Perez 1 TEST AND EVALUATION OF THE COOPERATIVE INTERSECTION COLLISION AVOIDANCE SYSTEM FOR VIOLATIONS (CICAS-V) DRIVER-VEHICLE INTERFACE

The Cooperative Intersection Collision Avoidance System for Violations (CICAS-V) project was conducted to develop and field-test a comprehensive system to assist drivers in reducing the number and severity of crashes at intersections due to violations at stop-sign and signal-controlled intersections. One essential component of such a system is the DriverVehicle Interface (DVI) to warn a driver of an impending violation. A series of test-track studies was conducted to support the selection of a DVI for subsequent on-road tests of the CICAS-V. In these tests, 18 naive drivers per interface were placed in a surprise intersection violation scenario and provided with a precisely timed warning presented through a variety of DVIs. Driver braking profiles and vehicle stop locations were collected and analyzed, with particular emphasis on behaviors that resulted in avoiding entering the intersection DVIs included combinations of visual, auditory, and haptic (brake pulse) warnings. Results from the tests showed that drivers exposed to a brake pulse tended to stop more often and with lower decelerations than drivers that were not exposed to the brake pulse. The effectiveness of the brake pulse warning, however, was partly moderated by the type of auditory warning that accompanied the brake pulse warning. A baseline trial was conducted to determine the benefit of the DVI over a non-warning condition. Overall, results supported the recommendation of a DVI containing the simultaneous presentation of a flashing visual (red stoplight/stop sign icon), a ‘Stop Light’ speech warning, and a single brake pulse. The best-performing DVI resulted in an 88% improvement over the baseline condition. Project participants included offices of the United States Department of Transportation, Daimler, Ford, General Motors, Honda, Toyota, and the Virginia Tech Transportation Institute. INTRODUCTION The Cooperative Intersection Collision Avoidance System for Violations (CICAS-V) project was conducted to develop and field-test a comprehensive system to assist drivers in reducing the number and severity of crashes at intersections due to violations of traffic control devices (TCD). These crashes account for almost 400,000 injuries and fatalities in the United States every year (National Highway Traffic Safety Administration, 2008). The approach selected to reduce these crashes is to present a timely and salient in-vehicle warning to those drivers predicted to violate a TCD. The warning is intended to elicit a behavior from the driver to avoid a potential violation. Supporting the warning are several subsystems that exchange, process, and present the required information from both the vehicle and the intersection. The Driver-Vehicle Interface (DVI), which is the means through which the warning information is presented to the potential violator, is one of these CICAS-V subsystems. The importance of this particular subsystem is based on its function: prompting the driver to take the appropriate violation avoidance maneuver. For this reason, a series of Human Factors (HF) test track studies were executed during the CICAS-V project to determine the DVI that would be integrated into the CICAS-V system for further on-road testing. To this end, experimental scenarios were developed to attain a set of test conditions that simulated a “representative” signal violation environment. Naive drivers were exposed to these scenarios while being aided by one of several DVI alternatives. Based on knowledge gaps remaining after past research efforts, these test scenarios were designed to address the following research questions: • Within the auditory modality, how does the effectiveness of speech warnings compare to non-speech warnings?