{"title":"A Literature Review of Simulation Fidelity for Autonomous-Vehicle\n Research and Development","authors":"Christopher Johnson, Elan Graupe, Maxfield Kassel","doi":"10.4271/01-16-03-0021","DOIUrl":null,"url":null,"abstract":"This article explores the value of simulation for autonomous-vehicle research and\n development. There is ample research that details the effectiveness of\n simulation for training humans to fly and drive. Unfortunately, the same is not\n true for simulations used to train and test artificial intelligence (AI) that\n enables autonomous vehicles to fly and drive without humans. Research has shown\n that simulation “fidelity” is the most influential factor affecting training\n yield, but psychological fidelity is a widely accepted\n definition that does not apply to AI because it describes how well simulations\n engage various cognitive functions of human operators. Therefore, this\n investigation reviewed the literature that was published between January 2010\n and May 2022 on the topic of simulation fidelity to understand\n how researchers are defining and measuring simulation fidelity\n as applied to training AI. The results reported herein illustrate that\n researchers are generally using agreed-upon terms such as physical\n fidelity, but there is an emerging definition of functional\n fidelity that is being adopted to replace the concept of\n psychological fidelity for training AI instead of\n humans.","PeriodicalId":44558,"journal":{"name":"SAE International Journal of Aerospace","volume":" ","pages":""},"PeriodicalIF":0.3000,"publicationDate":"2023-05-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"SAE International Journal of Aerospace","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.4271/01-16-03-0021","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"ENGINEERING, AEROSPACE","Score":null,"Total":0}
引用次数: 0
Abstract
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.