentives, such as coupons for truck stop showers, foods, and laundry services. The results also show that one way to motivate truck stop owners is to offer them fixed-rate revenues. Moreover, the best incentive for TSE/EPS system providers is supporting them to modernize and update TSE/EPS equipment.

Based on the consideration of VR related concepts and technologies, this paper makes use of its multi-perception, interaction, and autonomy to bring students an immersive user experience in order to improve the cognitive teaching. Through the friendly software interface, the panoramic roaming of the yard, buildings and other equipment and facilities of the marshalling station as well as the display of the daily operation content are completed. Taking Fengxi marshalling station as an example, the operation process of the operation in the station is studied, and on this basis, the whole railway system is expanded. Let the students feel the railway scene better and more intuitively, while ensuring the safety, let the cognitive practice process really gain. The technical innovation of this paper includes: using 3ds Max to build the infrastructure model, and combining with the on-site investigation data as reference to render; using the leading-edge UE4 game engine to build the virtual scene, and realizing the background logic control, and completely reproducing the operation process of the railway marshalling station; and providing a targeted training and testing system for the study of railway knowledge.

Despite the broad implications, it is not clear how the advent of emerging automated trucking technologies would impact the trucking industry in the United States (U.S.). Hence, this study investigated the potential impacts of autonomous trucking technologies on the future of the trucking industry in the U.S. The pertinent literature was reviewed to assess the state-of-the-art of different levels of automated trucking technologies and to identify their impacts on the critical issues of the trucking industry. The results of the review were then used to design survey questionnaires for truck drivers, trucking company managers/owners, automated trucking technology developers, and transportation planners. Results indicate that improved productivity, improved compliance, safety and accountability (CSA), and reduced crashes were the most expected advantages of autonomous trucking. However, analysis of the responses also imply a significant difference in the perception of the survey respondent groups regarding the degree of realization of the advantages. These results are expected to be highly valuable for transportation planners to coordinate with truck drivers, autonomous trucking technology developers, and trucking company managers for creating cohesive transportation plans aimed at maximizing the benefits of the early adoption of autonomous trucking technologies.

The goal of this research is to study the potential benefits of electric aircraft on short-range operations by using measured operations and performance. This is achieved by studying: noise, environmental parameters, maintenance costs, reliability of batteries, fuel efficiency, safety, and pilot training. Each category of study shows the impacts of all-electric aviation and how they compare to conventional aviation operations. The study will identify and discuss the prototypes developed by Uber, Bell, Pipistrel, Boeing, and Airbus. This research presents data collected for each prototype to show the current status of the technology and presents the strengths and drawbacks of each prototype. In conclusion, the benefits of all-electric aircraft include noise reductions 17%, a reduction in greenhouse emissions of 80%, operating cost and pilot training cost reductions 70%. The main limitations of the further advancement of electric aircraft technology is the battery. Battery density of electric aircraft requires further research, testing, and development to make fully-electric aviation a success. As of now, batteries of electric aircraft can only provide enough power for short-range flights.

Unmanned aerial vehicles have been at the center of innovation in surveying and engineering. However, adoption has been slow due to concerns over accuracy, along with complexities of managing aerial data. In order to see improvements in both accuracy and data processing, the industry needs to continue to explore new approaches. Artificial intelligence (AI) has significant potential to advance the profession in making sense of massive amounts of aerial data and answer to its request for automation. In a recent industry study, over 50% of the respondents mentioned that “lack of automation” is their number one pain point and desire that technology can address in the future. This paper describes the capabilities and limitations of AI applied to surveying, informed by the development of the AirWorks software, and the first tool that creates CAD drawings autonomously by extracting features of aerial images. It demonstrates significant time savings, and how these can be overcome collaboratively going forward.

Accurate measurement of fuel consumption and related CO₂ emissions are of great importance for stakeholders of the transportation systems. In this study, intelligent machine learning models were trained based on comprehensive field testing in California, to rectify the accuracy of fuel consumption measurements. In addition to conventional regression model, more complex model representations such as neural network and ensemble method were incorporated in the learning phase. To achieve optimal performance, the corresponding hyperparameters were tuned to the dataset. To avoid overfitting, first, the dataset was split into training and validation sets. Then, procedures such as early-stopping and regularization were applied to the learners to keep the model representation as simple as possible. The learned parameters were validated by tracking the accuracy of the predictions on the validation set. Finally, the performance of the candidate models was compared. The results reveal higher accuracy achieved by the ensemble method with additional generalization. The accuracy trends assured that the learned models are not prone to overfitting. The ensemble model can be used as a powerful tool in decision-making, planning, and management of transportation systems.

The electric vehicle (EV) market in China has expanded substantially thanks to a series of incentive policies. The operational characteristics and usage patterns of EVs are different from those of traditional fuel vehicles (TFVs), especially in terms of energy savings and cruising duration of battery. With required energy management devices, EVs inadvertently generate fine-grained and continuous trajectory data, which provides an unprecedented opportunity to study the multi-person and multi-day usage patterns of EVs. The data set used in this study includes GPS and on-board diagnostics (OBD) information of 8,000 EVs including both plug-in hybrid electric vehicle (PHEVs) and battery electric vehicles (BEVs) within one-week in Shanghai. A multi-day activity data set is created for each user. The concept of entropy is introduced and modified to describe and distinguish the usage stability of each vehicle through the multi-day activity data set. Based on indicators including usage duration, travel mileage, and spatial parking location, the usage patterns of EVs could be classified into three modes: individual, commercial, and semi-commercial. Among all activities, over 4.5% is used fully for commercial, and 20% is used partly for e-hailing services. Furthermore, results reveal that PHEVs’ average travel distance of individual groups is significantly higher than that of BEVs. The findings have a profound long-term implication on urban traffic demand with the growth of EVs. The research expands the understanding of the changes brought by transport technological evolutions to human activity patterns, and provides quantitative support for the adjustment of electric vehicle supporting policies.

This research focuses on modeling and assessing the influence of connected and automated vehicle (CAV) penetration rates on an arterial road operational performance. A 2.5-mi urban arterial road in the city of Charlotte, North Carolina, was chosen for microscopic simulation modeling using PTV VISSIM traffic simulation software and assessment in this research. The selected corridor, along N. Tryon St. from Mallard Creek Church Rd. to University Pointe Blvd., has eight intersections. During the model development process, key indicators such as driver behavior parameters, acceleration/deceleration profiles, time lead, and space headway were modified to incorporate the possible influence of non-CAVs and CAVs on arterial road operational performance. The base (or existing) scenario was first calibrated and validated using real-world travel times and turning movement counts. Hypothetical build scenarios (inclusion of CAVs) were then tested to assess the influence of varying CAV penetration rates (0%–100%, increments of 20) on operational performance measures such as travel time, delay, and the number of stops. The results indicate that CAVs are capable of improving the operational performance of urban arterial roads.

Aiming at improving the efficiency of urban intersection control, two signal control strategies based on Q-learning (QL) and deep Q-learning network (DQN), respectively. Overcoming the rough and passive defects of the traditional intersection timing control, the QL and DQN algorithm with intelligent real-time control is adopted. An algorithm framework with radar and video detector data as input, and optimal intersection control strategy as output is constructed. Based on the traffic simulation platform, a typical urban intersection is simulated and the control effect is tested. The results show that the proposed two intelligent control strategies can actively respond to different traffic states, converge in short training time, and find the optimal control strategy. QL-based control strategy and DQN-based control strategy can effectively reduce the travel time by more than 20%, and the stop delay by more than 30%. And DQN-based control strategy is more effective than the QL-based control strategy.

The transportation sector is recognized as one of the several sources of environmental problems produced by human activity in the U.S., according to the Environmental Protection Agency (EPA). The U.S. plays an outsized role: research shows that 45% of the world’s automotive CO₂ emissions come from automobiles in the U.S. Because of the pollutant emissions of internal combustion engines, the transportation sector is the main producer of greenhouse gas emissions. Researchers have considered technological solutions such as eco-driving to solve such problems. Eco-driving reduces excessive fuel consumption and greenhouse emissions by adjusting or enhancing drivers’ behaviors such as maintaining a steady speed, avoiding heavy acceleration and deceleration, anticipating the traffic flow ahead, and minimizing idling time. Several countries implemented eco-driving and demonstrated its efficient results; depending on the type of vehicle, the fuel consumption savings can be up to 30%. Implementing an eco-driving system that displays information such as average fuel efficiency and instantaneous acceleration results in a short-term reduction in fuel consumption. The result of a study using a driving simulator showed that faster but shorter acceleration/deceleration is more fuel-efficient than milder but longer acceleration/deceleration when compared over the same time span.

In this paper, three types of traffic noise barriers, interlocking steel panels, precast concrete panels, and Paragon panel 23-T, were evaluated in terms of insertion loss. Field test was conducted for the noise barriers recently installed as part of the Northwest express project in Georgia. The noise insertion loss was measured as noise difference in A-weighted decibels (dBA) immediately before and behind the barriers. To assess the effectiveness of noise barriers, the measured insertion loss was correlated with barrier types conditional upon other explanatory variables, including the separation distance of barriers from traffic, the level of traffic, wind speed, and pavement types. The results showed that under prevailing conditions, all three barrier types achieved an insertion loss in the range of 7.02 dBA to 13.58 dBA, exceeding the noise reduction design goal of 7 dBA as stated in the Georgia Department of Transportation’s noise abatement policy. Among the three barriers, the Paragon panel 23-T effected the highest insertion loss, followed by the precast concrete panel barriers and the interlocking steel panel barriers.

Intersections are key nodes of urban roads and constraints of traffic network capacity. And traffic data is the basis for intersection design, analysis, and management. The current measurement of vehicle speed is to use radar guns which could only measure point speed and volume measurement is manual counting which are time-and-energy-consuming. This paper introduces a method that collects the vehicle trajectories from videos taken by an unmanned aerial vehicle (UAV), with further processing of the trajectories to obtain useful traffic data such as intersection flow and vehicle speed. The program uses background subtraction and morphological operation for vehicle detection and uses KCF trackers to track vehicle targets. After extracting the trajectories of each vehicle in the field of view, it performs turning identification, counts the traffic flow of each entrance, and extracts more data such as vehicle speed and time headway. In order to assess the feasibility of the video processing method, case studies were performed based on the field data collected by a UAV at urban intersections in Nanjing, China. The results of these experiments proved that the approach can be applied to the efficient analysis of intersection volume, speed and other parameters based on UAV videos.

With the continuous development of intelligent transportation system, the research about the analysis and processing of road traffic flow have also launched through the prediction of future traffic flow information accurately in real time, the State and government can find appropriate controlling strategy to improve the traffic congestion, in order to make the road network unobstructed and operation efficiently. Therefore, it is of great significance to research the traffic flow prediction of urban traffic system. Based on the full consideration of the nonlinear characteristics of the traffic system, the support vector machine model and the BP neural network model optimized by genetic algorithm are used to analyze respectively the traffic flow of the road segment. Within this framework, the Bayes fusion is proposed to address the limitations of a single method used for prediction. The performances of proposed methods are evaluated by an experimental application with the measured data on the main road in Tangshan. The results show that the proposed prediction method solves the limitation of single method prediction and has higher prediction accuracy.

Connected and automated vehicle (CAV) plays an important role in the intelligent transportation system. The decision of CAV’s lane-changing behavior in heterogeneous traffic flow composed of CAV and regular vehicle (RV) becomes a key link. This paper builds a framework of lane-changing decision (LCDF) for CAV training and leaning. It consists of two parts: a micro simulation model based on cellular automata and a back-propagation neural network model that can quickly decide to change lane or still travel on the current lane. In LCDF, CAV targets a higher average speed in a certain time in the future based on what the on-board sensors and on-board communication equipment currently have access to. Then, the decision of CAV’s lane-changing behavior is trained within LCDF. In addition, normal prediction index (RMSE and MAPE) and the simulation result index (E_k and E_r) we defined were conducted to evaluate the influence of LCDF’s lane-changing decision on traffic. The results show that the efficiency of the trained CAV’s speed increases 6.7% compared to traditional model. It indicates that our framework is effective for CAV’s lane-changing decision.

This paper presents a preview of an ongoing study to develop an energy harvesting system based on thermoelectric technology embedded into the pavements structure. The asphalt pavement surface temperature in the summer can reach 60°C because of solar radiation. Two prototypes with different designs were developed to extract the heat energy from the temperature differential between the pavement surface and the subgrade soil. Designed prototype encompasses of thermoelectric generator, coolant module, heat collector, and conductor. Evaluation of prototypes was carried out through finite element analysis and laboratory testing. The results available to date suggest that this technology shows promise in powering LED traffic lights and wireless sensors embedded into pavement structures.

Travel times are an important measure for quantifying travel quality across different modes. However, collecting travel time data is a non-trivial and expensive task. Current practice involves using separate data collection methods for each mode. This paper presents a cost-effective and simple way to collect travel time data across multiple modes using media access control (MAC) matching detected by the mobile unit for sensing traffic (MUST) sensor. This technology detects personal electronic devices to determine people’s movement instead of traditional methods which detect singular modes, such as induction loops. This paper proposes a new travel-time calculation method for pedestrian, bicycle, and automobile travelers. A linear model distributes the travel time between different modes by weighting the travel time based on highest, lowest, and most likely speeds. Comparing estimated results for the modal distribution, an accuracy of approximately 83% is achieved, which is acceptable for most applications in transportation engineering.

Automated vehicles (AVs), as one of the most significant technological advancements of our century, are expected to improve traffic safety since most of the traffic crashes are related to human errors. However, their safety effectiveness in complex urban environments such as urban arterials and intersections has not yet been empirically demonstrated. Urban arterials are characterized by high traffic volumes and closely spaced signalized and unsignalized intersections such as driveways. Previous studies have found that increasing driveway density has a negative impact on highway safety because navigating in these environments can be extremely challenging. Advanced driver assistance systems (ADAS) and higher levels of automated driving systems (ADS) have the potential to overcome these challenges. This paper is developed to assess the safety impacts of AVs at urban unsignalized intersections and/or in the proximity of urban driveways. To accomplish this goal, a microsimulation model will be used to develop an urban arterial environment with an unsignalized access point. Two scenarios of a fully conventional vehicle and AV environments will be developed under various traffic levels of services (LOS). Afterward, the frequency and distribution of the conflicts will be extracted and compared for both conventional vehicles and AVs to determine how AVs will affect traffic safety under various LOS.

Pedestrian detection accuracy strongly impacts the effectiveness and reliability of intelligent pedestrian-related control systems. Traditional sensing technologies usually sense pedestrians based on the reflected signal of the transmitted infrared ray, sound wave, and electromagnetic wave which only can count the number of times that pedestrians passing a line of sight (LoS) but the moving feature monitoring, e.g., moving direction, speed, etc. For pedestrian monitoring based on computer vision-based sensing technology, the level of errors is relatively large and highly sensitive to environmental factors, such as illumination, weather conditions, and occlusion. Wi-Fi channel state information (CSI) represents the amplitudes and phases information for orthogonal frequency-division multiplexing (OFDM) subcarriers, which is mainly impacted by the static environment and moving object in surrounding areas. Previously, scholars utilized Wi-Fi CSI to analyzed multiple microscopic human movements, e.g., gesture, gait, and fall action in the indoor environment, but no application in the outdoor environment for pedestrian monitoring. The main objective of this research is to demonstrate the feasibility and reliability of the Wi-Fi CSI-based sensing method for pedestrian existence and moving direction recognition. The impacts of the CSI signal sampling ratio on the detection accuracy was investigated as well. The experiments were conducted in both indoor and outdoor environments. According to the results, the accuracy of pedestrian existence detection based on the data of the 100 Hz sampling ratio achieved 99.23% accuracy and 0.26% fast positive rate. For the moving direction recognition, the detection accuracy in the indoor environment achieved 100% and 96.92% for two directions, and got 92.21% and 93.51% in the outdoor environment. The findings of this research demonstrate the proposed Wi-Fi CSI signal is highly effective for pedestrian existence detection and moving direction recognition. The future research will continue in pedestrian moving speed estimation, overlapped pedestrian identification, and pedestrian, bicyclists, and wheelchair classification.

Many types of traffic studies require collecting field data, from traditional intersection turning movement count, travel time, queue length, and approach delay, to more detailed data in support of advanced studies like traffic conflicts, origin-destination volumes and travel times, and vehicle trajectories, etc. Many technologies are available, cost, accuracy, data collection frequency and duration, and ease of data maintenance are driving factors in selecting appropriate technologies to fulfill the needs of specific traffic studies. Aerial survey for traffic data collection can monitor a large area with few equipment and has been proven a viable approach. In recent years, drone based traffic data collection becomes increasingly popular due to its lower cost, clearer imagery, and larger field of view than ground based approach. This paper reviews the field data collection methods used in FHWA intersection field evaluation projects that involved collecting traffic data at various sized intersections. The technologies used evolved from the traditional on the ground technologies, to helicopter based aerial surveys, and drone based aerial surveys. Comparing to traditional on the ground approaches, aerial survey offers the advantages of less personnel, larger coverage area, no disruption to traffic flow, and often lower cost. The authors will discuss the potentials of drones in various types of traffic and driver behavior studies and its limitations.

This study intends to provide a thorough review of existing literature focusing on AV technologies, with respect to (1) the potential impacts of AVs on travel behavior and the transportation network; and (2) public perceptions toward AVs. Potential impacts of AVs were categorized into long-term impacts on location choices, urban development, and land use; mid-term lifestyle choices, such as auto ownership; and short-term daily travel choices, such as activity participation. Public perceptions can be viewed from several aspects, including technology adoption and willingness to pay, perceived benefits and concerns, and the perception of technologies. Besides discussing the trends in AV studies, several research gaps were recognized including the need for applying psychological models in perception analysis, taking advantage of advanced survey models like implicit association test instead of self-reporting surveys, and the necessity to use integrated land use and demand model with realistic assumptions to improve the predictions’ reliability.

Connected vehicles (CV) have the different car-following behavior from the traditional human vehicles (HV) because of the information sharing ability of CV, and it is an urgent issue to be studied that whether car-following behavior of CV can adapt to the actual complex traffic environment and improve the traffic system efficiency. Using the in-field test environment, the four conditions of start-up, acceleration, deceleration, and emergency braking are designed, and the car-following behavior based on long-term evolution-vehicle (LTE-V) communication by using in-field experimental data is studied. From the perspectives of man–machine environment, the driver’s reaction time, psychological field strength, acceleration intensity, space headway, and steady-state transfer time is proposed to establish the adaptability indicators. Using the adaptability indicators, the impact of car-following behavior of CV based on LTE-V communication under different running speed and operating conditions in actual complex traffic environment is studied considering the impact of human–machine environment. The experimental analysis results show that the car-following behavior of CV based on LTE-V communication in the actual complex traffic environment can effectively improve the safety, comfort, and stability of the whole traffic system.

Recent years have seen satellite technology progressed significantly, which enables free and efficient access to the latest satellite images in most urban areas throughout the world. Based on the existing research, this study establishes a remote sensing image dataset of China transportation hubs and land-use types within their passenger attraction range. The dataset contains 5 transportation hubs and 16 land-use child classes which are divided from eight parent categories. There are 300 images in each class for a total of 6,300. Compared with traditional remote sensing image dataset, the proposed one is mainly applied to identify and estimate passenger flow volume generated by transportation hubs in China. Therefore, the hubs and land-use categories that are closely related to passenger flow are selected. This dataset will provide data support for further development of passenger density assessment methods based on remote sensing images.

Vehicle re-identification of cross-region surveillance video is one of the current research hotspots, helping to analyze the trajectory and the travel rules of vehicles. In this paper, a feature select vehicle re-identification method without using the license plate is proposed. The method is based on a convolutional neural network (CNN) and is divided into three parts. In the first part, salient features are used in series with the original image as input for training in CNN, greatly enhancing the propagation of robust features. In the second part, the feature map selection mechanism selects and sends the features with low similarity to the next convolutional layer, reducing the propagation of irrelevant features in the network. Finally, the distance metrics between image features are compared and sorted in ascending order, as which of the same ID should be small. Experiments on datasets VRID-1 show that the proposed SSNet significantly outperforms the state-of-the-art methods for vehicle re-identification.

There are multiple vendors that now provide real-time probe data based interstate speed measurements at sub-1 mile, 1-min fidelity. This paper presents an analytical tool for visualizing interstates traffic conditions to validate congestion identified by the probe data with cameras at select locations. This visualization tool is then applied to the I-70 corridor in Indiana, an important east-west freight corridor. The tool is used to screen, at both a strategic and tactical level, the “platoon-ability” of sections, and time periods on I-70. Case studies are presented that illustrate recurring congestion, congestion associated with crashes, and congestion associated with moving workzones and maintenance operations. The 15-min median speed heat map was recommended as a quick and robust screening tool. A series of simulated trajectories, using 1- min segment speed archives, were used to demonstrate the robustness of this strategy by analyzing the frequency of speed changes greater than 10 mph as well as how often the speeds dropped below 55 mph. The paper concludes with a recommended visualization of median speeds for identifying strategic locations and time periods that platooning may be viable and a recommendation for a 10 mi real-time tactical “look ahead” visualization to identify likely areas of congestion or stopped traffic.