<rss version="2.0" xmlns:atom="https://www.w3.org/2005/Atom">
  <channel>
    <title>Research in Progress (RIP)</title>
    <link>https://rip.trb.org/</link>
    <atom:link href="https://rip.trb.org/Record/RSS?s=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" rel="self" type="application/rss+xml" />
    <description></description>
    <language>en-us</language>
    <copyright>Copyright © 2026. National Academy of Sciences. All rights reserved.</copyright>
    <docs>http://blogs.law.harvard.edu/tech/rss</docs>
    <managingEditor>tris-trb@nas.edu (Bill McLeod)</managingEditor>
    <webMaster>tris-trb@nas.edu (Bill McLeod)</webMaster>
    <image>
      <title>Research in Progress (RIP)</title>
      <url>https://rip.trb.org/Images/PageHeader-wTitle-RIP.jpg</url>
      <link>https://rip.trb.org/</link>
    </image>
    <item>
      <title>Exploring Top-Down Visual Attention for Transportation Behavior Analysis</title>
      <link>https://rip.trb.org/View/2553152</link>
      <description><![CDATA[This project stands at the intersection of cognitive psychology, artificial intelligence (AI) and computer vision, and transportation safety and efficiency. By focusing on the nuanced ways in which humans allocate their visual attention, and how this can inform the development of AI and machine learning (ML) to aid in self-driving cars, transportation safety automation, and transportation planning and scheduling in general, this project promises to contribute significantly to the field, ensuring safer, more intuitive driving experiences, and smoother traveling experiences for the traveling public. 
By performing human behavior analysis with visual attention, the research team aims to develop best practices for safe and efficient interaction of automated roadway vehicles with existing vehicles, roadside hardware, pedestrians, cyclists, and motorcyclists. The advantages of a top-down attention approach include: prioritizing relevance, improving accuracy, enhancing machine learning efficiency, adapting models to scenarios, and enabling better human interaction. By exploring top-down visual attention, the team aims to build machine learning models to achieve the following objectives that are coherently connected with each other, where the first two will be the objectives in the base phase of this proposal and the last two would be in a second phase for a follow-on effort:
(1) Develop human behavior analysis machine learning architectures that allow autonomous driving and other transportation systems to anticipate the attention and reaction patterns of both human drivers and pedestrians, thereby preventing accidents. These include the human behavior analysis of the interaction between a driver and their vehicle, driver and pedestrians,  humans with the existing vehicles and roadside hardware. The ML architectures explored will be CNN for image encoding for improving accuracy and reducing computation, GCN for relation reasoning focusing on human interaction and actions, and transformers for self-attention and feedback.
(2) Investigate the potential of using visual attention models to improve autonomous and/or automated vehicle navigation and decision-making processes in complex environments. The visual attention mechanisms will be driven by both data and knowledge, including dynamic transportation information, roadside hardware information, location-based information (maps, events, tasks). As a start point, the team will leverage the state-of-the-art (SOTA) model  such as Analysis-by-Synthesis Vision Transformer (AbSViT) to encode feature selection, higher-level feedback and top-down input, added on the typical bottom-up process in deep models.
(3) Develop multimodal human-machine interface dashboards in self-driving cars and vehicle safety automation system, making them more intuitive for human users. These include audio, visual and haptic features as well as accessibility functions that the team has studied for helping the navigation of people who are blind or have low vision. Supported by the AI/ML-based architectures and attention models, the interface as dashboards will also allow developers, engineers and users to access the intelligent transportation systems for interaction, interpretation and diagnosis.
(4) Furthermore, collaborative opportunities may arise with existing projects, especially in applying the findings to enhance the travel pattern analysis and other safety features of self-driving and/or existing vehicles and pedestrians. Collaboration could involve sharing data, methodologies, and insights to refine autonomous driving technologies' perception and decision-making capabilities.
]]></description>
      <pubDate>Tue, 13 May 2025 19:14:57 GMT</pubDate>
      <guid>https://rip.trb.org/View/2553152</guid>
    </item>
    <item>
      <title>Teen Driving Performance Associated with Distraction, ADHD, and Other Risk Factors



</title>
      <link>https://rip.trb.org/View/1996244</link>
      <description><![CDATA[BTSCRP Research Report 15: provides insights into traffic safety risks for teen drivers with different levels of attention-deficit/hyperactivity disorder (ADHD) screen scores. The report uses naturalistic driving data to assess the incidence of eyes off road (EOR) and crash risk for teen drivers with and without ADHD. The report concludes that teen drivers with ADHD tend to look away from the road more frequently and may be at an elevated risk for missed hazards. This report will be of interest to state highway safety offices (SHSOs) and other stakeholders concerned with young-driver safety.

Young drivers with neurodevelopmental disabilities may be at more risk for motor vehicle crashes due to behavior characteristics commonly associated with these conditions. In recent years, a growing body of research has examined driving risks for teens with autism and those with ADHD. Research has identified concerns about the driving skills of teenagers with ADHD, as well as their increased tendencies to become distracted while driving and to drive at higher speeds. Determining the role of distracted driving in crashes is difficult and inexact for many reasons, including a general lack of evidence. 

Under BTSCRP Project BTS-28, “Teen Driving Performance Associated with Distraction, ADHD, and Other Risk Factors,” Virginia Polytechnic Institute and State University was asked to (1) gauge the association between confirmed instances of distracting behaviors and inattention to the driving task by teen drivers with crash and near-crash (CNC) involvement, in relation to their incidence during baseline events; (2) determine whether these instances contribute to CNCs, and if and how these relationships change with increasing driving experience; and (3) compare exposure-based CNC involvement rates, and self-reported risky driving behaviors, for teen drivers with different levels of ADHD screen scores, taking into account the potential influence of other behavioral and demographic factors captured in naturalistic driving study (NDS) data. 

]]></description>
      <pubDate>Tue, 19 Jul 2022 12:47:38 GMT</pubDate>
      <guid>https://rip.trb.org/View/1996244</guid>
    </item>
    <item>
      <title>Safety Assurance System Utilizing Visual Attention for Advanced Driver-Assistance Systems</title>
      <link>https://rip.trb.org/View/1981123</link>
      <description><![CDATA[The research team will apply their results from neuroscience and safe control to improve driver-assistance technology as follows. First, the team will study the use of visual attention information to detect risks early, before the failure to detect risk-critical obstacles can be identified from the drivers' control action and vehicle states. Second, the team will find safer control actions even when conventional methods, which are unnecessarily conservative, become infeasible in finding safe intervention strategies.]]></description>
      <pubDate>Fri, 10 Jun 2022 14:40:44 GMT</pubDate>
      <guid>https://rip.trb.org/View/1981123</guid>
    </item>
    <item>
      <title>Effects of Cognitive Load on Takeover Requests in Conditionally Automated Driving</title>
      <link>https://rip.trb.org/View/1855169</link>
      <description><![CDATA[With increases in vehicle automation, drivers can engage in non-driving related tasks while trusting automation to maintain driving control. When the vehicle issues a takeover request (TOR), drivers must disengage attention from their non-driving task to direct attention toward the task of driving. Attentional disengagement takes time, making takeover requests limited by drivers’ attentional control. In the current project, the research team investigates how the cognitive complexity, or cognitive load, of the non-driving task impacts the time to disengage attention and respond to a TOR. Specifically, is the cost in disengaging attention from a non-driving task and switching to regaining vehicle control affected by the difficulty, or cognitive load, of the non-driving related task? Participants will perform a simulated automated drive while performance a secondary non-driving task under either a high- or low-cognitive load. The team will measure the impact of the cognitive load on driving parameters (e.g., lane position) and the time and quality of the driver’s takeover.]]></description>
      <pubDate>Fri, 28 May 2021 15:21:29 GMT</pubDate>
      <guid>https://rip.trb.org/View/1855169</guid>
    </item>
    <item>
      <title>Work Zone Safety III: Calibration of Safety Notifications through Reinforcement Learning and Eye Tracking</title>
      <link>https://rip.trb.org/View/1844340</link>
      <description><![CDATA[Despite increased regulations, restrictive measures, and devices used for warnings, work zone injuries and fatalities are still observed at highway construction projects with alarms/notifications being ignored. With a vision to reduce the number of injuries and fatalities, Phase 3 of the research team's worker safety project extends the original scope and adds two new main components, including the addition of eye-tracking for identifying worker attention under dangerous situations and a reinforcement learning model used to optimally send alarms to workers to maximize their attentions along with wide deployment and demonstration of the team's previous C2Smart research effort. This phase of the project aims to make sense of the biometric sensor data (i.e., heart rate and pupil movements while workers omit or accept safety notifications) through state-of-the-art reinforcement learning approaches. The outcomes of this research will bring an understanding to the unknowns of worker behaviors on why they decide to ignore/accept notifications for calibration of when and at what frequency to send notifications to workers for a better acceptance rate. Key questions this research answers are, at what conditions workers ignore/response to warnings at work zones? How we can calibrate notification systems for getting responsive actions from workers? What are the modalities, frequencies, and timings of pushing notifications in these calibrated systems? Through wearable sensors, hardware integrated realistic representations of work zones in virtual reality and eye tracking, in this phase of the project the team will widely have pilot demonstrations of the integrated platform to collect worker behavioral and biometric (heart rate, eye-tracking) responses to alarms/warnings/notifications issued under realistic scenarios and modalities of warning mechanisms (e.g., sensory, visual, audial) that were developed in earlier phases of this project, and mine these captured data towards understanding human behaviors in response to modalities of notifications. ]]></description>
      <pubDate>Thu, 01 Apr 2021 20:02:26 GMT</pubDate>
      <guid>https://rip.trb.org/View/1844340</guid>
    </item>
    <item>
      <title>Attention and Adaptation of Teen Drivers to Driving Automation Systems</title>
      <link>https://rip.trb.org/View/1705268</link>
      <description><![CDATA[Uncovering the effect of automation on driving for teens is important because given their high crash rate, they may benefit the most from driving automation systems (DASs). However, depending on their experience and abilities, they may be the most prone to distraction and they may have difficulty effectively partnering with DASs. Given that driving is a complex process with high cognitive demands, it’s beneficial to determine how heterogeneity in adolescents’ developing attentional skills, as well as their social awareness, influence engagement with DASs. The objective of this proposed research is to: (1) quantify how teen drivers adapt to DASs over a four-week exposure period; and (2) use neural markers of prefrontal development that reflect attention skills to predict teen drivers’ initial adaptation to DASs and subsequent adaption after repeated exposure. Teen drivers (n = 40) will be recruited from Massachusetts through driving education schools. Participants will begin with a visit to PI McDermott’s laboratory for assessment of behavioral and neural measures of attention and executive functions (e.g., sustained attention, selective attention, working memory, inhibitory control, attention shifting and response monitoring). Afterwards, four driving simulator studies (to be conducted in PI Roberts’ laboratory) spaced approximately one week apart will be conducted to examine teens’ behavioral adaptation to a level 2 DAS. Drivers’ adaptation responses using surrogate driving safety measures will be entered into a one-factor (exposure number) ANOVA to determine whether exposure has an effect on driver response. For the second set of analyses, behavioral and neural measures of attention skills will serve as independent variables to predict efficiency in adaption to DAS at initial introduction as measured by the surrogate driving safety measures.]]></description>
      <pubDate>Thu, 07 May 2020 08:14:02 GMT</pubDate>
      <guid>https://rip.trb.org/View/1705268</guid>
    </item>
    <item>
      <title>The Gap Effect in Conditionally Automated Driving</title>
      <link>https://rip.trb.org/View/1705266</link>
      <description><![CDATA[Attentional disengagement and reengagement is critical in automated driving, where drivers may need to move attention to and from the dynamic driving task. This study investigates the application of the gap effect to attention in the context of automated driving. A previous SaferSim research project (Gaspar & Shull, 2019) showed that it was possible to produce the gap effect in automated driving by extinguishing a secondary task display 300ms prior to a takeover request. However, these faster look up times in the gap condition resulted in a detrimental impact in terms of subsequent takeover performance and response time. The goal of this proposal is to investigate the gap effect in the context of conditionally automated driving for younger and older adults.]]></description>
      <pubDate>Thu, 07 May 2020 08:08:18 GMT</pubDate>
      <guid>https://rip.trb.org/View/1705266</guid>
    </item>
    <item>
      <title>Evaluating the Effects of Safety Vest Color on Visual Attention in Simulated Construction Work Zones</title>
      <link>https://rip.trb.org/View/1672463</link>
      <description><![CDATA[Vehicular traffic in construction work zones presents many threats to worker safety on North Carolina Department of Transportation (NCDOT) projects. One method to ensure workers are seen by passing vehicles is for workers to don high-visibility apparel. Although research suggests that safety vests may increase the conspicuity of construction workers, no research specifically examines the effects of workers’ safety vest color on driver’s visual attention. The proposed work will address this current knowledge gap. The research activities for this proposed study will occur in two distinct phases. Phase 1 will serve as a preliminary study to test the effects of safety vest color on driver’s visual attention to worker vests in a preexisting computerized simulated construction work zone environment. For Phase 1 of this study, 81 participants will use a Trans-Sit driving simulator by Advanced Therapy Products while fitted with a Tobii  2 Pro Glasses head mounted eye tracking system and will be asked to navigate a simulated construction work zone. Each participant will be randomly assigned a set of operational parameters including safety vest color and environmental condition via a Latin square design. Specifically, safety vest color parameters will be ANSI lime-yellow, ANSI fluorescent orange-red, and no vest. Environmental parameters will include day time, night time, and inclement weather. ANOVA analysis from eye tracking metrics will be used to determine if safety vest color and/or environmental conditions affect driver’s visual attention to workers. In Phase 1, the research team will also coordinate with NCDOT to obtain physical layout dimensions of an as-built construction work zone and will collect traffic density data of current NCDOT work zones using the iCone® radar-based vehicle monitoring system developed by iCone® Products, LLC. This data will be used to generate a reconstruction of an as-built construction work zone including realistic traffic density models in the simulator environment. During Phase 2, the research team will employ the reconstructed asbuilt work zone simulation for experimental testing in a realistic setting. In Phase 2 of this study, 81 different participants will be solicited. The experimental protocol for Phase 1 and Phase 2 is identical. However, Phase 2 will use the reconstructed as-built simulation environment to increase the ecological and external validity of the study.  The primary contribution of this study lies in the exploration of how the presence of safety vests and their colors affect the ability of a driver to identify onsite workers. The results from this work will contribute to NCDOT’s goal of improving safety in work zones by providing standardized markings and colors that facilitate the visual identification of NCDOT employees. In doing so, the public will have more opportunity to avoid collisions with NCDOT employees operating in work zones.   ]]></description>
      <pubDate>Wed, 11 Dec 2019 15:54:29 GMT</pubDate>
      <guid>https://rip.trb.org/View/1672463</guid>
    </item>
    <item>
      <title>Digital Advertising Billboards and Driver Distractions</title>
      <link>https://rip.trb.org/View/1474336</link>
      <description><![CDATA[There is growing concern that roadside advertising presents a real risk to driving safety, with conservative estimates putting external distractions responsible for up to 10% of all traffic incidents. Studies indicate that anything that distracts the driver from the forward roadway for more than two seconds significantly increases the chances of crashes and near-crashes. Reports confirm that 23% of crashes and near-crashes that occur in metropolitan environments are attributable to eyes off the forward roadway greater than two seconds. Nearly 80% of the crashes and 65% of near-crashes were caused by distractions that made the driver look away for up to three seconds. Digital billboards are composed of bright light, vibrant color, and image changes or motion and are designed to pull viewer's attention to the advertisement displays. As such, these displays may divert drivers' attention from the safe operation of the car thereby causing crashes. Studies sponsored by billboard advertising companies state that the presence of digital billboards does not cause a change in driver behavior in terms of visual behavior, speed maintenance, or lane keeping. In the past, attempts have been made to show the driver's diminished attention could result in more crashes in the vicinity of such billboards, but because of methodological problems, these studies have never been done in a sufficiently reliable manner. Due to the growing debate on this issue, an objective evaluation is needed to determine if the presence of digital billboards really distracts drivers' attention and, if distraction occurs, then to what extent.  This project will study digital advertising billboards and driver distraction and will determine the correlation between the presence of digital billboards and traffic safety through literature review, crash data analysis, driver survey, empirical study using a driving simulator, and statistical analysis.]]></description>
      <pubDate>Thu, 13 Jul 2017 01:01:55 GMT</pubDate>
      <guid>https://rip.trb.org/View/1474336</guid>
    </item>
    <item>
      <title>SPR-3901: Synthesis Study: Best Practices for Maximizing Driver Attention to Work Zone Signs (End of Queue Warning Devices)</title>
      <link>https://rip.trb.org/View/1356206</link>
      <description><![CDATA[Methods for attracting driver's attention either within or in advance of those work zones will be the study focus and the intent will be to provide a database that will comprehensively review solutions that will alert drivers entering work zones from previously conducted studies.  The document will potentially assist INDOT in identifying and selecting candidate solutions for improving driver alertness for future implementation and evaluation in construction work zones.]]></description>
      <pubDate>Wed, 03 Jun 2015 01:00:36 GMT</pubDate>
      <guid>https://rip.trb.org/View/1356206</guid>
    </item>
    <item>
      <title>Driver Safety and the Boomer Dilemma: A Pilot Study for "Advanced Driver Training" as Addressing the Attention Challenge</title>
      <link>https://rip.trb.org/View/1228249</link>
      <description><![CDATA[As the post-WWII generation enters middle age and early senior adult stage, there is increasing cause for alarm--and for systematic response. Research in traffic safety and human cognition and perception indicates, as we get older, driving becomes more challenging for a number of physiological and psychological reasons. Foremost among these is attention. By some counts, 43 percent of traffic accidents are caused by inattention to the road due to people either failing to process efficaciously or accurately the environment of roadways and other vehicles or their being diverted by activities ranging from cell phone use to eating to attending to kids in the back seat. It is unlikely that such numbers will change for a group that (a) values driving as a right and not a privilege, (b) is at this time among the safest drivers, and (c) does not respond positively to messages implying "you are getting old." Our interdisciplinary team will develop a multi-tiered "Advanced Driver Training" (ADT) course that ranges from an interactive media presentation (delivered via a CD or the Web) to actual on-the-road training. Integral to such a program will be a Social Marketing campaign that links driver training to friendly and familiar boomer messages of "self-improvement" and "professional training." We seek funds to produce a pilot project that: (1) develops an ADT program and tests the efficacy of an ADT interactive presentation in terms of improving simulated driver performance; (2) creates and tests the efficacy of a persuasion campaign to convince boomer-generation drivers to seek out and participate in ADT; and (3) serves as the basis for seeking greater outside private and government funding for larger versions of the ADT program.]]></description>
      <pubDate>Thu, 03 Jan 2013 13:17:24 GMT</pubDate>
      <guid>https://rip.trb.org/View/1228249</guid>
    </item>
  </channel>
</rss>