Driver Acceptance of Connected, Automation-Assisted Cruise Control: Experiment 1

10466813372018Philips, Brian H.PDFTech ReportFHWA-HRT-18-041HRDS-30/07-18(WEB)EDTFH61-13-D-00024Autonomous intelligent cruise controlDriver performanceAutomobiles--Automatic controlMotor vehicle driving--United States--AutomationUnited StatesUnited States. Federal Highway Administration. Office of Safety Research and DevelopmentInman, Vaughan W.Jackson, StevenChou, PeterLeidosUnited States. Federal Highway Administration. Office of Safety Research and DevelopmentTurner-Fairbank Highway Research CenterFederal Highway AdministrationThe results and conclusions of an initial experiment that examined human-factors issues in the use of adaptive cruise control (ACC) and a hypothetical cooperative ACC (CACC) in which ACC supplemented with vehicle-to-vehicle (V2V) communications to extend ACC capabilities. In this driving simulator\u2013based experiment, the CACC system was modeled to accelerate and decelerate less aggressively when the ACC radar lost track of the vehicle ahead on a curve. In addition to testing driver responses to two cruise-control systems, three cruise-control displays were tested. One display showed only whether cruise control was turned on or off. A second display showed not only whether the display was on or off, but also whether the system was tracking another vehicle. A third display incorporated a video of the road ahead with the second display and indicated which vehicle was being tracked. Drivers of the CACC-equipped vehicle rated their trust in cruise control higher than drivers of the ACC-equipped vehicle did. Drivers looked at the second and third display types about 2 percent of the time. However, no glance longer than 0.73 s to any cruise-control display was observed. It was concluded that supplementing ACC with V2V communications may increase system use relative to ACC and provide additional safety benefits. Providing ACC tracking information similar to that provided in this study does not appear to distract drivers; however, additional testing in more complex driving environments is recommended. This research should be of interest to researchers and developers of Level 1\u2013automated systems.74