Browse Publications Technical Papers 2000-01-1315

Partially-Braked Impact and Trajectory Benchmarks, and Their Application to CRASH3 and CRASHEX 2000-01-1315

This paper applies a known benchmark set for the performance of vehicular collision algorithms to two impact and trajectory reconstruction algorithms, for the case of partial braking. One algorithm assumes no vehicle movement during the collision (instantaneous impact) and no structural recovery (plastic crush; no restitution); the other revokes these conventional simplifications. The paper presents statistical comparison of each algorithm with the benchmark set, hence with one another.
A reference family of reasonably well simulated 2-car impacts in four classes, previously offered by Woolley and Kinney for the 100% braked case only, is here extended to the case of partial (40%) braking. As with the original (1994) set, authors of accident reconstruction programs are invited to demonstrate the performance of their programs in terms of deviation from this available, if arbitrary, benchmark set of reference cases.
For two reconstruction programs, CRASH and CRASHEX, this paper presents and statistically treats deviations from those 33 partially braked benchmark cases.
Departures from benchmark are similar to, although larger than, those previously reported for the fully braked Woolley and Kinney cases, continuing to show much less departure for CRASHEX than for CRASH3. Including all front and side impact cases and all but one offset collision, the larger extreme of 95% of CRASHEX departures was 2.5 mph. For the same 24 cases, adoption of the CRASH3 assumptions increased those departures by a factor of 3, and raised the increase in probable ultimate error, above equally likely 4 mph errors of observation, from 0.7 to 4.6 mph-increasing that critical contribution by a factor of 6.


Subscribers can view annotate, and download all of SAE's content. Learn More »


Members save up to 18% off list price.
Login to see discount.
Special Offer: Download multiple Technical Papers each year? TechSelect is a cost-effective subscription option to select and download 12-100 full-text Technical Papers per year. Find more information here.