University Transportation Centers Program

Modern control systems, like controllers for swarms of quadrotors, must satisfy complex control objectives while withstanding a wide range of disturbances, from bugs in their software to attacks on their sensors and changes in their environments. These requirements go beyond stability and tracking, and involve temporal and sequencing constraints on system response to various events. This work formalizes the requirements as formulas in Metric Temporal Logic (MTL), and designs a controller that maximizes the robustness of the MTL formula. Formally, if the system satisfies the formula with robustness r, then any disturbance of size less than r cannot cause it to violate the formula. Because robustness is not differentiable, this work provides arbitrarily precise, infinitely differentiable, approximations of it, thus enabling the use of powerful gradient descent optimizers. Experiments on a temperature control example and a two-quadrotor system demonstrate that this approach to controller design outperforms existing approaches to robustness maximization based on Mixed Integer Linear Programming and stochastic heuristics. Moreover, it is not constrained to linear systems.

Ninety percent of the world’s trade goods travel by surface transportation, using maritime, road and rail assets. The security of the goods in transit, the infrastructure supporting the movement, and the vehicles, are required to ensure that international commerce proceeds successfully. Much has been written about the surface supply chain itself, but little has focused on the security of these components. This report provides a guide for those wanting an increased understanding of the security issues that supply chain surface transportation systems confront and a blueprint to guide their future research.

Despite the sharp drop in transit ridership throughout the USA that began in March 2020, two different uses of land near transit stations continue to be implemented in the United States to promote ridership. Since 2010, transit agencies have given priority to multi-family residential construction referred to as transit oriented development (TOD), with an emphasis on housing affordability. In second place for urban planners but popular with suburban commuters is free or inexpensive parking near rail or bus transit centers, known as park-and-ride (PnR). Sometimes, TOD and PnR are combined in the same development. Public policy seeks to gain high community value from both of these land uses, and there is public interest in understanding the circumstances and locations where one of these two uses should be emphasized over the other. Multiple justifications for each are offered in the professional literature and reviewed in this report. Fundamental to the strategic decision making necessary to allocate public resources toward one use or the other is a determination of the degree to which each approach generates transit ridership. In the research reported here, econometric analysis of GIS data for transit stops, PnR locations, and residential density was employed to measure their influence on transit boardings for samples of transit stops at the main transit agencies in Seattle, Los Angeles, and San José. Results from all three cities indicate that adding 100 parking spaces close to a transit stop has a larger marginal impact than adding 100 housing units. Previous academic research estimating the higher ridership generation per floor area of PnR compared to multi-family TOD housing makes this show of strength for parking an expected finding. At the same time, this report reviews several common public policy justifications for TOD as a preferred land development emphasis near transit stations, such as revenue generation for the transit agency and providing a location for below-market affordable housing where occupants do not need to have a car. If increasing ridership is important for a transit agency, then parking for customers who want to drive to a station is an important option. There may also be additional benefits for park-and-ride in responding to the ongoing pandemic.

Despite growing interest in low-speed automated shuttles, pilot deployments have only just begun in a few places in the U.S., and there is a lack of studies that estimate the impacts of a widespread deployment of automated shuttles designed to supplement existing transit networks. This project estimated the potential impacts of automated shuttles based on a deployment scenario generated for a sample geographic area: Santa Clara County, California. The project identified sample deployment markets within Santa Clara County using a GIS screening exercise; tested the mode share changes of an automated shuttle deployment scenario using BEAM, an open-source beta software developed at the Lawrence Berkeley National Laboratory to run traffic simulations with MATSim; elaborated the model outputs within the R environment; and then estimated the related impacts. The main findings have been that the BEAM software, despite still being in its beta version, was able to model a scenario with the automated shuttle service: this report illustrates the potential of the software and the lessons learned. Regarding transportation aspects, the model estimated automated shuttle use throughout the county, with a higher rate of use in the downtown San José area. The shuttles would be preferred mainly by people who had been using gasoline-powered ride hail vehicles for A-to-B trips or going to the bus stop, as well as walking trips and a few car trips directed to public transport stops. As a result, the shuttles contributed to a small decrease in emissions of air pollutants, provided a competitive solution for short trips, and increased the overall use of the public transport system. The shuttles also presented a solution for short night trips—mainly between midnight and 2 am—when there are not many options for moving between points A and B. The conclusion is that the automated shuttle service is a good solution in certain contexts and can increase public transit ridership overall.

Nearly 5,000 people are killed and more than 418,000 are injured in weather-related traffic incidents each year. Assessments of the effectiveness of statistical models applied to crash severity prediction compared to machine learning (ML) and deep learning techniques (DL) help researchers and practitioners know what models are most effective under specific conditions. Given the class imbalance in crash data, the synthetic minority over-sampling technique for nominal (SMOTE-N) data was employed to generate synthetic samples for the minority class. The ordered logit model (OLM) and the ordered probit model (OPM) were evaluated as statistical models, while random forest (RF) and XGBoost were evaluated as ML models. For DL, multi-layer perceptron (MLP) and TabNet were evaluated. The performance of these models varied across severity levels, with property damage only (PDO) predictions performing the best and severe injury predictions performing the worst. The TabNet model performed best in predicting severe injury and PDO crashes, while RF was the most effective in predicting moderate injury crashes. However, all models struggled with severe injury classification, indicating the potential need for model refinement and exploration of other techniques. Hence, the choice of model depends on the specific application and the relative costs of false negatives and false positives. This conclusion underscores the need for further research in this area to improve the prediction accuracy of severe and moderate injury incidents, ultimately improving available data that can be used to increase road safety.

The intent of this study is to assess the readiness, resourcing, and structure of public transit agencies to identify, protect from, detect, respond to, and recover from cybersecurity vulnerabilities and threats. Given the multitude of connected devices already in use by the transit industry and the vast amount of data generated (with more coming online soon), the transit industry is vulnerable to malicious cyber-attack and other cybersecurity-related threats. This study reviews the state of best cybersecurity practices in public surface transit; outlines U.S. public surface transit operators’ cybersecurity operations; assesses U.S. policy on cybersecurity in public surface transportation; and provides policy recommendations that address gaps or identify issues for Congress, the Executive Branch, and the public surface transit agencies. Research methods include an online survey of public surface transit professionals in the United States and oral interviews conducted with members of the Executive Branch (e.g., U.S. Department of Transportation, U.S. Department of Homeland Security, The White House, and others), as well as research of literature published in periodicals.

The magnitude of the effect of adverse weather conditions on road operational performance varies with the type of weather condition and the road characteristics of the road links and adjacent links. Therefore, the relationship between weather and traffic is always a concern to traffic engineers and planners, and they have extensively explored ways to integrate weather information into transportation systems. Understanding the influence of weather on operational performance and safety helps traffic engineers and planners to proactively plan and manage transportation systems. The main objective of this research is to evaluate the effect of adverse weather conditions on travel time reliability and crash occurrence, by severity, using weather data, road data, travel time data, and crash data for North Carolina. The methodology and results from this research are useful for transportation system managers and planners to manage the traffic and improve safety under different weather conditions. They also help improve the functionality of weather-responsive management strategies like variable signs to indicate the change in reliability and safety under rainfall and low visibility conditions.

The first-mile, last-mile problem is a significant deterrent for potential transit riders, especially in suburban neighborhoods with low density. Transit agencies have typically sought to solve this problem by adding parking spaces near transit stations and adding stops to connect riders to fixed-route transit. However, these measures are often only short-term solutions. In the last few years, transit agencies have tested whether new mobility services, such as ridehailing, ridesharing, and microtransit, can offer fast, reliable connections to and from transit stations. However, there is limited research that evaluates the potential impacts of these projects. Concurrently, there is growing interest in the future of automated vehicles (AVs) and the potential of AVs to solve this first-mile problem by reducing the cost of providing these new mobility services to promote access to transit. This paper expands upon existing research to model the simulate the travel and revenue impacts of a fleet of automated vehicles that provide transit access services in the San Francisco Bay Area offered over a range of fares. The model simulates a fleet of AVs for first-mile transit access at different price points for three different service models (door-to-door ridehailing and ridesharing and meeting point ridesharing services). These service models include home-based drop-off and pick-up for single passenger service (e.g., Uber and Lyft), home-based drop-off and pick-up for multi-passenger service (e.g., microtransit), and meeting point multi-passenger service (e.g., Via).

This report summarizes the results from the twelfth year of a national public opinion survey asking U.S. adults questions related to their views on federal transportation taxes. A nationally-representative sample of 2,516 respondents completed the online survey from February 5 to 23, 2021. The questions test public opinions about raising the federal gas tax rate, replacing the federal gas tax with a new mileage fee, and imposing a mileage fee just on commercial travel. In addition to asking directly about support for these tax options, the survey collected data on respondents’ views on the quality of their local transportation system, their priorities for federal transportation spending, their knowledge about gas taxes, their views on privacy and equity matters related to mileage fees, travel behavior, and standard sociodemographic variables. This large set of variables is used to identify personal characteristics and opinions correlated with support for the tax options. Key findings include that large majorities supported transportation improvements across modes and wanted to see the federal government work towards making the transportation system well maintained, safe, and equitable, as well as to reduce the system’s impact on climate change. Findings related to gas taxes include that only 2% of respondents knew that the federal gas tax rate had not been raised in more than 20 years, and 71% of respondents supported increasing the federal gas tax by 10 cents per gallon if the revenue would be dedicated to maintenance. With respect to mileage fees, roughly half of respondents supported some form of mileage fee, whether that was assessed on all travel or just on commercial travel, 62% believe that low-income drivers should pay a reduced mileage fee rate, and 52% think that electric vehicles should pay a lower rate than gas and diesel vehicles. The analysis of trends across the survey series, which has run from 2010 to 2011, shows that support for both higher gas taxes and a hypothetical new mileage fee has risen slowly but steadily, and Americans’ experience with COVID over the past year has not disrupted those trends. Finally, support for the tax and fee options varies mostly by most personal characteristics, but there are frequently large differences correlated with age, community type, and political affiliation.

Connected and automated vehicles (CAVs) are expected to improve safety by gradually reducing human decisions while driving. However, there are still questions on their effectiveness as we transition from almost 0% CAVs to 100% CAVs with different levels of vehicle autonomy. This research focuses on synthesizing literature and identifying risk factors influencing fatal crashes involving level 1 and level 2 CAVs in the United States. Fatal crashes involving level 0 vehicles—ones that are not connected and automated—were compared to minimize unobserved heterogeneity and randomness associated with the influencing risk factors. The research team used the fatal crash data for the years 2016 to 2019 for the analysis. A partial proportionality odds model is developed using crash, road, and vehicle characteristics as the independent variables and the fatal crash involving a vehicle with a specific level of automation as the dependent variable. The results of this research indicate that level 1 and level 2 CAVs are less likely to be involved in a fatal crash at four-way intersections, on two-way routes with wide medians, at nighttime, and in poor lighting conditions when compared to level 0 vehicles. However, they are more likely than level 0 vehicles to be involved in a fatal crash with pedestrians and bicyclists. Comparative analysis between vehicles with smart features and other vehicles indicated that pedestrian automatic emergency braking (PAEB) and lane-keeping assistance (LKA) improve the safety by reducing possible collision with a pedestrian and roadside departure, respectively. Contrarily, vehicles with other smart features are still highly likely to be involved in fatal crashes. This research adds to the growing body of literature that will identify potential areas for improvement in the safety of vehicular technologies and road geometry.

Public transportation is an essential part of many older adults’ lives, but the pandemic presented new challenges for the vulnerable population. Adults aged 65 years and older experienced additional challenges, such as limited mobility options (e.g., lack of buses or trains in service due a combination of government lockdowns, fear of contracting or spreading the virus, and driver shortages in certain areas) because of the pandemic, which may have resulted in more age-related declines in perceptual, cognitive, and physical functioning. This study explores how older adults living in major metropolitan cities in the United States used and perceived public transportation during the COVID-19 pandemic. The research team conducted an online survey through the Amazon Mechanical Turk (MTurk) crowdsourcing marketplace, a platform that offers opportunities to recruit a larger number of participants from diverse geographic locations. 260 respondents completed the survey. Eligibility included: (1) residing in the United States, (2) being aged 55 years or older (the oldest age that can be selected on MTurk), and (3) having an approval rating of 90% or above (i.e., the percentage of the workers’ submitted tasks approved by survey requesters, offered by the MTurk platform). Overall, older adults reported that they had changed travel patterns since the onset of the COVID-19 pandemic, experienced challenges in using public transportation, and expressed concerns about catching the SARS-CoV-2 virus while using public transportation. Mobile technology (e.g., a transportation navigation app) was perceived as a good option for finding public transportation information, but needs improved user experience and accessibility. These findings may help transit agencies develop effective strategies for improving transportation services and increasing policymakers’ awareness of older adults’ need for accessible public transportation.

Trucks serve significant amount of freight tonnage and are more susceptible to complex interactions with other vehicles in a traffic stream. While traffic congestion continues to be a significant ‘highway’ problem, delays in truck travel result in loss of revenue to the trucking companies. There is a significant research on the traffic congestion mitigation, but a very few studies focused on data exclusive to trucks. This research is aimed at a regional-level analysis of truck travel time data to identify roads for improving mobility and reducing congestion for truck traffic. The objectives of the research are to compute and evaluate the truck travel time performance measures (by time of the day and day of the week) and use selected truck travel time performance measures to examine their correlation with on-network and off-network characteristics. Truck travel time data for the year 2019 were obtained and processed at the link level for Mecklenburg County, Wake County, and Buncombe County, NC. Various truck travel time performance measures were computed by time of the day and day of the week. Pearson correlation coefficient analysis was performed to select the average travel time (ATT), planning time index (PTI), travel time index (TTI), and buffer time index (BTI) for further analysis. On-network characteristics such as the speed limit, reference speed, annual average daily traffic (AADT), and the number of through lanes were extracted for each link. Similarly, off-network characteristics such as land use and demographic data in the near vicinity of each selected link were captured using 0.25 miles and 0.50 miles as buffer widths. The relationships between the selected truck travel time performance measures and on-network and off-network characteristics were then analyzed using Pearson correlation coefficient analysis. The results indicate that urban areas, high-volume roads, and principal arterial roads are positively correlated with the truck travel time performance measures. Further, the presence of agricultural, light commercial, heavy commercial, light industrial, single-family residential, multi-family residential, office, transportation, and medical land uses increase the truck travel time performance measures (decrease the operational performance). The methodological approach and findings can be used in identifying potential areas to serve as truck priority zones and for planning decentralized delivery locations.

This project developed a simple methodology for using Twitter data to explore public perceptions about misconduct on public transit in California. The methodology allows future researchers to analyze tweets to answer questions such as: How frequent are tweets related to assault, abuse, or other misconduct on public transit? What concerns arise most frequently? What are the types of behaviors discussed? We collected and analyzed data from Twitter posts in California about various types of public transit misconduct from January 2020 to March 2023 to identify the nature and frequency of reported misconduct. Our findings reveal that harassment, uncivil behavior, and assault are the commonly reported concerns; far fewer tweets mention obscene behavior, threats, or theft. It appears that at times the victims had been targeted on the basis of their race, gender, or sexual identity, or because they were transit employees. The tweets indicate that both genders are victimized, though women were targeted more often than men (57.5% vs. 42.5%). As for the alleged perpetrators of transit misconduct, more than three-quarters were male (78%). Transit agencies and researchers can use the results of these analyses to strategically improve safety measures for the benefit of passengers and transit operators.

Nearly 499,000 motor vehicle crashes involving trucks were reported across the United States in 2018, out of which 22% resulted in fatalities and injuries. Given the growing economy and demand for trucking in the future, it is crucial to identify the risk factors to understand where, when, and why the likelihood of getting involved in a severe or moderate injury crash with a truck is higher. This research, therefore, focuses on capturing and exploring risk factors associated with surrounding land use and demographic characteristics in addition to crash, driver, and on-network characteristics by modeling injury severity of crashes involving trucks. Crash data for Mecklenburg County in North Carolina from 2013 to 2017 was used to develop partial proportionality odds model and identify risk factors influencing injury severity of crashes involving trucks. The findings from this research indicate that dark lighting condition, inclement weather condition, the presence of double yellow or no-passing zone, road sections with speed limit >40 mph and curves, and driver fatigue, impairment, and inattention have a significant influence on injury severity of crashes involving trucks. These outcomes indicate the need for effective geometric design and improved visibility to reduce the injury severity of crashes involving trucks. The likelihood of getting involved in a crash with a truck is also high in areas with high employment, government, light commercial, and light industrial land uses. The findings can be used to proactively plan and prioritize the allocation of resources to improve safety of transportation system users in these areas.

This research focused on analyzing the association between transit service reliability indicators and ridership. Further, the effect of road network, demographic, socioeconomic, and land use characteristics on transit service reliability was analyzed. The analysis was conducted at a bus stop level. Bus arrival/departure and ridership data from the Charlotte Area Transit System (CATS) was obtained. The road network, demographic, socioeconomic, and land use characteristics were captured within 0.25-mile and 0.50-mile buffers. Pearson correlation analysis was conducted to understand the association between road network, demographic, socioeconomic, and land use characteristics and bus transit service reliability measures. The results show that bus transit service reliability has a substantial impact on ridership and is influenced by road network, demographic, socioeconomic, and land use characteristics within the bus stop vicinity. The findings help public transportation agencies to effectively utilize available resources, plan, and provide equitable services to all riders.

There is an existing issue in human-machine interaction, such that drivers of semi-autonomous vehicles are still required to take over control of the vehicle during system limitations. A possible solution may lie in tactile displays, which can present status, direction, and position information while avoiding sensory (e.g., visual and auditory) channels overload to reliably help drivers make timely decisions and execute actions to successfully take over. However, limited work has investigated the effects of meaningful tactile signals on takeover performance. This study synthesizes literature investigating the effects of tactile displays on takeover performance in automated vehicles and conducts a human-subject study to design and test the effects of six meaningful tactile signal types and two pattern durations on drivers’ perception and performance during automated driving. The research team performed a literature review of 18 articles that conducted human-subjects experiments on takeover performance utilizing tactile displays as takeover requests. Takeover performance in these studies were highlighted, such as response times, workload, and accuracy. The team then conducted a human-subject experiment, which included 16 participants that used a driving simulator to present 30 meaningful vibrotactile signals, randomly across four driving sessions measuring for reaction times (RTs), interpretation accuracy, and subjective ratings. Results from the literature suggest that tactile displays can present meaningful vibrotactile patterns via various in-vehicle locations to help improve drivers’ performance during the takeover and can be used to assist in the design of human-machine interfaces (HMI) for automated vehicles. The experiment yielded results illustrating higher urgency patterns were associated with shorter RTs and higher intuitive ratings. Also, pedestrian status and headway reduction signals presented shorter RTs and increased confidence ratings compared to other tactile signal types. Finally, the signal types that yielded the highest accuracy were the surrounding vehicle and navigation signal types. Implications of these findings may lie in informing the design of next-generation in-vehicle HMIs and future human factors studies on human-automation interactions.

There are many aspects of the transportation industry that can be focused on, but the lack of resiliency is one of the most urgent. Enhancing resiliency and creative problem-solving is essential to the industry’s growth and survival. But it cannot happen without building a more diverse workforce. Women still make up a small fraction of transportation workers, and African American and Hispanic employees are even less represented. These disparities are increasingly pronounced in many senior positions, particularly in STEM fields. Meanwhile, the public transportation industry is experiencing a severe and worsening workforce shortage and many agencies have reported substantial difficulty recruiting, retaining, and developing skilled workers. Considering the transit industry’s existing diversity and inclusion toolkits and guidelines, this project emphasizes lessons from in-depth interviews with leaders from 18 transit agencies across the country. The interviews illuminate the existing challenges and creative solutions around transit workforce diversity and inclusion. From the interviews, we discovered: 1) the critical factors that impact the current level of diversity and career mobility within transit agencies; 2) how diversity efforts help explore resources and provide opportunities for effective and robust employee engagement; and 3) the significance of evaluation systems in creating a more transparent recruitment process that initiates structural shifts, resulting in better recruiting. Moving towards inclusive and equitable workforce environments is a healing process that starts with understanding these gaps. We call this effort Healing the Workforce through Diversification.

People rely on transportation every day to access food, work, and social activities. Transportation insecurity—the lack of regular access to adequate transportation—can therefore cause significant disruptions to livelihoods. Understanding how people experience transportation insecurity in metropolitan areas may contribute to building better transportation systems and help formulate ways to alleviate persistent and underlying transportation issues. In this study, the researchers interviewed San José residents who experience transportation insecurity to better understand their experiences and identify the major ways that they cope with lack of adequate transportation. The researchers then used inductive techniques for thematic text analysis to identify patterns major themes in people's experiences and coping strategies. Findings suggest that people experience transportation insecurity as excess time consumption through congested traffic, convoluted travel schedules, and service complications, which causes worry, anxiety, and missed opportunities due to wasting or losing personal time. Overall, people's experiences and reactions allude to what could be improved in San José’s transportation infrastructure.

In the United States, public transit ridership in 2020 declined by 79% compared to 2019 levels. With lockdowns implemented during the early days of the pandemic, direct human-to-human interactions migrated to virtual platforms (e.g., Facebook, Twitter, and Reddit). Social media platforms have aided researchers in answering numerous questions about current societal dilemmas, including COVID-19. This study investigates the public’s perception of transit systems via a social media analysis given the emergence of vaccines and other COVID-19 preventive measures. Findings revealed themes of fear and confusion concerning the use of public transportation during the pandemic. The public had doubts regarding the vaccines’ impact on transportation and movement throughout 2021, with most users concerned about the proliferation of new variants. Twitter users were concerned about the travel bans placed on African countries amidst the Omicron variant and urged the government to remove the bans. These findings will help bridge the gap between public health, transport, and commuter needs by helping transportation authorities and city planners better understand the social perception of transit systems during a pandemic.

The energy-water nexus (i.e., availability of potable water and clean energy) is among the most important problems currently facing society. Ammonia is a carbon-free fuel that has the potential to reduce the carbon footprint in combustion related vehicles. However, ammonia production processes typically have their own carbon footprint and do not necessarily come from sustainable sources. This research examines wastewater filtration processes to harvest ammonia for transportation processes. The research team studied mock wastewater solutions and was able to achieve ammonia concentrations above 80%(nanofiltration) and 90% (reverse osmosis). The research team also investigated the influence of transmembrane pressure and flow rates. No degradation to the membrane integrity was observed during the process. This research used constant pressure combustion simulations to calculate the ignition delay times for NH3-air flames with expected impurities from the wastewater treatment processes. The influence of impurities, such as H2O, CO, CO2, and HCl, were studied under a range of thermodynamic conditions expected in compression ignition engines. The team observed carbon monoxide and water vapor to slightly decrease (at most 5%) ignition delay time, whereas HCl, in general, increased the ignition delay. The changes to the combustion chemistry and its influence of the reaction mechanism on the results are discussed. The experimental wastewater treatment study determined that reverse osmosis produced higher purity ammonia. The findings of the combustion work suggest that ignition delays will be similar to pure ammonia if HCl is filtered from the final product.