· Chris Thompson · Hamilton Turner · Brian Dougherty · Douglas C. Schmidt
Download 0.67 Mb. View original pdf
|
| Table 1 Analysis of wreck watch false negatives on real crash data Wreck description Speed (mph) Acceleration (Gs) False negative Car front/side impact from oncoming truck 40–75 No Stopped truck rear-ended 65 80–160 No Car striking stationary object 100–120 No Car knocked in culvert 40–60 No SUV rollover 30–70 No Truck bounces off guardrail 25–45 No Car side-impact from truck pulling out 70–200 No Car rear-ended by another car 60–90 No Stopped car rear-ended by truck and sandwiched No between truck and school bus SUV roll onto side 30 No Armored car rear-ended by another armored car 20–25 >500 No Van rear-ended 15 50–100 No Small truck collision with car in reverse 20–40 No Car rear-ended at an angle No a standstill to over 330 mph in about 4.5 s. This type of acceleration yields roughly 3.34 Gs. A formula one race car driving through a turn with aft radius at mph experience approximately 4 Gs. The McLaren F1 sports car’s minimum 60 to 0 mph breaking time is 2.8 s, which yields roughly G. Clearly, for normal drivers, it is highly unlikely that 4 Gs, which is the accident detection threshold for WreckWatch, would be experienced by the driver’s phone without an accident. Since we did not test all possible phone positions in the vehicle, it certainly possible that there maybe scenarios where an unusually high deceleration or fast turn could hurl the phone into the dashboard or other hard surface fast enough to produce a false positive. However, in these situations, WreckWatch’s builtin confirmation screen would allow the driver to cancel the accident notification before it was reported. There is still the potential that the driver might forget to cancel the notification before the it was reported. Download 0.67 Mb. Share with your friends:
|