Detecting abnormal driving behavior is critical for road traffic safety and the evaluation of drivers’ behavior. With the advancement of machine learning (ML) algorithms and the accumulation of naturalistic driving data, many ML models have been adopted for abnormal driving behavior detection (also referred to in this paper as “anomalies”). Most existing ML-based detectors rely on (fully) supervised ML methods, which require substantial labeled data. However, ground truth labels are not always available in the real world, and labeling large amounts of data is tedious. Thus, there is a need to explore unsupervised or semi-supervised methods to make the anomaly detection process more feasible and efficient. To fill this research gap, this study analyzes large-scale real-world data revealing several abnormal driving behaviors (e.g., sudden acceleration, rapid lane-changing) and develops a hierarchical extreme learning machine (HELM)-based semi-supervised ML method using partly labeled data to accurately detect the identified abnormal driving behaviors. Moreover, previous ML-based approaches predominantly utilized basic vehicle motion features (such as velocity and acceleration) to label and detect abnormal driving behaviors, while this study seeks to introduce event-level safety indicators as input features for ML models to improve the detection performance. Results from extensive experiments demonstrate the effectiveness of the proposed semi-supervised ML model with the introduced safety indicators serving as important features. The proposed semi-supervised ML method outperforms other baseline semi-supervised or unsupervised methods as far as various metrics are concerned: for example, it delivers the best accuracy at 99.58% and the best F1-score at 0.9913. The ablation study further highlights the significance of safety indicators for advancing the detection performance of abnormal driving behaviors.
The problem of highway accidents involving snowplows has arisen again and again over the years. Snowplow operators work in a harsh environment when winter winds blow snow across the road, and snow from the plow blows over the top of the hood, further obscuring visibility. Each winter motorists travel at speeds higher than conditions dictate into conditions where visibility is reduced to nearly zero. It is during these harsh driving conditions that snowplow operators are most vulnerable to being hit by motorists. The results can be thousands of dollars in private and state vehicle damages, injuries and fatalities. A flexible moldboard plow manufactured by Frink America was evaluated for its ease of use and effectiveness at reducing the snow cloud obscuring the plow. Video and interviews of snowplow operators reveal the flexible plow is simple to use and effective at directing the snow in a narrow column off of the roadway. The capability to adjust the plow's curvature and angle to centerline on the fly allows the operator to adapt the plow to changing conditions quickly. Narrowing the discharge end of the plow directs the snow away from the plow rather than over the hood and windshield. greatly improving the operator's visibility. This allows the operator to drive faster, more closely matching the speed of traffic and plowing more roadway in a given time. The higher cost of the plow can be recovered by plowing highways in a shorter time and avoiding accidents. It is recommended that the South Dakota Department of Transportation (SDDOT) purchase flexible moldboard plows when higher highway speeds and visibility are required, and the plow's cost can be recovered within three years.
This report summarizes four Shared Transportation Goals Workshops held by the Center for Transportation Studies in April through June 2024. These included the Equity Workshop, the Climate Change and Natural Systems Workshop, the Our Region is Dynamic and Resilient Workshop and the Our Communities are Healthy and Safe Workshop. Participants were representatives of the University of Minnesota and local, regional, and state transportation professionals. Metropolitan Council's Transportation Policy Plan 2050 (TPP) was used as the scaffolding for the discussion.
High rates of speed are a large problem in construction work zones nationwide. Speeding puts the construction workers and the drivers of the vehicles in grave danger. Unfortunately, traffic accidents occur every day in rural and urban construction work zones, and some of them cause severe injuries and even death. The LaVideo system used in this study utilizes a video camera to capture the image of the speeding vehicle, and Laser to measure speed. For the Evaluation of Laser Video Speed Measurement Equipment project, four objectives were chosen to fully study the methods proposed to fix the speeding problem in construction zones. Those four objectives were: (1) To evaluate the effectiveness of the Video/Lidar system in work zones; (2) To evaluate the functional capabilities and limitations of the Video/Lidar system in work zones; (3) To evaluate the cost effectiveness of the Video/Lidar system in work zones; and (4) To develop recommendations for other applications of the Video/Lidar system. Recommendations include: (1) Much more effective if used as intended – to issue citations; (2) Much more effective use of law enforcement – Officers are free to pursue more serious matters; (3) Manufacturer must provide more user-friendly system – fewer cables; (4) Replace poorly built cables in video system; (5) Continue use of DOT COP program; and (6) Continue use of HP officers – Enforcement presence is still the most effective method.
While pedestrian safety is found in many Strategic Highway Safety Plans across the country, there may be potentially competing values and beliefs that influence the deployment of effective pedestrian safety strategies. Understanding shared values and beliefs among transportation stakeholders about pedestrian safety is critical to growing a positive traffic safety culture, deploying effective strategies to improve pedestrian safety, and ultimately achieving our nation’s goal of zero deaths on our roadways. This project sought to improve pedestrian safety by developing resources to assess and grow beliefs among transportation stakeholders to support the deployment of effective pedestrian safety strategies. This report includes a literature review, results from 10 stakeholder interviews, a survey created to reveal stakeholder beliefs about pedestrian safety and their understanding, support for, and engagement in pedestrian safety strategies, survey results, and a toolkit entitled: Tools and Resources to Improve Pedestrian Safety. The toolkit includes a resource focused on growing public participation and tools that address growing supportive beliefs to improve pedestrian safety, ways to prioritize pedestrian safety and bolster current approaches, telling your pedestrian safety story, engaging in meaningful conversations about pedestrian safety, and promoting pedestrian safety across the social environment.
Work zone safety has been a major concern for many stakeholders including the state Departments of Transportation (DOTs). To prevent DOT workers, especially truck mounted attenuator (TMA) drivers from injuries, the Autonomous Truck Mounted Attenuator (ATMA) technology is developed. To understand the current testing and deployment status of ATMA, the authors began with interviews with project managers and researchers in four state DOTs who have experience with the ATMA system. Then, the authors systematically reviewed, tested, and evaluated the ATMA system in both simulation and real-world environments under various operating scenarios, traffic conditions, and mobile work zone operations. Furthermore, Indiana Department of Transportation (INDOT) workers’ perception of using the ATMA system was also collected. Results from this study suggest that the ATMA system can be successfully implemented in most of the selected INDOT road maintenance activities with good performance. INDOT workers are comfortable to operate the ATMA system after training and welcome this new technology. They believe ATMA can improve both safety and productivity. Longer exposure to the ATMA system is a key factor impacting workers' behavior and evaluation toward ATMA. While the ATMA system has demonstrated significant benefits in saving transportation workers' lives and reducing their risk of injuries, one limitation is that the cost of the system is relatively high at the current stage based on a cost-benefit analysis.
Nationwide, the Asian American Pacific Islander (AAPI) community is projected to constitute 11 percent of people 65 years and older in the United States by 2050 (He et al., 2005). The challenges limiting the transportation and mobility of AAPI older adults include, but are not limited to, language barriers, cultural barriers, anti-Asian hate, accessibility to public transit, traffic safety and public security concerns, and changes to mobility due to the COVID-19 pandemic. This project conducted an extensive literature review and a preliminary multi-language survey in Southern California to better understand mobility-related challenges for Asian American and Pacific Islander older adults. The results of this project can provide government agencies and organizations with recommendations for policy and program changes to benefit AAPI older adults and the broader communities.
Abstract of the final report is stated below for reference: This study explored new vehicle to infrastructure (V2I) technology in construction work zones (CWZ), where speeding, unsafe driving behaviors, and drivers' failure to obey traffic signs contribute significantly to elevated accident rates and fatalities. The objective of this research to advance CWZ safety by evaluating the potential of 3-axis magnetometers attached to a moving cart and traversing over a pavement-assisted passive sensing system can improve vehicle lateral positioning and warning in CWZ. Secondly, to develop a process to implement a programmable ferromagnetic oxide material for roadway coatings to interface with vehicles containing magnetometers on a field site. The research testing used a custom-built cart equipped with multiple 3- axis magnetometer to detect EM signals from invisible markings composed of 10% and 20% CrO₂, that were created to alert for speed, lane merges, and lane-keeping. The invisible marking strips were oriented and positioned in various ways to test the repeatability and ability to reliable detect a signal and signature that could be interpreted with automated algorithm. The experimental test results were acquired in a parking and signal-processing technique was established that normalized the raw signals, removed background EM signals not related to the created EM signatures, filtered high- and low-frequency noise, and took the derivative of the EM flux density with respect to the number of points. The V2I signals in the Y and Z-axes occasionally failed to exceed the minimum threshold set for the experiments, but the X-axis signals consistently exceeded the minimum value of ±200nT throughout the testing. The minimum threshold signals were used to calculate the speed of the cart, indicate a lane merge, and determine the lateral lane position of the cart. The detected speed signals closely correlated with the GPS speed measurements on the cart as well as provided accurate cart positioning and maneuvering actions. This pilot study demonstrated the potential of V2I communication specifically EM pavement signatures to enhance CWZ safety and provide detectable and actionable feedback to the vehicle.
Ensuring an inclusive and equitable public transport system remains a major challenge for many cities. Spaces in crowded terminals and vehicles can compel riders to interact close to each other. Some evidence shows that public transport users from racial- or ethnic-minority groups, especially individuals who wear religious attire, often encounter higher rates of discrimination and harassment. The present study contributes to the limited knowledge on their travel experiences by providing insights into perceived personal security and ridership frequency, specifically focusing on riders who wear religious attire. This study uses data involving 524 participants from an online survey administered in Auckland, New Zealand. Results showed that racial and ethnic groups are dependent on public transport in their daily lives, and those who wear religious attire are often frequent riders. Findings show statistically significant differences in perceived personal security between those who wear religious attire and others. Individuals with a religious appearance express greater concern about their personal security while riding public transport and are found to avoid using it during off-peak hours. These findings provide evidence that such groups are marginalized because of their appearance. Othering behaviors do not go unnoticed and can create a hostile environment for public transport users who wear religious attire. Findings from this study add to the growing evidence that an inclusive environment is necessary for all riders to feel safe, particularly those who are dependent on public transport. Service providers have an ethical responsibility to ensure a no-tolerance culture for discrimination toward riders that explicitly protects those of marginalized identities, including those who wear religious attire.
Construction work zones (CWZ) on roadways can lead to significant safety risks for drivers and workers. Maintaining driver attention, reducing vehicle speeds, and providing advanced lane closure warnings are actions crucial for minimizing accident risks. This research study focused on the feasibility of passive sensors embedded in concrete pavement to alert drivers to maneuver their vehicles and reduce speeds in CWZ. Electromagnetic materials strategically placed or embedded in concrete pavement can alert vehicles to upcoming lane closures, provide actual to posted speed levels, warn of imminent lane departure, and help vehicles merge safely at the beginning and end of the CWZ. The proposed concrete pavement to vehicle communication (V2I) passive sensing system for CWZ also allows for in-vehicle communication (visual and audio) and engagement of the driver. The proposed passive material sensing approach still complements existing work zone signage and pavement markings to maximize overall safety of drivers and workers.
