摘要: |
The objectives of the research described in this report were twofold: evaluate whether crash imminent braking (CIB) false positives can be consistently observed on the test track using maneuvers representative of real-world driving scenarios, and to assess the practicality of accurately and repeatably performing these maneuvers. To satisfy these objectives, five later model light vehicles were evaluated using eight test maneuvers performed at two test speeds. Six of the scenarios were identified during CIB research conducted by the Crash Avoidance Metrics Partnership (CAMP) Crash-Imminent Braking Consortium. The remaining scenarios originated from an ISO 22839 draft that provides recommendations for test methods and requirements necessary for validating forward collision warning (FCW) and CIB system performance. Generally speaking, CIB false positives were not consistently observed on the test track. Of the eight maneuvers used in this study, only Object in Roadway Steel Trench Plate tests produced CIB false positives, and only for one vehicle (although they were present during each test trial for this vehicle). Whether the corresponding reduction in vehicle speed would create a real-world safety problem is not known. FCW alerts always preceded the CIB activations for this vehicle. Three maneuvers produced FCW alerts but no CIB false activations: the Stationary Vehicle at Curve Entrance, Stationary Roadside Vehicles, and Decelerating Vehicle in an Adjacent Lane (Curve) scenarios. Alert frequency varied considerably by vehicle and maneuver; some vehicles consistently issued alerts while others did not respond at all. Four maneuvers failed to elicit either FCW or CIB activation: The Decelerating Vehicle in an Adjacent Lane (Straight), Driving Under an Overhead Bridge, Objects in Roadway Botts Dots, and Stationary Vehicle at Curve Exit scenarios. With regards to practicality, the only maneuver found to produce CIB false positives, the Object in Roadway Steel Trench Pla |