Koen van de Merwe, S. Mallam, Ø. Engelhardtsen, Salman Nazir
{"title":"Operationalising Automation Transparency for Maritime Collision Avoidance","authors":"Koen van de Merwe, S. Mallam, Ø. Engelhardtsen, Salman Nazir","doi":"10.12716/1001.17.02.09","DOIUrl":null,"url":null,"abstract":": Automation transparency is a means to provide understandability and predictability of autonomous systems by disclosing what the system is currently doing, why it is doing it, and what it will do next. To support human supervision of autonomous collision avoidance systems, insight into the system’s internal reasoning is an important prerequisite. However, there is limited knowledge regarding transparency in this domain and its relationship to human supervisory performance. Therefore, this paper aims to investigate how an information processing model and a cognitive task analysis could be used to drive the development of transparency concepts. Also, realistic traffic situations, reflecting the variation in collision type and context that can occur in real ‐ life, were developed to empirically evaluate these concepts. Together, these activities provide the groundwork for exploring the relation between transparency and human performance variables in the autonomous maritime context","PeriodicalId":46009,"journal":{"name":"TransNav-International Journal on Marine Navigation and Safety of Sea Transportation","volume":null,"pages":null},"PeriodicalIF":0.7000,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"TransNav-International Journal on Marine Navigation and Safety of Sea Transportation","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.12716/1001.17.02.09","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"TRANSPORTATION SCIENCE & TECHNOLOGY","Score":null,"Total":0}
引用次数: 1
Abstract
: Automation transparency is a means to provide understandability and predictability of autonomous systems by disclosing what the system is currently doing, why it is doing it, and what it will do next. To support human supervision of autonomous collision avoidance systems, insight into the system’s internal reasoning is an important prerequisite. However, there is limited knowledge regarding transparency in this domain and its relationship to human supervisory performance. Therefore, this paper aims to investigate how an information processing model and a cognitive task analysis could be used to drive the development of transparency concepts. Also, realistic traffic situations, reflecting the variation in collision type and context that can occur in real ‐ life, were developed to empirically evaluate these concepts. Together, these activities provide the groundwork for exploring the relation between transparency and human performance variables in the autonomous maritime context