79 research outputs found
Evaluation of Control Modalities in Highly Automated Vehicles
No full textAutonomous driving vehicles are classified by researchers into six levels, ranging from 0 to 5. The level of autonomy increases with the level number, with vehicles at levels 4 or 5 possessing the capability for full self-driving without human intervention. In high-level autonomous driving, user control model can be adapted to meet user demands since drivers are not required to focus on the road. Thus, measuring the metrics and trade-offs of control modalities under this new driving paradigm is crucial. This study proposes an evaluation framework for control modalities in level 4 and 5 autonomous vehicles, particularly in distraction scenarios.
The research comprises two parts. The first part is a user requirement study. A questionnaire, which surveyed 150 participants, investigated potential control modalities and features in self-driving vehicles. Following this, a user study that incorporated both between-participant and within-participant designs was conducted. The between-participant design aimed to compare three control modalities: physical buttons, voice, and hand gesture. Additionally, a within-participant design tested each participant's performance while being distracted. The study collected both objective and subjective data, including user error rates, physiological data, the NASA TLX rating scale, and interview feedback.
The evaluation revealed that the hand gesture control modality yielded the lowest user performance without distractions and was least affected by distractions compared to the other models. Users who engaged with the voice control modality experienced a lower error rate and workload but were also more susceptible to distractions
Toward Adaptive and User-Centered Intelligent Vehicles: AI Models with Granular Classifications for Risk Detection, Cognitive Workload, and User Preferences
No full textAs artificial intelligence (AI) increasingly integrates into our transportation systems, intelligent vehicles have emerged as research topics. Many advancements aim to enhance both the safety and comfort of drivers and the reliability of intelligent vehicles. The main focus of my research is addressing and responding to the varying states and needs of drivers, which is essential for improving driver-vehicle interactions through user-centered design. To contribute to this evolving field, this thesis explores the use of physiological signals and eye-tracking data to decode user states, perceptions, and intentions. While existing studies mostly rely on binary classification models, these approaches are limited in capturing the full spectrum of user states and needs. Addressing this gap, my research focuses on developing AI-driven models with more granular classifications for cognitive workload, risk severity levels, and user preferences for self-driving behaviours. This thesis is structured into three core domains: collision risk detection, cognitive workload estimation, and perception of user preferences for self-driving behaviours. By integrating AI techniques with multi-modal physiological data, my studies develop ML (Machine Learning) models for the domains introduced above and achieve high performance of the ML models. Feature analytical techniques are employed to enhance model interpretability for a better understanding of features and to improve the model performance. These findings pave the way for a new paradigm of intelligent vehicles that are not only more adaptive but also more aligned with user needs and preferences. This research lays the groundwork for the future development of user-centered intelligent companion systems in vehicles, where adaptive, perceptive, and interactive vehicles can better meet the complex demands of their users
Dynamic Alert Design Based on Driver’s Cognitive State for Take-over Request in Automated Vehicles
No full textThis thesis investigates the effectiveness of dynamic alert systems tailored to drivers' cognitive states in automated driving environments, focusing on enhancing takeover readiness during critical transitions. Utilizing a large-scale immersive driving simulation, the study evaluated drivers' response times and physiological measures when reacting to various alert intensities and the presence of a secondary typing task.
The experiment revealed that dynamic alerts significantly improved response times and takeover performance, especially in high-distraction scenarios. Drivers responded more effectively when alerts were adjusted to their cognitive load, with strong alerts resulting in the fastest reaction times under distracted conditions. On average, dynamic alerts reduced response times by approximately 1.75 seconds compared to static alerts. Additionally, higher lateral accelerations were observed under strong alerts, indicating more decisive maneuvering.
Self-rated attention-capturing scores were notably higher with dynamic alerts, particularly under strong alert conditions and in the presence of secondary tasks. The ANOVA results showed significant improvements in attention capturing and overall alert effectiveness when dynamic alerts were employed, demonstrating the robust design’s ability to capture attention and enhance driver responsiveness. The study confirmed that adaptive alert designs, which adjust based on the driver's cognitive state, can markedly enhance overall driving experience and safety. Participants reported higher levels of confidence with dynamic alerts, especially in scenarios involving secondary tasks. Despite the strong alerts, annoyance levels remained low, indicating that dynamic alerts are effective without causing undue stress.
These results underscore the potential of using adaptive systems to improve safety and efficiency in automated driving, advocating for a more nuanced approach to system alerts that considers the variable cognitive states of drivers. Future research should validate these findings with on-road studies, explore a broader range of alert modalities, and refine physiological monitoring techniques to further enhance adaptive alert systems
Comparing 2-level and 3-level graded collision warning systems under distracted driving conditions
No full textThis study delves into a comprehensive exploration of driver performance by comparing the effects of a 3-level graded collision warning system with those of a 2-level graded system. Employing a within-between-subject design, the experiment seeks to unravel the impact of graded warning levels (2-stage and 3-stage) on driving performance in both normal and critical driving conditions. Forty participants were recruited to undergo precise testing within a controlled driving simulator environment.
The experimental setup involves dividing participants into two groups, each exposed to distinct collision warning paradigms. The first group experiences a two-level graded warning system, while the second group encounters a three-level graded warning system, structured based on Time to Collision (TTC) metrics. Each participant drove eight scenarios, including four normal and four critical scenarios. This strategic design allows for a comprehensive evaluation of the influence of warning system intricacies on various facets of driving behavior. The study encompasses an array of dependent variables, including eye-tracking data, wristband-derived physiological metrics, driver response times, and the incidence of collisions. This multifaceted approach ensures a holistic understanding of the drivers’ reactions under different collision warning paradigms.
Results indicated that the 3-level graded system significantly reduced response times and collision frequencies compared to the 2-level system across both normal and critical driving conditions. Additionally, the 3-level system demonstrated better mitigation of driver distraction. While driving conditions did not significantly affect eye-tracking data, the warning level had a significant impact, with the 3-level system showing superior results. However, neither warning level nor driving condition significantly affected physiological data, including Electrodermal Activity (EDA), Heart Rate (HR) and Heart Rate Variability (HRV). Subjective evaluations highlighted the impact of collision warnings on driver performance, particularly in high-speed scenarios. Moreover, auditory warning modalities were preferred by a majority of participants.
These findings provide valuable insights for the development of advanced collision warning systems, emphasizing the importance of multi-level warnings and preferred warning modalities in enhancing driver safety and reducing collision risks in diverse driving environments
Trends in Electrodermal Activity, Heart Rate and Temperature during Distracted Driving among Young Novice Drivers.
No full textDriver distraction, defined as the scattering of attention from critical activities for safe driving,
is among the key globally recognized contributing factors to road crashes. The trend keeps
increasing with in-vehicle information systems and hand-held devices, leading to inattention.
Of people in all age groups, young novice teenagers are prone to the risk of road crashes and
are also more likely to exhibit risky and unsafe driving behavior. Data shows that the
involvement of distracted drivers in fatal & injury collisions is higher for people aged between
16 -34, which is about 55%. Therefore, young drivers are of great concern for the research
about driving and evaluation of safe driving conditions, which is vital in upcoming
advancements in autonomous vehicles.
Several research studies have explored the effects of distracted driving using face tracking and
eye glance monitoring. Previous research [50] did not consider much about the effect of
distraction on physiological factors and their impact during driving. The current study used
data collected from a previous thesis work titled “Detection of Driver Cognitive Distraction
Using Machine Learning Methods” by Apurva Misra and conducted new data analysis
focusing on new research questions. The main objective of this thesis is to study, identify and
discuss the effects on physiological factors like heart rate (HR), electrodermal activity (EDA),
body temperature, and motion sickness during distracted driving among young drivers.
The data was collected from a driving simulator study comprising 42 participants aged
16 – 23 under normal and distracted driving conditions. Their driving experience ranges from
0 to a maximum of 5 years. Each participant navigated six scenarios, three with distraction and the rest without distraction. Each scenario has a hidden, latent hazard depending on the
surrounding; for example, in the work zone scenario, a worker is hidden behind the bulldozer
in the work zone. The distraction task is a spoken task for which the driver has to respond
verbally, which exerts a workload similar to that observed in conversations using a hands-free
mobile phone. The physiological data collected through the Empatica4 wristband was analyzed
and compared against age, gender, driver experience, and another parameter like motion
sickness score (MSS) obtained from a questionnaire the participants completed after the
experiment. Of the physiological factors stated above, it was found that HR and EDA play a
significant role while studying distraction. Data analysis showed that HR and EDA increase
more during distraction than baseline events. Nearly 80% of drivers with 0 or 1 year of
experience tend to have a higher range of HR and EDA, which reveals that they are more
distracted than their peers with more experience. From the results of the Load index
questionnaire and Motion Sickness susceptibility questionnaire, it is inferred that when MSS
increases, there is an increase in HR and EDA. These findings will provide insights into
physiological factors for developing distraction mitigation systems or in-vehicle warning
systems for distracted drivers
Comparative Data Analysis of Older Driver's vs Younger Driver's Gap Acceptance Behavior at signalized left turns - A driving Simulator Study
No full textDrivers aged 65 and older are particularly prone to motor vehicle crashes, with approximately 20% of traffic fatalities occurring at intersections [11]. Intersections appear to be hazardous for drivers in this age group due to cognitive, perceptual, and psychomotor challenges. Older drivers find it particularly difficult to safely navigate left turns at signalized permissive intersections, having problems adequately detecting, perceiving, and accurately judging the safety of gaps. The increase in the number of elderly drivers has been paralleled by an increase in road-related accidents due to age-related fragility. By 2030, more than 21% of the adult population is projected to be over 65 years old [1]. However, previous studies have not adequately considered the combined effects of the randomized gap, queue length, traffic volume, pedestrians, and physiological factors on driving.
The current study aims to address the gap in the literature by explicitly examining older and younger drivers’ gap acceptance behaviors during permissive left turns at four-way intersections. The main objective of this thesis is to study, identify and analyze the effect of Gap Acceptance Behavior on age, traffic volume, queue length, and physiological factors such as heart rate variability (HRV), electrodermal activity (EDA), and motion sickness among older and younger drivers. The data was collected from a driving simulator study comprising 40 participants aged between 20-30 for younger and 65 years for older. The collected data was used for comparative analysis, with the Gap Accepted by the drivers calculated from the video data. The gap is calculated as the distance between the left turning vehicle and the oncoming traffic. All recruited drivers were healthy.
Each participant navigated twelve scenarios, six with lower traffic conditions and six with higher traffic conditions. Each lower and higher traffic scenario varied in queue length, with the number of cars in front of the ego vehicle varying from 0, 1, and 2. All varying queue lengths also had one with a pedestrian and another without. The physiological data collected through the Empatica4 wristband was also considered to study the gap acceptance behavior. Another parameter, motion sickness susceptibility score (MSSQ), was obtained from a questionnaire the participants completed after the experiment. Of these factors, queue length, traffic volume, and pedestrians play a significant role in studying gap acceptance. There is a significant difference in accepting and rejecting the gap between young and older drivers. Older drivers’ decision is affected more by factors, such as traffic volume, age, queue length, HRV, EDA, MSSQ score and the presence of pedestrians.
This study showed that older drivers exhibited longer gap acceptance times than their younger counterparts while turning left across traffic at permissive intersections. Researchers may use the findings to better understand gap acceptance behaviors, while policymakers may utilize the results to design mobility guidelines
Machine Learning in Driver Drowsiness Detection: A Focus on HRV, EDA, and Eye Tracking
No full textDrowsy driving continues to be a significant cause of road traffic accidents, necessi- tating the development of robust drowsiness detection systems. This research enhances our understanding of driver drowsiness by analyzing physiological indicators – heart rate variability (HRV), the percentage of eyelid closure over the pupil over time (PERCLOS), blink rate, blink percentage, and electrodermal activity (EDA) signals. Data was collected from 40 participants in a controlled scenario, with half of the group driving in a non- monotonous scenario and the other half in a monotonous scenario. Participant fatigue was assessed twice using the Fatigue Assessment Scale (FAS).
The research developed three machine learning models: HRV-Based Model, EDA- Based Model, and Eye-Based Model, achieving accuracy rates of 98.28%, 96.32%, and 90% respectively. These models were trained on the aforementioned physiological data, and their effectiveness was evaluated against a range of advanced machine learning models including GRU, Transformers, Mogrifier LSTM, Momentum LSTM, Difference Target Propagation, and Decoupled Neural Interfaces Using Synthetic Gradients.
The HRV-Based Model and EDA-Based Model demonstrated robust performance in classifying driver drowsiness. However, the Eye-Based Model had some difficulty accurately identifying instances of drowsiness, likely due to the imbalanced dataset and underrepre- sentation of certain fatigue states. The study duration, which was confined to 45 minutes, could have contributed to this imbalance, suggesting that longer data collection periods might yield more balanced datasets.
The average fatigue scores obtained from the FAS before and after the experiment showed a relatively consistent level of reported fatigue among participants, highlighting the potential impact of external factors on fatigue levels.
By integrating the outcomes of these individual models, each demonstrating strong performance, this research establishes a comprehensive and robust drowsiness detection system. The HRV-Based Model displayed remarkable accuracy, while the EDA-Based Model and the Eye-Based Model contributed valuable insights despite some limitations. The research highlights the necessity of further optimization, including more balanced data collection and investigation of individual and external factors impacting drowsiness. Despite the challenges, this work significantly contributes to the ongoing efforts to improve road safety by laying the foundation for effective real-time drowsiness detection systems and intervention methods
Age Differences in the Situation Awareness and Takeover Performance in a Semi-Autonomous Vehicle Simulator
No full textResearch on young and elderly drivers indicates a high crash risk amongst these drivers in comparison to other age groups of drivers. Young drivers have a greater propensity to adopt a risky driving style and behaviors associated with poor road safety. On the other hand, age-related declines can negatively impact the performance of older drivers on the road leading to crashes and risky maneuvers. Thus, autonomous vehicles have been suggested to improve the road safety and mobility of younger and older drivers. However, the difficulty of manually taking over control from semi-autonomous vehicles might vary in different driving conditions, particularly in those that are more challenging. Hence, the present study aims to examine the effect of road geometry and scenario, by investigating young, middle-aged and older drivers' situation awareness (SA) and takeover performance when driving a semi-autonomous vehicle simulator on a straight versus a curved road on a highway and an urban non-highway road when engaged in a secondary distracting task.
Due to the impact of COVID-19, data from only the young (n=24) and middle-aged (n=24) adults were collected and analyzed. Participants drove a Level 3 semi-autonomous simulator vehicle and performed a secondary non-driving related task in the distracted conditions. The results indicated that the participants had significantly longer hazard perception times on the curved roads and autopilot drives, but there was no significant effect of driver age and road type. Their Situation Awareness Global Assessment Technique (SAGAT) scores were higher in the highway scenarios, on the straight roads, and in the manual drive compared to the autopilot with distraction drive. Young drivers were also found to have significantly higher SAGAT scores than middle-aged drivers. While there was a significant interaction effect between road type and road geometry on takeover time, there was no significant main effect of road geometry, drive type and driver’s age. For the takeover quality metrics, road geometry and drive type had an effect on takeover performance. The resulting acceleration was higher for the straight road and in the autopilot drives, and the lane deviation was higher on the curved road and autopilot only drive compared to the autopilot with distraction drive. There was no significant main effect of road type and driver’s age on resulting acceleration and lane deviation.
Overall, while there were age differences in some aspects of SA, young and middle-aged drivers did not differ in their takeover performance. The participants' SA was impacted by the road type and geometry and their takeover quality varied according to the road geometry and drive type. The outcomes of this research will aid vehicle manufacturing companies that are developing Level 3 semi-autonomous vehicles with appropriately designing the lead time of the takeover request to meet the driving style and abilities of younger and middle-aged drivers. This will also help to improve road safety by reducing the crash rate of younger drivers
Recommended from our members
HOW MUCH DO IN-VEHICLE TASKS WITH SWAPPING, SWITCHING AND SPILLOVER EFFECTS INTERFERE WITH DRIVERS’ ABILITY TO DETECT AND RESPOND TO THREATS ON THE FORWARD ROADWAY?
No full textDistractions have long been associated with crashes. A review of the literature shows drivers engaging in secondary tasks to be three times as likely to crash as compared to attentive drivers. Although several studies report that excessively long glances away from the forward roadway elevate the risk of crashes, little research has been conducted to determine how long a driver needs to glance towards the forward roadway in between glances inside the vehicle to perform a secondary task in order to detect threats present in or emerging from the forward roadway. To determine this, drivers were asked to perform simulated in-vehicle tasks requiring glances alternating inside and outside the vehicle. The glance inside was limited to 2 s. The glance outside was varied between 1 and 4 s. Eighty five participants were evaluated across two experiments involving one continuous view and three alternating view (baseline, low load and high load) conditions. Drivers in all alternating conditions were found to detect far more hazards when the forward roadway duration between two in-vehicle glances was the longest (4 s). The decrease in hazard detection at the shorter roadway durations was a combined consequence of the drivers having to devote more resources to their driving (swapping), and having to switch their attention between the primary (driving) and secondary (in-vehicle) tasks (switching). There was an additional carry over effect of load observed in the alternating high load condition when drivers were loaded even while looking at the forward roadway (spillover). There was an effect of type of processing (bottom up versus top down) and eccentricity (central versus peripheral). The asymptotic estimation of the threshold duration indicated that the drivers’ minimum glance duration on the forward roadway be at least 4 seconds when engaged with an in-vehicle task that elicits swapping effects and at least 7 seconds when engaged with an in-vehicle task eliciting switching effects.Industrial Engineering & Operations ResearchDoctor of Philosophy (Ph.D.
Impact of Aging on Driver Preferences for Self-Driving Modes and Behaviors in Two Traffic Complexities
No full textThis study investigates age-related distinctions in preferences for self-driving vehicles, exploring their connections with traffic-related elements and individual perceptions. Analyzing two groups (23-44 and 60+ years old), the research uncovers nuanced findings that offer valuable insights for designing driver preference-based autonomous driving. The elder (Old: 60+) group, despite displaying elevated trust levels, exhibits lower preferences for self-driving compared to the young-to-middle (Y-M: 23-44) aged group. This discrepancy is highlighted alongside a significant significance between perceived difficulty and self-driving preferences in both age groups. In each traffic situation, the elder group lacks statistical significance between traffic complexity and perceived difficulty, signaling a more intricate traffic perception process. Correspondence analysis underscores age-specific preferences for extreme human-engaged or -disengaged actions, handing over control and no informing, emphasizing the significance of situation-specific considerations. Correlations between current trust and preference choices align in both groups. As a result, we suggest various design considerations that could potentially improve driver-preference factors; customizable actions, gradual automation transitions, factor-specific scaling, and specific cut-off threshold, etc. This study not only reveals age-related variations but also provides potential design principles for preference-based decision-making systems in autonomous vehicles (AVs)
- …
