請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/57028
標題: | 自動駕駛車輛之車道維持系統安全性量化分析 Quantitative analysis of safety of the lane keeping system of autonomous driving vehicles |
作者: | You-Peng Jhang 張友鵬 |
指導教授: | 李綱(Kang Li) |
關鍵字: | 自動駕駛,安全評估,安全分析, Autonomous vehicle,Safety Evaluation,Safety Analysis, |
出版年 : | 2020 |
學位: | 碩士 |
摘要: | 本研究基於最劣情況下(Worst-Case Scenario)之錯誤注入(Fault Injection),對自駕駛系統進行安全性量化分析,利用自動駕駛實車運行數據,探討封閉場域下,自駕載具在無外部事件影響下,其自體發生的最劣錯誤是否影響載具運行時最基本的車道維持功能,並根據其位置與姿態、速度與加速度進行安全性量化,進而提出三個安全指標來提供描述自駕系統之安全程度,並可依此提供自駕系統改善的方向,包含位置安全餘裕、速度安全餘裕與加速度安全餘裕,而這些安全指標未來亦可作為安全例證論證時的證據之一。 Based on the fault injection under the worst-case scenario, a quantitative analysis of the safety of the autonomous driving system is proposed. Using the real-time driving data of the autonomous driving vehicle to analysis whether the occurrence of the worst fault of the autonomous driving system will affect the operation of the most important function of autonomous driving vehicle which is so called lane keeping in the closed field without the influence of external events. Under this scenario, evaluation procedure of the safety is designed according to its position, posture, speed and acceleration which is the most essential properties of physical system. In the end, three safety indicators is provided to give a description of the safety level of the autonomous driving system and it can provide the direction of improvement of the autonomous driving system, including position safety margin, speed safety margin and acceleration safety margin. These safety indicators can also be used as one of the evidences for the argument of safety case in the future. |
URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/57028 |
DOI: | 10.6342/NTU202001758 |
全文授權: | 有償授權 |
顯示於系所單位: | 機械工程學系 |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
U0001-2307202000320200.pdf 目前未授權公開取用 | 4.72 MB | Adobe PDF |
系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。