車載語音人機介面互動之風險評估

Risk Assessment for In-Vehicle Speech Interaction Systems: Talking Car System

陳虹君(Hung-Chun Chen)

指導教授 ： 黃雪玲

國立清華大學/工學院/工業工程與工程管理學系/碩士(2014年)

摘要

隨著科技的進步和資訊流動加快，駕駛對車用資通訊功能的需求也大幅提高，龐大的市場因此讓汽車電子被視為帶動電子產業商機的第四C，進而加速相關技術的發展。很多駕駛在開車的時候，往往會透過車用資訊系統來取得路況、音樂、廣播等資訊，然在取得這些相關資訊的同時，駕駛可能因為人機介面的設計不妥等因素造成使用車用資通訊系統時發生失誤而影響其行車安全。本研究以一款由工研院開發之全盲式中文車用語音系統Talking Car為研究對象，使用風險評估相關技術來找出此語音介面設計上的潛在問題。
在本研究中，首先進行一次駕駛模擬實驗，共計有16位受試者，每位受試者須一邊開車一邊操作Talking Car來完成指定的任務，並在實驗後填寫主觀問卷及接受訪談。實驗中，同時設有錄影機記錄受試者的所有行為，包括開車和操作語音系統。透過影片觀察以及訪談可以蒐集、分析駕駛模擬實驗中發生的所有的失效模式、錯誤操作以及其造成原因與產生效應，再進行納入人為操作考量的修正型失效模式與效應分析，並針對每個失效模式與操作動作計算其風險優先數，由高到低排列出其需改善的順序。
研究結果顯示現行Talking Car語音系統確實存在需要修改之處，分別是「系統引導語句」、「方向盤上的控制按鍵」及「部分系統介面」。而此修正後的失效模式與效應分析也被認為是可用來評估操作語音系統作業之系統化工具。本研究亦針對Talking Car語音系統不足之處提出改善建議，預期透過介面改善，降低Talking Car失誤發生之頻率並提升其操作績效與安全性。

關鍵字

風險評估 ； 車用語音系統 ； 失效模式與效應分析 ； 駕駛安全

並列摘要

The purpose of this study was to identify the potential problems of an in-vehicle speech interaction system called Talking Car by using risk assessment techniques.
In this study, 16 participants took part in a driving simulation experiment and were asked to complete the assigned tasks by controlling the Talking Car system while driving. The participants also had to answer a subjective questionnaire and sit for an interview after the experiment. By conducting the experiment, the failure modes, incorrect operations, failure causes, and effects were observed and collected. Then, a modified failure mode and effects analysis (FMEA) method which took human operation into account was used to determine and analyze these failure modes and incorrect operations. After completing the analysis, the risk priority number (RPN) was computed to detect which failure had a significant effect on driving safety. RPNs and subjective questionnaires indicated that “System Dialogue with Elicitation,” “Control Button on Steering Wheel,” and “Partial System Interface” urgently need improvement.
This study revealed significant deficiencies in the current Talking Car system. Additionally, the results also revealed that the modified FMEA method was feasible and can be proposed as a systematic technique to assess the task of controlling in-vehicle speech interaction systems while driving. Furthermore, recommendations to address the design problems of the Talking Car system are presented to facilitate driving safety.

並列關鍵字

Risk Assessment ； In-Vehicle Speech Interaction System ； FMEA ； Driving Safety

參考文獻

Bowles, J. B., & Peláez, C. E. (1995). Fuzzy logic prioritization of failures in a system failure mode, effects and criticality analysis. Reliability Engineering & System Safety, 50(2), pp. 203-213.

Braglia, M., Frosolini, M., & Montanari, R. (2003). Fuzzy TOPSIS approach for failure mode, effects and criticality analysis. Quality and Reliability Engineering International, 19(5), pp. 425-443.

Chang, K. H. (2009). Evaluate the orderings of risk for failure problems using a more general RPN methodology. Microelectronics Reliability, 49(12), pp. 1586-1596.

978-0-387-73819-2_11, DOI: 10.1007/978-0-387-73819-2_11

Chiozza, M. L., & Ponzetti, C. (2009). FMEA: a model for reducing medical errors. Clinica Chimica Acta, 404(1), pp. 75-78.