Browse Topic: Psychiatry and psychology

Items (516)
Pilot fatigue represents a critical concern in aviation safety, as it can significantly impair cognitive functions, decision-making abilities, and reaction times. In addition to decreasing performance, in-flight chronic fatigue has negative long-term health effects. Possible causes of fatigue include sleep loss, extended time awake, circadian phase irregularities and workload. Conventionally, the risk due to fatigue in aerospace is reduced by flight time limits and controlled rest requirements. Despite regulations limiting flight time and enabling optimal rostering, fatigue cannot be prevented completely. Hence, there is need to detect pilot fatigue in real time. There is ongoing research to detect pilot fatigue using devices that can capture Electroencephalogram (EEG) and Electrocardiogram (ECG). Though these devices have high fidelity, they are intrusive and can limit pilot activity. This limitation could potentially be overcome by non-intrusive devices such as a smart watch/wrist band/goggles which can measure physiological parameters that provide insights into pilot’s mental health. Heart rate variability (HRV) is one such physiological marker of interest for detecting pilot fatigue in real time. HRV can be effectively derived by processing raw Photoplethysmography (PPG) signals to gain insights into the autonomic nervous system, enabling the assessment of physiological state. Wearable devices such as a wristwatch are used in the current study to measure PPG data. Time and frequency domain analysis were performed to evaluate the potential of HRV indices. The analysis of R-R intervals and the Low Frequency / High Frequency (LF/HF) ratio plots, derived from HRV signals, revealed distinct characteristics that differentiate between an alert and a fatigued pilot. This study demonstrates a reliable non-intrusive method for detecting pilot fatigue and enhancing flight safety.
Nyamagoudar, VinayakP R, NamrathaRamachandran, Venkataramani
Augmented Reality (AR) and multimodal human–machine interfaces (MMI)— combining visual overlays, voice, gesture, eye- tracking, and biometric sensing—are maturing into flight-relevant technologies capable of transforming astronaut training and in-orbit operations. These interfaces can reduce task time, lower procedural errors, and mitigate cognitive workload, thereby strengthening crew autonomy and mission safety. Global operational experiences from International Space Station (ISS) augmented- reality trials and related international programs are synthesized to inform the proposed system architecture and validation framework: (i) an overview of India’s current AR/MMI-related ecosystem relevant to human spaceflight, including astronaut training pipelines and research collaborations; (ii) a mission-grade AR/MMI system architecture and multimodal fusion/decision logic suitable for human-rated operations; (iii) algorithms and programming examples for AR-driven finite-state-machine (FSM) procedures and workload-sensitive adaptation; and (iv) simulation-backed datasets across representative procedures indicating approximately 20 to 30 percent task-time reduction and approximately 40 to 50 percent error- rate reduction under controlled conditions (based on ten procedures and twenty-four simulated sessions for workload analysis). The findings reinforce that AR/MMI deployment can improve training throughput, reduce crew fatigue, and increase safety margins when designed with evidence gating, conservative confidence thresholds, and robust fallback modes. Recommendations include establishing a Human Space Flight Centre (HSFC) AR/MMI laboratory, conducting structured A/B validation trials, and committing resources for progressive demonstrations aligned with future in-orbit operations.
Yadav, Anoop Singh
Rotorcraft pilots operating in degraded visual environments encounter significant challenges during hover flight, where the absence of critical visual cues increases the risk of spatial disorientation. At low altitudes and in obstacle-rich environments, even minor losses in situational awareness can have severe consequences. Understanding the visual cues that support stable hover in good visual environments, and how their absence impacts performance and cognitive workload, is essential for mitigating these risks. This study examined key human factors in hover flight, focusing on the role of peripheral vision and microtextures in supporting pilot performance. It evaluated whether naturally relied-upon visual cues in good visual environment conditions can be artificially replicated to restore visual dominance in simulated degraded visual environments. Analysis included flight performance metrics, control inputs, physiological workload indicators, subjective assessments, and pilot feedback. The findings contribute to improved understanding of visual cueing and pilot adaptation in degraded conditions.
Mayfield, MaggieJohnson, CharlesDiMeo, Karen
This paper investigates the use of full-body vibrotactile cueing to augment operator perception during swarm teleoperation tasks. Piloted simulations are conducted in a virtual reality (VR) flight simulation environment using a quadcopter swarm model and a nonlinear dynamic inversion (NDI) flight control architecture. A scaled version of the ADS-33 slalom Mission Task Element (MTE) is implemented to evaluate swarm formation maintenance and obstacle avoidance under four experimental conditions: Good Visual Environment (GVE), Degraded Visual Environment (DVE), and each of these conditions augmented with haptic feedback. Haptic cues are delivered through vibrotactile vests and sleeves to convey information on formation deformation and gate proximity. Experimental results involving human participants indicate that haptic feedback improves formation maintenance and increases operators’ situational awareness of follower drone positions without increasing perceived mental workload. While haptic cues provided modest assistance in gate localization, visual conditions remained the dominant factor influencing obstacle avoidance performance. Overall, the results indicate that full-body haptic feedback provides an effective modality for augmenting operator perception and supporting swarm supervision tasks, particularly in visually degraded environments.
Morcos, MichaelCrane, CliftonBreed, AdamKubik, StephenGeiger, DerekLuzzani, GabrieleGary, EvanSaetti, Umberto
This paper introduces a sensorless approach for data-driven modeling of in-cabin CO2 concentration to optimize air recirculation flap control without the need for a dedicated CO2 sensor. Elevated CO2 concentrations, resulting from passenger exhalation, can impair occupants’ cognitive function and comfort. Current state-of-the-art solutions rely either on time-based control strategies, which lack responsiveness to actual cabin conditions, or on direct CO2 measurements via sensors, which increase system complexity and costs. In contrast, the proposed approach aims to replicate the benefits of sensor-based control without requiring physical sensors. In this study, a model-based methodology is presented, utilizing empirical CO2 measurement data collected from real-world test drives at varying occupancies, fan stages, vehicle speeds, and flap positions. Data acquisition involves a multi-gas analyzer positioned within the passengers’ breathing zone under controlled operation of the vehicle’s climate control unit. Based on these measurements, time-dependent CO2 concentration profiles are represented using exponential functions. These regression curves capture CO2 accumulation, depletion, and balancing behaviors, considering factors such as cabin leakage, pressure differentials at varying speeds, and ventilation conditions. These influences are inherently included in the calibration curves due to their empirical basis. The derived regression curves are implemented into a control model to simulate CO2 concentration throughout the drive, including situations where outside pollution is high and prolonged air recirculation is necessary – such as when driving through tunnels or behind trucks. On the baseline of this simulation, the sensorless control strategy adjusts flap positions accordingly, thereby minimizing both excessive CO2 buildup and unnecessary energy losses due to overventilation. By omitting CO2 sensors and relying solely on existing in-vehicle databus signals, this approach offers a cost-effective solution for cabin air quality management. Future work will focus on real-world validation of the control model and integration of exterior air quality monitoring as a complementary input.
Stürmer, MichaelGeier, BertramHofstetter, MartinHirz, Mario
Integrating intelligent and connected technologies in vehicles has significantly enriched the information environment for drivers, aiding them in making comprehensive driving decisions. However, inadequate information display may lead drivers to miss crucial information or increase their cognitive load, thereby affecting driving safety and user experience. It is essential to study drivers’ preferences for in-vehicle information display, the factors influencing these preferences, and to present information through appropriate modalities and carriers. Drawing on 695 valid questionnaire responses, this study investigates drivers’ preferences for recommendatory, explanatory, alerting, and warning information across three display modalities and six display carriers. A multivariate ordered probability model was further developed to examine the influence of user characteristics on these preferences. The results showed that drivers preferred visual cues over auditory ones, with a selection frequency that was 5.253 times higher (p < 0.001). Additionally, auditory cues were preferred 3.265 times more than tactile cues (p < 0.001). In terms of the interface, drivers favored the center console, which was preferred 1.058 times more than dashboard (p < 0.001). Furthermore, the HUD was found to be significantly better than steering wheel vibrations, being preferred 2.899 times more (p < 0.001). The study found that the choice of message type influences user preferences. Warning messages had a visual choice preference that was 1.669% higher than that for alert messages (p = 0.042). Additionally, auditory choices for alert messages were significantly enhanced, being 11.079% higher than regular messages (p < 0.001). User characteristics also played a significant role in these preferences. Women showed a lower preference for visual messages compared to men, with a ratio of 0.62 (p < 0.05). Senior drivers were less likely to choose visual dashboards, with the likelihood decreasing to 0.82 for each age group (p = 0.017). Furthermore, individuals with higher levels of education showed a preference for auditory messages, with the preference increasing to 1.23 for each education stratum (p < 0.05). The findings provide theoretical support for selecting appropriate modalities and carriers in in-vehicle information displays, particularly for tailoring displays to various information types and user groups.
He, GangDiao, KaiLuo, LongfeiXie, BingjunZhong, YixinQi, Jianping
With the advent of digital displays in driver cabins in commercial vehicles, drivers are being offered many features that convey some useful or critical information to drivers or prompt the driver to act. Due to the availability of a vast number of features, drivers face decision fatigue in choosing the appropriate features. Many are unaware of all available functionalities displayed in the Human Machine Interface (HMI) System, leading to a bare minimum usage or complete neglect of helpful features. This not only affects driving efficiency but also increases cognitive load, especially in complex driving scenarios. To alleviate the fatigue faced by drivers and to reduce the induced lethargy to choose appropriate features, we propose an AI driven recommendation agent/system that helps the driver choose the features. Instead of manually choosing between multiple settings, the driver can simply activate the recommendation mode, allowing the system to optimize selections dynamically. The novelty of this proposal focuses on introducing Intelligence in HMI Systems in such a way that it will maximize the operational usage and reduce decision fatigue in drivers. In this paper, we aim to propose a novel metric - “Decision fatigue index” to conceptualize both – the reduction in driver's cognitive load and AI models to capture, train based on the data from the driver preferences, road conditions, vehicle dynamics and user customizations. The most relevant mitigation/intervention strategies will be augmented in the HMI, which enhances ease of use, improves safety, and ensures that drivers receive the most relevant assistance.
K, SunilDhoot, Disha
In emerging markets, especially in India and other similar countries, the growing traffic density on the roads leads to different types of accidents, including frontal head-on collisions, rear-end collisions, side-impact collisions, collisions with fixed objects such as electric poles, trees, road guard rails, road dividers, and accidents involving pedestrians, cyclists, and two-wheelers. These accidents could be due to over speeding, distracted driving, violation of traffic rules, and inadequate road infrastructure etc. Providing the necessary safety restraint systems (Airbags and Seat belts) in vehicles and ensuring their robust functionality in different real-world accident scenarios will be challenging for vehicle manufacturers. It is high time to redefine the traditional collision-sensing architecture strategies with a logical approach based on a thorough study of available accident data statistics, types of objects, and scenarios leading to severe accidents. Among these, rear-end collisions (such as car-to-truck under-ride), side collisions, and head-on collision accident cases are increasing day by day. Ineffective sensing of collision signals and inadequate functionality of the safety restraint system can lead to severe injuries or fatalities. It is imperative to improvise the collision sensing system architecture and place the crash sensors inventively at optimal locations in the vehicle with an innovative approach for the early detection of collision signals and for the robust functionality of the safety restraint systems to mitigate occupant injuries and reduce fatalities to the maximum extent. This technical paper describes the thought process and methodology used to improve collision sensing techniques for the robust functionality of safety restraint systems in rear-end collisions (mainly car-to-truck underride), side impact collisions, head-on collisions, and undercarriage/underbody scraping scenarios (especially for EV battery packs). This innovative collision-sensing architecture system can be introduced in vehicles to cater to the needs of robust functionality of safety restraint systems (Airbags and Seat belts) in different real-world accident scenarios to protect vehicle users with improvisation in the existing situations while ensuring fuel cut-off in the case of ICE and high-voltage cut-off in the case of EV vehicles.
KOVALAM, SUNIL KUMAR
A low-cost, portable biosensor can quickly identify a protein whose altered levels are associated with psychiatric disorders, such as depression, schizophrenia, and bipolar disorder. When it becomes commercially available in the future, it may contribute to early detection, which is essential for treating and monitoring patients’ clinical conditions.
Lane change plays a critical role in autonomous driving and directly affects traffic safety and efficiency. Although deep learning-based lane-change decision-making frameworks have achieved promising results, they still face fundamental challenges in producing human-consistent and trustworthy behavior, mainly due to: 1) Inadequate psychology-informed personalization, as most frameworks focus on physical variables but neglect psychological factors (e.g., risk tolerance, urgency), limiting their ability to capture individual differences in lane-change motivations. 2) Limited holistic understanding of traffic context, most frameworks lack consideration of high-level and interpretable indicators (e.g., traffic pressure) in comprehensively assessing dynamic traffic scenarios, limiting their capacity for human-like contextual understanding. 3) Lack of transparent and interpretable decision logic, as many frameworks operate as black boxes with opaque reasoning processes, hindering human-aligned explanation, weakening user trust, reducing accident traceability, and impeding model refinement. To this end, a policy-oriented contextual-reasoning fuzzy neural network (POCR-FNN) is proposed as a deep learning-based decision-making framework for personalized and interpretable autonomous lane-change. First, we develop a psychology-informed driving style classification by learning distinct fuzzy membership functions to enable style-specific policy learning. Second, we design a human-inspired local interaction-aware module that estimates traffic tension by combining interaction salience and contextual risk, enhancing contextual understanding. Finally, we integrate fuzzy logic with a deep learning-based policy network to enable rule-level decision reasoning with real-time interpretability and transparent traceability. Extensive experiments on multiple public highway and urban datasets demonstrate that POCR-FNN achieves state-of-the-art performance while significantly improving personalization and interpretability across various driving styles and scenarios.
Chen, YanboChen, JiaqiYu, HuilongXi, Junqiang
As intelligent cockpit technology continues to evolve, the ways in which information is presented and interacted with within vehicle systems are becoming increasingly diverse, driving the development of driver-machine interaction toward multi-modal integration, proactive sensing, and personalized responses. As the core perception object of the intelligent cockpit, the accuracy of driver state recognition directly impacts the intelligence level of cockpit interaction and driving safety. In response to the increasing trend of task diversity and behavioral response complexity in natural driving scenarios, there is an urgent need to develop a driver multimodal data collection and processing tool with high timeliness, non-intrusiveness, and multi-source synchronization capabilities, serving as the key foundation for driver state modeling and intelligent interaction support. Based on multiple resource theory (MRT) and driver status perception mechanisms, this study designs and develops a multi-modal driver behavior and vehicle driving state data collection and processing apparatus tailored for natural driving scenarios. The apparatus adopts a model-view-view model (MVVM) architecture to achieve functional module decoupling, integrates hardware and software co-design, and incorporates key modules such as visual attention detection, hand operation tracking, and vehicle longitudinal and lateral driving state perception. It supports synchronous collection, real-time processing, and instantaneous/task-level feature extraction of multi-source heterogeneous data. The apparatus boasts excellent scalability, deployment flexibility, and interface visualization capabilities, making it suitable for typical intelligent cockpit application scenarios such as driver behavior modeling, risk identification, distracted driving detection, and human-machine interaction research. It enables comprehensive driver state perception across the entire process of “information acquisition—operation execution—behavior output,” providing high-quality data foundations and methodological support for cognitive decision-making and personalized control strategies in intelligent driving systems.
Chen, KeLi, XinyiCheng, JiahaoGuo, GangLi, Wenbo
In order to reduce traffic accidents and losses in long downhill sections of expressways, giving drivers reasonable prevention and control means of information induction can improve the safety of long downhill sections. The location of the accompanying information service of the driver's vehicle terminal and the rationality of the intervention information are worth studying. This study takes a high-speed long downhill road as an example, divides the risk level of the long downhill road based on the road safety risk index model, and verifies it with the help of driving behavior data. Secondly, three coverage schemes of sensing devices are designed according to the results of risk classification, and the HMI interface of accompanying information service is designed according to the different coverage degrees of sensing devices. Finally, a driving simulation experiment was carried out based on the driving simulator, and the speed control level, psychological comfort level, operational stability level and risk awareness level were selected to compare the impact and effect analysis of the three prevention and control schemes, and the optimal accompanying information service scheme was selected with the help of TOPSIS comprehensive evaluation method. The results show that the early-warning prevention and control information provided by the full-coverage accompanying service has a positive effect on the safety prevention and control of long downhill road sections.The results can provide theoretical support for the location and number of sensing devices in the practice of long and steep downhill projects, and provide a new perspective for the design of accompanying service terminals.
Wang, YuejiaWeng, WenzhongLuan, SenDai, Yibo
The rapid evolution of autonomy in Off-Highway Vehicles (OHVs)—spanning agriculture, mining, and construction—demands robust cybersecurity strategies. Sensor-control systems, the cognitive core of autonomous OHVs, operate in harsh, connectivity-limited environments. This paper presents a structured approach to applying threat modeling to these architectures, ensuring secure-by-design systems that uphold safety, resilience, and operational integrity.
Kotal, Amit
In recent years, the vibration comfort of automobiles has become a key consideration for consumers when purchasing vehicles. This study introduces human electrocardiogram (ECG) signals and blood pressure, and proposes a comfort prediction model based on physiological indicators. The research steps include: obtaining riding indicators and subjective feelings on flat and bumpy roads, and analyzing the differences in heart rate variability indicators and blood pressure under different road conditions through paired sample tests; playing different sound signals on bumpy roads, and using repeated measures analysis of variance to explore their impacts on physiological indicators and subjective evaluations; conducting data validity tests on the subjective evaluation results, and constructing a comfort prediction model based on correlation analysis and support vector regression algorithm. The results show that there are significant differences in indicators such as the average RR interval and standard deviation of normal-to-normal intervals (SDNN) under different riding environments; music in the frequency band of 200Hz to 600Hz can significantly improve comfort, and the average relative error of the prediction model is 8.209%. This study can provide data support for automobile manufacturers to optimize the design of suspension systems and seats. At the same time, by monitoring the physiological indicators of passengers, the vehicle system can adjust the sound signals in real time to alleviate the discomfort caused by bumps and enhance the driving experience.
Hu, LiChen, HaoWan, YeqingTian, RuiliXu, Jiahao
In order to explore the actual safety management effect of safety signs and better carry out on-site safety management, this article independently developed an evaluation scale for the management effect of safety signs. Taking a certain marine engineering equipment manufacturing enterprise as the object, the management of safety signs was evaluated and analyzed. Firstly, 11 questions from the SPSSAU online analysis scale were selected as measurement indicators to test safety label management. Factor analysis was used to select three factors: cognitive function, compliance behavior, and leadership attitude. Secondly, a safety identification management model was constructed based on structural equation modeling (SEM) with three factors as latent variable factors. Through fitting tests, it was found that cognitive effects, compliance behaviors, and leadership attitudes have a certain impact on management effectiveness, and there is a positive correlation between the three latent variable factors. Finally, taking the marine engineering equipment manufacturing enterprise as an example, the score of the enterprise’s safety label management effectiveness was analyzed. It was found that the enterprise had good safety label management, which was consistent with the on-site evaluation of experts. The results can provide reference for the safety label management of related enterprises.
Wang, ChunyuanYang, GuihuaLi, XinyaoZhu, Jie
Evaluation of integrated human-machine systems depends on having accurate human performance models. However, such models often provide only instantaneous snapshots of cognitive state and fail to account for ongoing dynamics. We argue that generative AI solutions can be used to alleviate this problem. Generative AI tools have been successful when applied to problems that have repeatable structure captured by a low-dimensional lexicon and associated with large amounts of training data. These properties apply to human performance modeling as well. Here, we introduce our Generative Cognitive Modeling Tool, a prototype human performance model developed using strategies from the generative AI community. We demonstrate the utility of our approach using simulated driving data. Our results show that cognitive states associated with driving errors are not randomized events but rather the outcome of continuous dynamics and are predictable up to 25 secs prior to the error event. We also find that the voluntary utilization of autonomous driving aids can be predicted, in part, by the disruption of ongoing dynamics. Overall, this underscores the importance of ongoing dynamics for human performance modeling and establish that generative AI approaches can provide one way to account for such factors.
Gordon, S. M.Lawhern, V. J.Touryan, J.
Chronic stress can lead to increased blood pressure and cardiovascular disease, decreased immune function, depression, and anxiety. Unfortunately, the tools we use to monitor stress are often imprecise or expensive, relying on self-reporting questionnaires and psychiatric evaluations.
A study of mental workload and the resultant cognitive-motor behavior is essential to understanding the intrinsic limitations of the human information processing system, the results of which have impact on the design of safety-critical systems. While the effects of increased task demand on mental workload and the quality of cognitive-motor performance has been previously investigated, it remains unclear how system controllability (i.e., expected handling qualities) impacts perceptual workload and performance. Furthermore, traditional EEG spectral metrics lack the temporal specificity to capture dynamic workload. Consequently, the purpose of this experiment was to examine objective brain dynamics, task performance, and subjective ratings during piloting tracking tasks of varying complexity while also challenging participants with different expected levels of handling qualities. Our results revealed a trend suggestive of increasing mental workload related to increased task complexity and varying levels of expected handling qualities. To examine dynamic operator workload with increased temporal fidelity, we introduce a time-resolved cross-correlation based approach to assess synchronous dynamics between cortical activity and behavioral performance. The findings herein highlight the practical significance of including analyses of the time domain in workload assessment, in addition to the functional utility of a combination of metrics in the study of the temporally linked cognitive-motor output associated with increased mental workload.
Hatfield, Bradley
This paper investigates the use of multi-modal cueing through full-body haptic feedback to enhance pilot-vehicle system (PVS) performance, reduce mental workload (MWL), and increase situational awareness (SA) in both good and degraded visual environments (GVE/DVE). Piloted simulations were conducted using an H-60-like flight dynamics model in a virtual reality (VR) motion-based simulator, evaluating two ADS-33-like mission task elements (MTEs) – precision hover and slalom – under visual-only and combined visual and haptic feedback conditions in both GVE and DVE. The H-60 flight dynamics were augmented with a dynamic inversion (DI)- based stability augmentation system (SAS), implementing rate-command/attitude hold (RCAH) response type on the roll, pitch, and yaw axes and altitude hold response type on the vertical axis. The SAS was designed to achieve Level 1 handling qualities per ADS-33 standards. The full-body haptic cueing strategy leveraged an outer-loop DI control law, which provided vibrotactile feedback to cue desired roll, pitch, and yaw attitudes to the pilot. Roll cues were delivered via tactors mounted on the upper arms, pitch cues via tactors on the chest and back, and yaw cues via tactors on the calves. Eight test subjects participated in the piloted simulations, including three U.S. Navy test pilots and five subjects with different flying experiences. Results indicated that haptic feedback significantly improved hover performance, reducing MWL and enhancing SA, particularly in DVE. However, in the slalom task, predefined haptic guidance misaligned with pilots’ individual control strategies, leading to performance degradation. This finding highlights the need for pilot-specific adaptive haptic feedback to mitigate inconsistencies in dynamic maneuvering tasks.
Morcos, Michael T.Saetti, UmbertoGeiger, Derek H.Kubik, Stephen T.Breed, Adam R.Crane, Clifton J.Luzzani, GabrieleFischer, Madeline R.Jun, DogyuGary, Evan
The multifaceted, fast-paced evolution in the automotive industry includes noise and vibration (NVH) behavior of products for regulatory requirements and ever-increasing customer preferences and expectations for comfort. There is pressing need for automotive engineers to explore new and advanced technologies to achieve a ‘First Time Right’ product development approach for NVH design and deliver high-quality products in shorter timeframes. Artificial Intelligence (AI) and Machine Learning (ML) are trending transformative technologies reshaping numerous industries. AI enables machines to replicate human cognitive functions, such as reasoning and decision-making, while ML, a branch of AI, employs algorithms that allow systems to learn and improve from data over time. The purpose of the paper is to show an approach of using machine learning techniques to analyze the impact of variations in structural design parameters on vehicle NVH responses. The study begins by executing the Design of Experiments (DoE) involving systematic variation of connection parameters between different vehicle subsystems employing Latin HyperCube algorithm, a statistical method for generating a near-random sample of parameter values from a multidimensional distribution. The generated designs are leveraged to train multiple machine learning models which are in turn tested against unseen data. The most accurate ML model achieved a remarkable more than 95% accuracy rate using R-squared method. This optimized ML algorithm was further employed to predict performance outcomes at arbitrary input points in space and subsequently validated against traditional Finite Element (FE) based solver (OptiStruct) output data. This framework enhances predictive accuracy and significantly accelerates the analytical workflow, empowering engineers with actionable insights for informed decision-making in structural and acoustic design processes.
Miskin, Atul R.Parmar, AzanRaj, SoniaHimakuntla, Uma Maheswar
Advancements in sensor technologies have led to increased interest in detecting and diagnosing “driver states”—collections of internal driver factors generally associated with negative driving performance, such as alcohol intoxication, cognitive load, stress, and fatigue. This is accomplished using imperfect behavioral and physiological indicators that are associated with those states. An example is the use of elevated heart rate variability, detected by a steering wheel sensor, as an indicator of frustration. Advances in sensor technologies, coupled with improvements in machine learning, have led to an increase in this research. However, a limitation is that it often excludes naturalistic driving environments, which may have conditions that affect detection. For example, reductions in visual scanning are often associated with cognitive load [1]; however, these reductions can also be related to novice driver inexperience [2] and alcohol intoxication [3]. Through our analysis of the research, we discover that the tendency to explore these singular driver states with only a comparison to “normal” driving is common. Additionally, research on interventions for these driver states is relatively scarce (fewer than 10% of cognitive load-related papers we examined assessed or discussed intervention solutions) and narrowly tailored to specific states [e.g., 4, vis-à-vis cognitive load]. States that share common behavioral and physiological markers tend to be explored independently when a more universal and integrated approach may be warranted. In this paper, we identify the need for a driver state and intervention framework that addresses these limitations by exploring state indicators and their overlap, interventions for one or multiple states, and major research gaps. Our framework offers practical approaches for handling one or many driver states, including interventions that may be deployed at different timings during a trip.
Seaman, SeanZhong, PeihanAngell, LindaDomeyer, JoshuaLenneman, John
The proliferation of intelligent technologies in the future battlefield necessitates an exploration of crew workload balancing strategies for human-machine integrated formations. Many current techniques to measure cognitive workload, through qualitative surveys or wearable sensors, are too brittle for the harsh, austere operational environments found in military settings. Non-invasive workload estimation techniques, such as those that analyze physiological effects from video feeds of the crew, present a way forward for workload-aware Soldier-machine interfaces that could trigger events – such as task reallocation – if limits on crew or individual workload are exceeded. One such technique that is being explored is the use of facial expression analysis for workload estimation. We present the performance results of regression and classification models developed from supervised machine learning algorithms that predict pNN50, a common heart rate variability metric used as a physiological measure for workload, from action unit intensity data of the Facial Action Coding System (FACS). Drawing from these results, we propose implementation recommendations for leveraging facial expressions to inform crew workload in workload-aware Soldier-machine interfaces. We conclude with a discussion on open challenges and areas of exploration for non-invasive workload estimation in military vehicle applications.
Mikulski, ChristopherRiegner, Kayla
Comprehensive requirements generation is a critical stage of the design process. Requirements are used to bound the design space and to guide the selection and evaluation of various solutions. Requirements can be categorized as either functional, defining things that the solution must do (such as produce a certain amount of horsepower), or non-functional, defining desirable qualities of the solution (such as weigh less than a particular value). Functional requirements are relatively easy to define and are often associated with particular components or subsystems within the design. As such, they can be the main focus of academic design instruction and therefore the design projects undertaken by novice designers. However, non-functional requirements (NFRs) capture important characteristics of the design solution and should not be ignored. Because of their nature, they are also difficult to assign to a particular subset of components or subsystem within the system. In this study, a group of graduate-level vehicle design students were surveyed evaluating their understanding and use of NFRs in the design process. Subsequently, they attended a lesson on defining and implementing NFRs in their specific vehicle design project. This was then followed by another survey, which gauged the impact of the targeted NFR lesson on designers’ understanding. Ultimately, the NFR lesson developed by the research team was shown to improve designer understanding of the purpose and definition of NFRs and increase their consideration of NFRs as part of the overall requirement set.
Sutton, MeredithAnbuvanan, AadithanCastanier, Matthew P.Turner, CameronKurz, Mary E.
The research activity aims at defining specific Operational Design Domains (ODDs) representative of Italian traffic environments. The paper focuses on the human-machine interaction in Automated Driving (AD), with a focus on take-over scenarios. The study, part of the European/Italian project “Interaction of Humans with Level 4 AVs in an Italian Environment - HL4IT”, describes suitable methods to investigate the effect of the Take-Over Request (TOR) on the human driver’s psychophysiological response. The DriSMI dynamic driving simulator at Politecnico di Milano has been used to analyse three different take-over situations. Participants are required to regain control of the vehicle, after a take-over request, and to navigate through a urban, suburban and highway scenario. The psychophysiological characterization of the drivers, through psychological questionnaires and physiological measures, allows for analyzing human factors in automated vehicles interactions and for contributing to advance AD technologies. Physiological signals, including electrocardiographic (ECG) and electroencephalographic (EEG) are acquired synchronously with eye-tracking and instrumented steering wheel signals throughout the entire test. The use of dynamic driving simulation enhances the study’s efficacy, facilitating early-stage development insights crucial for the advancement of AD technologies.
Gobbi, MassimilianoBoscaro, LindaDe Guglielmo, VeronicaFossati, AndreaGalbiati, AndreaMastinu, LedaPonti, MarcoMastinu, GianpieroPreviati, GiorgioSabbioni, EdoardoSignorini, Maria GabriellaSomma, AntonellaSubitoni, LucaUccello, Lorenzo
For many psychiatric patients, the traditional hospital setting can feel isolating and, at times, even stigmatizing. The high costs, limited access to care, and the disruption of being away from home often make inpatient treatment feel like a last resort.
Nowadays, cognitive distraction in the process of driving has become a frequent phenomenon, which has led to a certain proportion of traffic accidents, causing a lot of property losses and casualties. Since the fact that cognitive distraction is mostly reflected in the driver's reception and thinking of information unrelated to driving, it is difficult to recognize it from the driver's facial features. As a result, the accuracy of prediction is usually lower relying solely on facial performance to detect cognitive distraction. In this research, fifty participants took part in our simulated driving experiment. And each participant conducted the experiment in four different traffic scenarios using a high-fidelity driving simulator, including three cognitive distraction scenarios and one normal driving scenarios. Firstly, we identified the facial performance indicators and vehicle performance indicators that had a significant effect on cognitive distraction through one-way ANOVA. Then we applied the YOLOv5s model to detect cognitive distraction by combining the above two types of performance indicators. The results showed that in deep learning models, the accuracy of detecting driver cognitive distraction by combining the facial data and the vehicle data was 61.69%, and the recall rate was 83.28%, which were 8.91% and 15.1% higher than the ones using only the facial data.
Qu, ChixiongBao, QiongQu, QikaiShen, Yongjun
Biofeedback training is a technology that enhances cognitive and emotional capabilities, empowering peak performance. What sets it apart is the Biocybernetics adaptation systems, which not only collect biofeedback data but also dynamically adjust your environment based on physiological signals. Imagine surroundings adapting — changing lighting, sounds, and more — in response to your biofeedback. Traditionally confined to clinical or training rooms, the real innovation is its integration into daily life. This system offers a new level of self-regulation. Users can navigate daily life venues with real-time insights into their physiological signals, providing continuous feedback and motivation for cognitive and emotional control. Efforts yield positive surroundings, fostering well-being and peak performance.
Aerospace engineering programmes typically cover airworthiness philosophies, principles, structures, processes, and procedures. The industry has recently recognized the need to enhance the graduate engineers’ skills around airworthiness. This has led to introduction of standards acting as guides for developing curricula and content for university airworthiness courses. Concept maps, a visual mapping of concepts in a hierarchical way, enjoy wide use in engineering education (teaching and assessment). Airworthiness courses are both technical and legalistic, presenting challenges to students when it comes to understanding complex and intertwined regulations. Schematic representations of concepts can foster the cognitive processes of learning. Concept maps can assess efficiently and comprehensively a multitude of airworthiness topics. This study examines the feasibility of applying concept maps in airworthiness education. Fill-in-a-map concept maps were developed as assessment tools for an airworthiness course, covering topics in aircraft type design, production organization approvals, and maintenance documentation. They were used in tests alongside multiple-choice questions. The results were analyzed via descriptive and inferential statistics, complemented by qualitative evaluation of the concept map results. The students gained experience in concept maps, while the multiple-choice questions were present to compensate the overall test grade. Prior exposure to concept maps can assist the students to familiarize with their structure and function. The gradual increase in the concept maps’ difficulty was consistent with the increasing complexity of the airworthiness material. When concept maps are not accompanied by lists of concept words they can be challenging to complete.
Kourousis, KyriakosChatzi, Anna
The planning of mountain campus bus routes needs to take into account user demand, convenience, and other factors. This study adopts a comprehensive research method that combines quantitative and qualitative viewpoints. From the perspective of university students, this article studies the demand of campus public transportation and proposes the layout of campus bus routes in mountainous universities to meet the needs of users. The psychological needs questionnaire was used to investigate college students’ expectation of bus station service function. Taking three mountain universities as examples, the integration and selectivity of campus road networks are evaluated by using space syntax analysis, which provides valuable insights into the quality of bus stop areas. This article discusses the correlation between psychological needs assessment of college students and objective conditions of campus road network. The study concludes with the following findings: (1) The pedestrian environment quality at the university bus station in the mountain falls below the standard, highlighting a considerable disparity between the current conditions and the expectations of students and faculty. (2) Both the teaching and residential areas of the mountainous university exhibit a heightened reliance on public transportation and usage frequency. (3) Enhancements in public transport support infrastructure have the potential to significantly reduce transportation costs. (4) Considering the constraints imposed by unalterable elements of the existing pedestrian network, the positive impact of bus stop facilities and green landscapes can alleviate challenges associated with limited road network options, thus facilitating pedestrian mobility. This research lays the groundwork for further optimization of the campus layout in mountainous colleges and universities.
Duan, RanTang, RuiWang, ZhigangZhao, YixueWang, QidaYang, JiyiSu, Jiafu
This paper presents a comprehensive implementation of various Conduit frameworks designed to manage the hygiene of Simulink models in control systems and enhance them to meet industry standards such as MAB, MISRA, Polyspace, and CERT. The core challenge addressed is the minimization of repetitive work and the elimination of cognitive workload. Beginners often struggle to create Simulink models that adhere to industry standards, and keeping track of all the standards can be challenging. Given the complexity and size of these models, manual processing is time-consuming. Our Conduit frameworks help enhance their models for adherence to those standards, improving efficiency by up to 95% and utilizing machine intelligence to process large amounts of code efficiently. The Conduit frameworks also automate non value added (NVA) activities, including updates in properties of variables, checking for unwanted data types that develop during internal calculations of Simulink blocks, and variable name changes according to industry standards.
Agrawal, VipulTE, HarikrishnaN, PrajithaKumar, KosalaramanVenkat, HarishShaji, Anish
University of Utah Salt Lake City, UT
This study aims to explore the multifaceted influencing factors of market acceptance and consumer behavior of low-altitude flight services through online surveys and advanced neuroscientific methods (such as functional magnetic resonance imaging fMRI, electroencephalography EEG, functional near-infrared spectroscopy fNIRS) combined with artificial intelligence and video advertisement quantitative analysis. We conducted an in-depth study of the current trends in low-altitude flight vehicle development and customer acceptance of low-altitude services, focusing particularly on the survey methods used for market acceptance. To overcome the influence of strong opinion leaders in volunteer group experiments, we designed specialized surveys targeting broader online and social media groups. Utilizing specialized knowledge in aviation psychology, we designed a distinctive questionnaire and, within just 7 days of its launch, gathered a significant number of valid responses. The data was then analyzed using AI to provide original, insightful data on the acceptance of low-altitude services. Furthermore, we addressed the limitations of traditional manual survey methods by designing an advanced system combining EEG and AI analysis to automatically generate surveys by measuring neural and physiological responses while subjects watched video advertisements for low-altitude services. Our research offers a comparison with existing online survey forms and proposes specific predictions to potentially improve the accuracy of online surveys.
Ma, XinDing, ShuitingLi, Yan
Meet CARMEN — short for Cognitively Assistive Robot for Motivation and Neurorehabilitation — a small, tabletop robot designed to help people with mild cognitive impairment (MCI) learn skills to improve memory, attention, and executive functioning at home.
Nagoya University Nagoya, Japan
The advancements towards autonomous driving have propelled the need for reference/ground truth data for development and validation of various functionalities. Traditional data labelling methods are time consuming, skills intensive and have many drawbacks. These challenges are addressed through ALiVA (automatic lidar, image & video annotator), a semi-automated framework assisting for event detection and generation of reference data through annotation/labelling of video & point-cloud data. ALiVA is capable of processing large volumes of camera & lidar sensor data. Main pillars of framework are object detection-classification models, object tracking algorithms, cognitive algorithms and annotation results review functionality. Automatic object detection functionality creates a precise bounding box around the area of interest and assigns class labels to annotated objects. Object tracking algorithms tracks detected objects in video frames, provides a unique object id for each object and performs distance ranging. A unique feature of cognitive algorithms is the elimination of non-realistic objects of interests which appear in billboards or advertisements on buses/trucks. The framework also has a feature of event detection like overtaking scenarios or pedestrians/animals crossing the roads. Annotation review functionality is provided where assessment and correction of auto annotated data can be done manually. The results can be saved in standard file formats such as txt, csv, Json and open ASAM, ensuring compatibility across different systems. ALiVA replaces traditional annotation methods, thereby reducing the effort, the need for skilled resources and the time required to annotate large datasets. This eliminates human biases, manual errors and inconsistencies. ALiVA is validated for numerous customer requirements and offers a large amount and variety of data to quantify the benefits offered. Some of the distinguishing features are models and functionalities that are optimized for Asian road scenarios, which are typically characterized by very high road density. It is platform independent, adaptable to newer requirements, complements newer event definitions for data segmentation and works both in cloud environments for Data as a service and as a standalone desktop application.
Mardhekar, AmoghPawar, RushikeshMohod, RuchaShirudkar, RohitHivarkar, Umesh N.
Neurostimulators, also known as brain pacemakers, send electrical impulses to specific areas of the brain via special electrodes. It is estimated that some 200,000 people worldwide are now benefiting from this technology, including those who suffer from Parkinson’s disease or from pathological muscle spasms. According to Mehmet Fatih Yanik, professor of neurotechnology at ETH Zurich, further research will greatly expand the potential applications: instead of using them exclusively to stimulate the brain, the electrodes can also be used to precisely record brain activity and analyze it for anomalies associated with neurological or psychiatric disorders. In a second step, it would be conceivable in future to treat these anomalies and disorders using electrical impulses.
Engineers at the University of California San Diego in collaboration with clinicians, people with MCI, and their care partners have developed CARMEN, short for Cognitively Assistive Robot for Motivation and Neurorehabilitation — a small, tabletop robot designed to help people with mild cognitive impairment (MCI) learn skills to improve memory, attention, and executive functioning at home.
Using electrical impedance tomography (EIT), researchers have developed a system using a flexible tactile sensor for objective evaluation of fine finger movements. Demonstrating high accuracy in classifying diverse pinching motions, with discrimination rates surpassing 90 percent, this innovation holds potential in cognitive development and automated medical research.
Piloted flight involves the participation of both an air vehicle and a pilot. Modeling humans is becoming more and more important due to the growing integration of automatic systems interacting with the pilot at an ever deeper cognitive level. This research wants to offer a systematic approach to analyzing pilot behavior during a mission. The development of the pilot behavior meta-model was preceded by a careful definition of the required terminology, starting from the important assumption that human behavior is inherently teleological and therefore directed towards either explicit or implicit goals. A custom test performed with two pilots showed how humans respond to an unexpected event with a perceptible variation in muscle activation; the mechanism falls within the pilot's voluntary behavior as it aims at achieving a goal that changed following the event itself. The proposed meta-model can be adapted and used to analyze complex pilot behavior, possibly leading to new paradigms in the design of training procedures or novel cockpit layouts.
Marchesoli, DavideMasarati, PierangeloZanoni, Andrea
Temporal light modulation (TLM), colloquially known as “flicker,” is an issue in almost all lighting applications, due to widespread adoption of LED and OLED sources and their driving electronics. A subset of LED/OLED lighting systems delivers problematic TLM, often in specific types of residential, commercial, outdoor, and vehicular lighting. Dashboard displays, touchscreens, marker lights, taillights, daytime running lights (DRL), interior lighting, etc. frequently use pulse width modulation (PWM) circuits to achieve different luminances for different times of day and users’ visual adaptation levels. The resulting TLM waveforms and viewing conditions can result in distraction and disorientation, nausea, cognitive effects, and serious health consequences in some populations, occurring with or without the driver, passenger, or pedestrian consciously “seeing” the flicker. There are three visual responses to TLM: direct flicker, the stroboscopic effect, and phantom array effect (also called the “beads effect”). Metrics for the first two have limitations in both calculation and application. The phantom array effect has no established visibility measure at all, and this is the effect most associated with vehicular flicker because of the viewing conditions and frequency, plus the widespread use of PWM. Conventional wisdom from the recent past, especially concerning acceptable driver frequency ranges, needs to be reconsidered and replaced with improved guidelines to protect health and comfort. Four principal TLM waveform characteristics affect TLM visibility: frequency, modulation depth, duty cycle, and waveshape. This paper proposes much higher frequency operation if PWM control cannot be avoided; but it may be possible to modify the four principal waveform characteristics together to achieve reduced TLM visibility and improved health and comfort.
Miller, NaomiIrvin, Lia
Researchers have invented sensor-based noninvasive medical devices to make the monitoring and treatment of certain physiological and psychological conditions timelier and more precise.
Advances in healthcare and medical treatments have led to longer life expectancies in many parts of the world. As people receive better healthcare and management of other health conditions, they are more likely to reach an age where neurodegenerative diseases become a greater risk. Neurodegenerative diseases, such as Alzheimer's disease (AD), Parkinson's disease (PD), amyotrophic lateral sclerosis (ALS), and Huntington's disease (HD), are complex and can affect various aspects of a person's cognitive, motor, and sensory functions.
Effective smart cockpit interaction design can address the specific needs of children, offering ample entertainment and educational resources to enhance their on-board experience. Currently, substantial attention is focused on smart cockpit design to enrich the overall travel engagement for children. Recognizing the contrasts between children and adults in areas such as physical health, cognitive development, and emotional psychology, it becomes imperative to meticulously customize the design and optimization processes to cater explicitly to their individual requirements. However, a noticeable gap persists in both research methodologies and product offerings within this domain. This study employs user survey to delve into children’s on-board experiences and utilization of current child-centric in-cockpit interaction solutions (C-SI Solutions), that over 50% of the interviewees (children) got on-board at least several times per week and over half of the parents would pay for C-SI Solutions, but less than 8% of the interviewees reported actual usage. By employing an interdisciplinary approach that harmonizes Design Thinking and Developmental Psychology, this research reveals that the traditional cockpit is actually a liminal space for children, and introduces the ICE Model (Evaluation Model for In-Cockpit Child-Centric Interaction Solutions) for providing insights into C-SI solution design. This model is consisted of two modules: IPO-Based Structured Module and I&C (Intelligence & Consciousness) Evaluation Module. IPO-Based Structured Module is based on the IPO (Input-Process-Output) Model and for interpreting C-SI Solution’s structure, so that to realize the paradigm shift in Design Thinking. I&C Evaluation Module, the second one, is for analyzing C-SI Solution’s psychological developmental function. The ICE model is then applied to conduct market research, aiming to identify challenges and shortcomings with current C-SI Solutions. Subsequently, this research offers recommendations and possibilities for the improvement of designing C-SI Solutions, that it requires not only seamless cooperation between designers and engineers, but also interdisciplinary collaboration.
Xu, JinghanHui, XinruWang, YixiangJia, Qing
In this study, a novel assessment approach of in-vehicle speech intelligibility is presented using psychometric curves. Speech recognition performance scores were modeled at an individual listener level for a set of speech recognition data previously collected under a variety of in-vehicle listening scenarios. The model coupled an objective metric of binaural speech intelligibility (i.e., the acoustic factors) with a psychometric curve indicating the listener’s speech recognition efficiency (i.e., the listener factors). In separate analyses, two objective metrics were used with one designed to capture spatial release from masking and the other designed to capture binaural loudness. The proposed approach is in contrast to the traditional approach of relying on the speech recognition threshold, the speech level at 50% recognition performance averaged across listeners, as the metric for in-vehicle speech intelligibility. Results from the presented analyses suggest the importance of considering speech recognition accuracy across a range of signal-to-noise ratios rather than the speech recognition threshold alone, and the importance of considering individual differences among listeners when evaluating in-vehicle speech intelligibility.
Samardzic, NikolinaLavandier, MathieuShen, Yi
Simulation-Based Application of Safety of The Intended Functionality to Mitigate Foreseeable Misuse in Automated Driving SystemsSAE-PP-0037111/28/2023
The development of Automated Driving Systems (ADS) has the potential to revolutionize the transportation industry, but it also presents significant safety challenges. One of the key challenges is ensuring that the ADS is safe in the event of Foreseeable Misuse (FM) by the human driver. To address this challenge, a case study on simulation-based testing to mitigate FM by the driver using the driving simulator is presented. FM by the human driver refers to potential driving scenarios where the driver misinterprets the intended functionality of ADS, leading to hazardous behavior. Safety of the Intended Functionality (SOTIF) focuses on ensuring the absence of unreasonable risk resulting from hazardous behaviors related to functional insufficiencies caused by FM and performance limitations of sensors and machine learning-based algorithms for ADS. The simulation-based application of SOTIF to mitigate FM in ADS entails determining potential misuse scenarios, conducting simulation-based testing, and evaluating the effectiveness of measures dedicated to preventing or mitigating FM. The major contribution includes defining (i) test requirements for performing simulation-based testing of a potential misuse scenario, (ii) evaluation criteria in accordance with SOTIF requirements for implementing measures dedicated to preventing or mitigating FM, and (iii) approach to evaluate the effectiveness of the measures dedicated to preventing or mitigating FM. In conclusion, an exemplary case study incorporating driver-vehicle interface and driver interactions with ADS forming the basis for understanding the factors and causes contributing to FM is investigated. Furthermore, the test procedure for evaluating the effectiveness of the measures dedicated to preventing or mitigating FM by the driver is developed in this work.
patel, milinJung, Rolf
Artificial intelligence (AI) has become prevalent in many fields in the modern world, ranging from vacuum cleaners to lawn mowers and commercial automobiles. These capabilities are continuing to evolve and become a part of more products and systems every day, with numerous potential benefits to humans. AI is of particular interest in autonomous vehicles (AVs), where the benefits include reduced cognitive workload, increased efficiency, and improved safety for human operators. Numerous investments from academia and industry have been made recently with the intent of improving the enabling technologies for AVs. Google and Tesla are two of the more well-known examples in industry, with Google developing a self-driving car and Tesla providing its Full Self-Driving (FSD) autopilot system. Ford and BMW are also working on their own AVs.
Through connectivity with the electric grid, electric vehicles (EVs) minimize or eliminate the need for fossil fuels. Despite the rapid adoption of EVs in recent times, most government adoption objectives have not been attained. This article aims to comprehend the reasons behind the limited uptake of electric scooters in India and the driving aspects. This research used a grounded theory methodology. Using a snowball sampling technique, we conducted 25 in-depth interviews with EV owners, mainly based in Delhi and Mumbai. As an outcome of the study, four drivers and four impediments to the adoption of EVs have been formulated. The study shows that there are Financial, Technological, Operational, and Psychological drivers and Technological/Infrastructural, Operational, and Psychological impediments to the adoption. The study identifies the key concern areas in the form of categories of drivers and impediments, which can be considered in industrial and public policymaking. This research broadens our understanding of India’s uptake of EVs and provides key insights to organizations and policymakers regarding EV adoption in India.
Suri, AnkitDeepthi, B.Sharma, Yogesh
Prior investigations of swarm robot control focus on optimizing communication and coordination between agents, with at most one human control scheme, or with discrete (rather than continuous) human control schemes. In these studies, focus tends to be on human-robot interactions, including human-machine gesture interfaces, human-machine interaction during conversation, or evaluation of higher-level mental states like comfort, happiness and cognitive load. While there is early work in human control of Unmanned Arial Vehicles (UAVs) and interface design, there are few systematic studies of how human operators perceive fundamental properties of small swarms of ground-based semi-autonomous robots. Therefore, the goal of this study is to better understand how humans perceive swarms of semi-autonomous agents across a range of conditions.
This article explores the value of simulation for autonomous-vehicle research and development. There is ample research that details the effectiveness of simulation for training humans to fly and drive. Unfortunately, the same is not true for simulations used to train and test artificial intelligence (AI) that enables autonomous vehicles to fly and drive without humans. Research has shown that simulation “fidelity” is the most influential factor affecting training yield, but psychological fidelity is a widely accepted definition that does not apply to AI because it describes how well simulations engage various cognitive functions of human operators. Therefore, this investigation reviewed the literature that was published between January 2010 and May 2022 on the topic of simulation fidelity to understand how researchers are defining and measuring simulation fidelity as applied to training AI. The results reported herein illustrate that researchers are generally using agreed-upon terms such as physical fidelity, but there is an emerging definition of functional fidelity that is being adopted to replace the concept of psychological fidelity for training AI instead of humans.
Johnson, ChristopherGraupe, ElanKassel, Maxfield
Items per page:
1 – 50 of 516