Pub Date : 2023-10-25DOI: 10.1177/21695067231192435
Sakshi Pranay Taori, Sol Lim
The objective of this study was to evaluate the performance of machine learning (ML) algorithms developed using surface electromyography (EMG) armband sensor data in predicting hand-load levels (5 lb and 15 lb) from diverse lifting trials. Twelve healthy participants (six male and six female) performed repetitive lifting with three different lifting conditions, i.e., symmetric (S), asymmetric (A), and free-dynamic (F) lifts. ML models were developed with four lifting datasets (S, A, S+A, and F) and were cross-validated using F as the test dataset. Mean classification accuracy was significantly lower in models developed with the S dataset (78.8%) compared to A (83.2%) and F (83.4%). Findings indicate that the ML model developed with controlled symmetric lifts was less accurate in predicting the load of more dynamic, unconstrained lifts, which is common in real-world settings.
{"title":"Comparing Armband EMG-based Lifting Load Classification Algorithms using Various Lifting Trials","authors":"Sakshi Pranay Taori, Sol Lim","doi":"10.1177/21695067231192435","DOIUrl":"https://doi.org/10.1177/21695067231192435","url":null,"abstract":"The objective of this study was to evaluate the performance of machine learning (ML) algorithms developed using surface electromyography (EMG) armband sensor data in predicting hand-load levels (5 lb and 15 lb) from diverse lifting trials. Twelve healthy participants (six male and six female) performed repetitive lifting with three different lifting conditions, i.e., symmetric (S), asymmetric (A), and free-dynamic (F) lifts. ML models were developed with four lifting datasets (S, A, S+A, and F) and were cross-validated using F as the test dataset. Mean classification accuracy was significantly lower in models developed with the S dataset (78.8%) compared to A (83.2%) and F (83.4%). Findings indicate that the ML model developed with controlled symmetric lifts was less accurate in predicting the load of more dynamic, unconstrained lifts, which is common in real-world settings.","PeriodicalId":74544,"journal":{"name":"Proceedings of the Human Factors and Ergonomics Society ... Annual Meeting. Human Factors and Ergonomics Society. Annual meeting","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135215970","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-25DOI: 10.1177/21695067231192262
Hannah Larson, Leia Stirling
Autonomous rendezvous and docking (ARD) maneuvers are challenging tasks that require collaboration between a human and a spacecraft to be successful. As automation becomes more integrated into ARD systems, it is important to consider when and why a human may take control. Intrinsic human characteristics can influence these decisions. We consider how human spatial orientation capacity affects participants when monitoring a simulated ARD maneuver and initiating takeover when the system is perceived to be failing. Participants’ spatial reasoning capability was assessed and compared to performance in the monitoring task and perceived mental workload. While participants showed high rates of success in the task, they showed a wide range in spatial reasoning capacity and perceived mental demand. Spatial reasoning capacity did not indicate participants’ mental workload, which has implications for the human as the supervisor. These results inform future work on augmentative displays that may incorporate exocentric and egocentric views.
{"title":"Examination of Human Spatial Reasoning Capability and Simulated Autonomous Rendezvous and Docking Monitoring Performance","authors":"Hannah Larson, Leia Stirling","doi":"10.1177/21695067231192262","DOIUrl":"https://doi.org/10.1177/21695067231192262","url":null,"abstract":"Autonomous rendezvous and docking (ARD) maneuvers are challenging tasks that require collaboration between a human and a spacecraft to be successful. As automation becomes more integrated into ARD systems, it is important to consider when and why a human may take control. Intrinsic human characteristics can influence these decisions. We consider how human spatial orientation capacity affects participants when monitoring a simulated ARD maneuver and initiating takeover when the system is perceived to be failing. Participants’ spatial reasoning capability was assessed and compared to performance in the monitoring task and perceived mental workload. While participants showed high rates of success in the task, they showed a wide range in spatial reasoning capacity and perceived mental demand. Spatial reasoning capacity did not indicate participants’ mental workload, which has implications for the human as the supervisor. These results inform future work on augmentative displays that may incorporate exocentric and egocentric views.","PeriodicalId":74544,"journal":{"name":"Proceedings of the Human Factors and Ergonomics Society ... Annual Meeting. Human Factors and Ergonomics Society. Annual meeting","volume":"414 3","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135218207","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-25DOI: 10.1177/21695067231192669
Alexandra B. Proaps, Jeremiah D. Still
The lack of alignment between drivers’ and cyclists’ road-sharing knowledge results in unsafe interactions. To address this issue, educational countermeasures must clearly present and apply evidence-based practices to increase the likelihood that drivers will learn how to share the road safely with cyclists. In this study, we redesigned an existing Virginia road-sharing safety educational handbook to support a series of experiments. The redesign was based on established principles of instructional, organizational, and visual design. Virginia drivers completed a comprehension test after reviewing road-sharing educational material online. Results showed that reviewing the redesigned brochure did not improve global comprehension, law-based knowledge, and procedural knowledge about sharing the road with cyclists. However, the improved design of the educational material enhanced drivers’ declarative knowledge of road-sharing laws and safety. Further research is needed to determine the effectiveness of transferring these design choices to other transportation domains, so policymakers and instructors can effectively prioritize approaches for improving road safety.
{"title":"Redesigning Educational Countermeasures to Increase Virginia Drivers’ Road-Sharing Safety Knowledge","authors":"Alexandra B. Proaps, Jeremiah D. Still","doi":"10.1177/21695067231192669","DOIUrl":"https://doi.org/10.1177/21695067231192669","url":null,"abstract":"The lack of alignment between drivers’ and cyclists’ road-sharing knowledge results in unsafe interactions. To address this issue, educational countermeasures must clearly present and apply evidence-based practices to increase the likelihood that drivers will learn how to share the road safely with cyclists. In this study, we redesigned an existing Virginia road-sharing safety educational handbook to support a series of experiments. The redesign was based on established principles of instructional, organizational, and visual design. Virginia drivers completed a comprehension test after reviewing road-sharing educational material online. Results showed that reviewing the redesigned brochure did not improve global comprehension, law-based knowledge, and procedural knowledge about sharing the road with cyclists. However, the improved design of the educational material enhanced drivers’ declarative knowledge of road-sharing laws and safety. Further research is needed to determine the effectiveness of transferring these design choices to other transportation domains, so policymakers and instructors can effectively prioritize approaches for improving road safety.","PeriodicalId":74544,"journal":{"name":"Proceedings of the Human Factors and Ergonomics Society ... Annual Meeting. Human Factors and Ergonomics Society. Annual meeting","volume":"83 3","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135112760","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-25DOI: 10.1177/21695067231192528
Keum Joo Kim, Eugene Santos
Humans learn from both successful and unsuccessful experiences, because useful information about how to solve complex problems can be gleaned not only from success but also from failure. In this paper, we propose a method for investigating this difference by applying Preference based Inverse Reinforcement Learning to Double Transition Models built from replays of StarCraft II. Our method provides two advantages: (1) the ability to identify integrated reward distributions from computational models composed of multiple experiences, and (2) the ability to discern differences between learning by successes and failures. Our experimental results demonstrate that reward distributions are shaped depending on the trajectories utilized to build models. Reward distributions based on successful episodes were skewed to the left, while those based on unsuccessful episodes were skewed to the right. Furthermore, we found that players with symmetric triple reward distributions had a high probability of winning the game.
{"title":"Learning by Successful or Unsuccessful Experiences?","authors":"Keum Joo Kim, Eugene Santos","doi":"10.1177/21695067231192528","DOIUrl":"https://doi.org/10.1177/21695067231192528","url":null,"abstract":"Humans learn from both successful and unsuccessful experiences, because useful information about how to solve complex problems can be gleaned not only from success but also from failure. In this paper, we propose a method for investigating this difference by applying Preference based Inverse Reinforcement Learning to Double Transition Models built from replays of StarCraft II. Our method provides two advantages: (1) the ability to identify integrated reward distributions from computational models composed of multiple experiences, and (2) the ability to discern differences between learning by successes and failures. Our experimental results demonstrate that reward distributions are shaped depending on the trajectories utilized to build models. Reward distributions based on successful episodes were skewed to the left, while those based on unsuccessful episodes were skewed to the right. Furthermore, we found that players with symmetric triple reward distributions had a high probability of winning the game.","PeriodicalId":74544,"journal":{"name":"Proceedings of the Human Factors and Ergonomics Society ... Annual Meeting. Human Factors and Ergonomics Society. Annual meeting","volume":"33 8","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135112838","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-25DOI: 10.1177/21695067231192530
Nathan Sanders, Elif Sener, Karen B. Chen
People are increasingly using virtual reality (VR) for work. As a result of extended use, fatigue and musculoskeletal disorders affecting the upper arms and shoulders are already becoming common among VR users. This pilot study presented a “virtual working area” (VWA) to reduce the risk of fatigue resulting from using gestures obtained in gesture elicitation studies, and explored how the distance to the user interface (UI) interacted with different functions (select, scroll) during a mock reading task. Results showed that keeping the hands within the VWA had the potential to reduce Rapid Upper-Body Limb Assessment (RULA) and Borg CR10 scores at clinically significant levels. Scores were worse when the UI was far away and for the select function, suggesting the design of virtual UIs can play a role in eliciting naturalistic yet ergonomic interactions. The results also provide effect sizes and variance estimates to plan future work.
{"title":"Eliciting Ergonomic User-Defined Gestures for Virtual Reality: A Pilot Study","authors":"Nathan Sanders, Elif Sener, Karen B. Chen","doi":"10.1177/21695067231192530","DOIUrl":"https://doi.org/10.1177/21695067231192530","url":null,"abstract":"People are increasingly using virtual reality (VR) for work. As a result of extended use, fatigue and musculoskeletal disorders affecting the upper arms and shoulders are already becoming common among VR users. This pilot study presented a “virtual working area” (VWA) to reduce the risk of fatigue resulting from using gestures obtained in gesture elicitation studies, and explored how the distance to the user interface (UI) interacted with different functions (select, scroll) during a mock reading task. Results showed that keeping the hands within the VWA had the potential to reduce Rapid Upper-Body Limb Assessment (RULA) and Borg CR10 scores at clinically significant levels. Scores were worse when the UI was far away and for the select function, suggesting the design of virtual UIs can play a role in eliciting naturalistic yet ergonomic interactions. The results also provide effect sizes and variance estimates to plan future work.","PeriodicalId":74544,"journal":{"name":"Proceedings of the Human Factors and Ergonomics Society ... Annual Meeting. Human Factors and Ergonomics Society. Annual meeting","volume":"8 2","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135112854","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-25DOI: 10.1177/21695067231199689
Faiza Tazi, Archana Nandakumar, Josiah Dykstra, Prashanth Rajivan, Sanchari Das
The COVID-19 pandemic has significantly transformed the healthcare sector, with telehealth services being among the most prominent changes. The adoption of telehealth services, however, has raised new challenges, particularly in the areas of security and privacy. To better comprehend the telehealth needs and concerns of medical professionals, particularly those in private practice, we conducted a study comprised of 20 semi-structured interviews with telehealth practitioners in audiology and speech therapy. Our findings indicate that private telehealth practitioners encounter difficult choices when it comes to balancing security, privacy, usability, and accessibility, particularly while caring for vulnerable populations. Additionally, the study revealed that practitioners face challenges in ensuring HIPAA compliance due to inadequate resources and a lack of technological comprehension. Policymakers and healthcare providers should take proactive measures to address these challenges, including offering resources and training to ensure HIPAA compliance and enhancing technology infrastructure to support secure and accessible telehealth.
{"title":"Privacy, Security, and Usability Tradeoffs of Telehealth from Practitioners’ Perspectives","authors":"Faiza Tazi, Archana Nandakumar, Josiah Dykstra, Prashanth Rajivan, Sanchari Das","doi":"10.1177/21695067231199689","DOIUrl":"https://doi.org/10.1177/21695067231199689","url":null,"abstract":"The COVID-19 pandemic has significantly transformed the healthcare sector, with telehealth services being among the most prominent changes. The adoption of telehealth services, however, has raised new challenges, particularly in the areas of security and privacy. To better comprehend the telehealth needs and concerns of medical professionals, particularly those in private practice, we conducted a study comprised of 20 semi-structured interviews with telehealth practitioners in audiology and speech therapy. Our findings indicate that private telehealth practitioners encounter difficult choices when it comes to balancing security, privacy, usability, and accessibility, particularly while caring for vulnerable populations. Additionally, the study revealed that practitioners face challenges in ensuring HIPAA compliance due to inadequate resources and a lack of technological comprehension. Policymakers and healthcare providers should take proactive measures to address these challenges, including offering resources and training to ensure HIPAA compliance and enhancing technology infrastructure to support secure and accessible telehealth.","PeriodicalId":74544,"journal":{"name":"Proceedings of the Human Factors and Ergonomics Society ... Annual Meeting. Human Factors and Ergonomics Society. Annual meeting","volume":"22 6","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135112902","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-25DOI: 10.1177/21695067231192432
Randall Spain, Benjamin Goldberg, Shannon Bailey, Stephanie Fussell, Allison Bayro, Kelly Hale, Aaron Jones, Rachel Regina, Bob Thomas, Kevin Owens, Nathan Lau, Abhraneil Dam, Karen Chen, Luke Sturgeon, Monifa Vaughn-Cooke, Nuela Chidubem Enebechi
This alternative format session provides a forum for human factors scholars and practitioners to showcase how state-of-the-art extended reality (XR) applications are being used in academia, defense, and industry to address human factors research. The session will begin with short introductions from each presenter to describe their XR application. Afterward, session attendees will engage with the presenters and their demonstrations, which will be set up around the demonstration floor room. This year’s showcase features XR applications in STEM education, medical and aviation training, agricultural data visualization, homeland security, training design, and visitor engagement in informal learning settings. Our goal is for attendees to experience how human factors professionals use XR to support human factors-oriented research and to learn about the exciting work being conducted with these emerging technologies.
{"title":"Human Factors Extended Reality Showcase","authors":"Randall Spain, Benjamin Goldberg, Shannon Bailey, Stephanie Fussell, Allison Bayro, Kelly Hale, Aaron Jones, Rachel Regina, Bob Thomas, Kevin Owens, Nathan Lau, Abhraneil Dam, Karen Chen, Luke Sturgeon, Monifa Vaughn-Cooke, Nuela Chidubem Enebechi","doi":"10.1177/21695067231192432","DOIUrl":"https://doi.org/10.1177/21695067231192432","url":null,"abstract":"This alternative format session provides a forum for human factors scholars and practitioners to showcase how state-of-the-art extended reality (XR) applications are being used in academia, defense, and industry to address human factors research. The session will begin with short introductions from each presenter to describe their XR application. Afterward, session attendees will engage with the presenters and their demonstrations, which will be set up around the demonstration floor room. This year’s showcase features XR applications in STEM education, medical and aviation training, agricultural data visualization, homeland security, training design, and visitor engagement in informal learning settings. Our goal is for attendees to experience how human factors professionals use XR to support human factors-oriented research and to learn about the exciting work being conducted with these emerging technologies.","PeriodicalId":74544,"journal":{"name":"Proceedings of the Human Factors and Ergonomics Society ... Annual Meeting. Human Factors and Ergonomics Society. Annual meeting","volume":"48 3","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135113307","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-25DOI: 10.1177/21695067231192871
Kamolnat Tabattanon, Aaron Sun, Bernard J.
Simulated impairment refers to requiring persons without impairments to substitute persons with impairments in endeavors related to training, design creation, or usability testing. However, disability voices and research suggest limited effectiveness, greater distancing, and exclusion. Despite this, the simulation of physical limitations, including that of aging, continue to be used under the assumption that physical tasks and usability ratings do not significantly differ in simulation. In this experiment, age- and sex-matched older adults who use and do not use a manual wheelchair (MWC) are instructed to independently perform a self-paced parallel park using an MWC. The total clearance between obstacles required to perform a collision-free trial was recorded. Thirty-eight volunteers were recruited (MWC-user n=15; Simulated Impairment [SI] n=23). Higher clearance was required by the MWC group, suggesting the use of simulated impairment for motor tasks may result in bias. Open-ended questions revealed self-centered viewpoints, supporting literature that raises inclusion concerns regarding views of an “Other” group. Overall, our results support the direct engagement of target populations during early design to appropriately define user perspectives and needs. Designers should work with the community of people who face limitations rather than substituting their voices with those who may not accurately represent all of their consumer needs.
{"title":"Unintended Biases due to Simulated Impairment within Inclusive Mobility Research and Design","authors":"Kamolnat Tabattanon, Aaron Sun, Bernard J.","doi":"10.1177/21695067231192871","DOIUrl":"https://doi.org/10.1177/21695067231192871","url":null,"abstract":"Simulated impairment refers to requiring persons without impairments to substitute persons with impairments in endeavors related to training, design creation, or usability testing. However, disability voices and research suggest limited effectiveness, greater distancing, and exclusion. Despite this, the simulation of physical limitations, including that of aging, continue to be used under the assumption that physical tasks and usability ratings do not significantly differ in simulation. In this experiment, age- and sex-matched older adults who use and do not use a manual wheelchair (MWC) are instructed to independently perform a self-paced parallel park using an MWC. The total clearance between obstacles required to perform a collision-free trial was recorded. Thirty-eight volunteers were recruited (MWC-user n=15; Simulated Impairment [SI] n=23). Higher clearance was required by the MWC group, suggesting the use of simulated impairment for motor tasks may result in bias. Open-ended questions revealed self-centered viewpoints, supporting literature that raises inclusion concerns regarding views of an “Other” group. Overall, our results support the direct engagement of target populations during early design to appropriately define user perspectives and needs. Designers should work with the community of people who face limitations rather than substituting their voices with those who may not accurately represent all of their consumer needs.","PeriodicalId":74544,"journal":{"name":"Proceedings of the Human Factors and Ergonomics Society ... Annual Meeting. Human Factors and Ergonomics Society. Annual meeting","volume":"25 6","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135113550","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-25DOI: 10.1177/21695067231193689
Noor Jajo, Francesco N. Biondi
The purpose of this study is to investigate how partially automated vehicles affect cognitive load. The study involved an on-road experiment where 29 participants drove a Tesla in both partially automated and manual mode for up to 45 minutes. The researchers objectively measured the participants' cognitive workload using the Detection Response Task (DRT) and subjectively assessed it using NASA Task Load Index (NASA-TLX). The findings revealed that there was no significant difference in cognitive workload between the partially automated mode and manual mode in both objective and subjective measures. Our study expands the literature on the effects of partially automated vehicles on cognitive workload by using DRT and NASA-TLX. Further studies should adopt similar methodology with the addition of physiological and ocular measures.
{"title":"Using Detection Response Task and NASA-TLX to Measure the Difference in Cognitive Workload Between Partially Automated Mode and Manual Mode: An On-Road Study","authors":"Noor Jajo, Francesco N. Biondi","doi":"10.1177/21695067231193689","DOIUrl":"https://doi.org/10.1177/21695067231193689","url":null,"abstract":"The purpose of this study is to investigate how partially automated vehicles affect cognitive load. The study involved an on-road experiment where 29 participants drove a Tesla in both partially automated and manual mode for up to 45 minutes. The researchers objectively measured the participants' cognitive workload using the Detection Response Task (DRT) and subjectively assessed it using NASA Task Load Index (NASA-TLX). The findings revealed that there was no significant difference in cognitive workload between the partially automated mode and manual mode in both objective and subjective measures. Our study expands the literature on the effects of partially automated vehicles on cognitive workload by using DRT and NASA-TLX. Further studies should adopt similar methodology with the addition of physiological and ocular measures.","PeriodicalId":74544,"journal":{"name":"Proceedings of the Human Factors and Ergonomics Society ... Annual Meeting. Human Factors and Ergonomics Society. Annual meeting","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135113718","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-25DOI: 10.1177/21695067231192612
Manhua Wang, Ravi Parikh, Myounghoon Jeon
Ensuring a safe transition between the automation system and human operators is critical in conditionally automated vehicles. During the automation-to-human transition process, hazard avoidance plays an important role after human drivers regain the vehicle control. This study applies the multilevel Hidden Markov Model to understand the hazard avoidance processes in response to static road hazards as continuous processes. The three-state model—Approaching, Negotiating, and Recovering—had the best model fitness, compared to the four-state and five-state models. The trained model reaches an average of 66% accuracy rate on predicting hazard avoidance states on the testing data. The prediction performance reveals the possibility to use the hazard avoidance pattern to recognize driving behaviors. We further propose several improvements at the end to generalize our models into other scenarios, including the potential to model hazard avoidance as a basic driving skill across different levels of automation conditions.
{"title":"Using Multilevel Hidden Markov Models to Understand Driver Hazard Avoidance during the Takeover Process in Conditionally Automated Vehicles","authors":"Manhua Wang, Ravi Parikh, Myounghoon Jeon","doi":"10.1177/21695067231192612","DOIUrl":"https://doi.org/10.1177/21695067231192612","url":null,"abstract":"Ensuring a safe transition between the automation system and human operators is critical in conditionally automated vehicles. During the automation-to-human transition process, hazard avoidance plays an important role after human drivers regain the vehicle control. This study applies the multilevel Hidden Markov Model to understand the hazard avoidance processes in response to static road hazards as continuous processes. The three-state model—Approaching, Negotiating, and Recovering—had the best model fitness, compared to the four-state and five-state models. The trained model reaches an average of 66% accuracy rate on predicting hazard avoidance states on the testing data. The prediction performance reveals the possibility to use the hazard avoidance pattern to recognize driving behaviors. We further propose several improvements at the end to generalize our models into other scenarios, including the potential to model hazard avoidance as a basic driving skill across different levels of automation conditions.","PeriodicalId":74544,"journal":{"name":"Proceedings of the Human Factors and Ergonomics Society ... Annual Meeting. Human Factors and Ergonomics Society. Annual meeting","volume":"32 5","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135170428","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：