In-Vehicle Voice Control Interface Performance Evaluation

The purpose of the National Highway Traffic Safety Administration (NHTSA) project titled, In-Vehicle Voice Control Interface Performance Evaluation was to conduct empirical research about the use of voice control systems (VCS) by drivers and potential measures that could be used for evaluating possible distraction from using these systems while driving. An on-road, contextual interview study was conducted in Rockville, Maryland, and Seattle, Washington, to identify drivers’ existing patterns of use and interaction errors encountered with VCS while driving. Differences were observed between those who used original equipment VCS and those who used portable smart devices that were paired to the vehicle. In total, the research team noted 22 themes that characterized participants’ interactions with VCS. Most notably, drivers often had trouble using their VCS but did not necessarily blame the system for the errors or the lengthy system interactions that they experienced. Interactions frequently included several types of errors including speech recognition errors. These results suggest that an evaluation protocol based solely on error free trials would not be representative of many VCS interactions commonly experienced by users while driving. Two other studies were conducted in controlled laboratory environments. Participants interacted with a “Wizard of Oz” voice control system while they performed a surrogate driving task with a driving simulator or with a computer-based collision detection task (CDT). Cognitive load was measured by performance on the ISO Tactile Detection Response Task (TDRT). Eye glance measures, based on Visual-Manual NHTSA Driver Distraction Guidelines for In-Vehicle Electronic Devices were also used. Results indicated that both TDRT performance and eye glance measures may be appropriate for evaluation of VCS and that the CDT protocol yielded similar results to the driving simulator protocol. The findings of these studies will inform development of Phase 3 of NHTSA’s Driver Distraction Guidelines.

• Record URL:
  http://www.nhtsa.gov/DOT/NHTSA/NVS/Crash Avoidance/Technical Publications/2016/8122314_In_VehVoiceCntrlItrfacePerformEval.pdf
• 
• Corporate Authors:

  Westat, Incorporated

  1650 Research Boulevard
  Rockville, MD  United States  20850

  National Highway Traffic Safety Administration

  1200 New Jersey Avenue, SE
  Washington, DC  United States  20590
• Authors:
  • Jenness, James W
  • Boyle, Linda Ng
  • Lee, John D
  • Chang, Chun-Cheng
  • Venkatraman, Vindhya
  • Gibson, Madeleine
  • Riegler, Kaitlin E
  • Kellman, Daniel
• Publication Date: 2016-10

Language

• English

Media Info

• Media Type: Digital/other
• Edition: Final Report
• Features: Appendices; Figures; Maps; Photos; References; Tables;
• Pagination: 112p

Subject/Index Terms

• TRT Terms: Distraction; Driver performance; Driver vehicle interfaces; Drivers; Errors; Speech recognition
• Uncontrolled Terms: Cognitive load; Glance behavior; In-vehicle technologies; Voice control systems
• Subject Areas: Highways; Safety and Human Factors; Vehicles and Equipment;

Filing Info

• Accession Number: 01616256
• Record Type: Publication
• Report/Paper Numbers: DOT HS 812 314
• Contract Numbers: DTNH22-11-D-00237 /0007
• Files: HSL, TRIS, ATRI, USDOT
• Created Date: Nov 15 2016 11:34AM