"Dogleg" Trajectories Which Can Confuse Trajectory/Momentum Solution Procedures

General Crash Reconstruction Tecnhiques
Crash tests, Pole Impacts
Post Reply
MSI
Site Admin
Posts: 2005
Joined: Thu Jun 18, 2009 12:37 pm

"Dogleg" Trajectories Which Can Confuse Trajectory/Momentum Solution Procedures

Post by MSI »

August 12, 2022: A post today on the NAPARS page of Facebook included a discussion of situation where there is a rollout of the vehicle at the end of rotation. It brought back memories as follows:

A long long time ago (like Don McLean American Pie song)
    • One of my first projects in 1979-1980 was a NHTSA research project working on improvements to the trajectory option of the CRASH3 program. The final report was published in 1981: As part of that research we came up with many option for handling many different types of trajectories. The main issue was that in most crashes the 'separation angle' assumed by simplified momentum solution procedures is not directly from the point of impact to the point of rest.
      Our research looked at a number of different scenarios and came up with several refinements, however...
      • we concluded that due to residual linear velocity at the end of rotation and the possibility of a 'dogleg' type runout that:
        • "Therefore, a realistic appraisal of the achieved results led to the recommendation that future developments of CRASH should focus attention on the existing step-by-step, time ­history forn of trajectory analysis, in which fundamental physical laws are applied directly. "
    • in 1997, we added and improved:
      • 1997 "CRASH-97 - Refinement of the Trajectory Solution Procedure", SAE 970949
        We improved the scatter of the CRASH3 Momentum results (see below) However we never released a commercial version since working on many other projects and not time.
        It also had some sensitivities and we believed that the automatic iteration/optimization of SMAC would be the best solution to test and refine the scatter in simplified momentum results.
        Crash97 results.jpg
        Crash97 results.jpg (77.7 KiB) Viewed 128 times
    • IN 2003 in response to the major advances in PC computers we worked on the automatic iteration/optimization of SMAC:
      • We developed an automatic iteration/optimization routine for the SMAC program:
        • 2003 SMAC2003 - The Automatic Iteration of SMAC SAE 2003-01-0486
          Which used the CRASH3 results form the trajectory collision and iterated/optimized to the following correlation with the full scale test result and measurements for damage, trajectory, impact speed and impact speed change
          • and note we verified the veracity and accuracy of the RICSAC full scale test result and deltaV measurements in our 1997 SAE paper RICSAC-97 :
          Smac2003 Results.png
          Smac2003 Results.png (58.84 KiB) Viewed 128 times
        • which included in the discussion:
          • "The collision and trajectories of vehicles can be highly non-linear events. Minor variations in starting conditions (i.e., speed, impact location) can produce major changes in the resulting rest positions (X, Y, PSI) and discontinuities in the calculated error evaluation terms. For example, during decelerations of the linear and angular velocities, as a vehicle rotates while it travels from separation to rest, the vehicle may “shoot off” tangentially in what has been described as a “dog leg” type of trajectory at any time that the velocity vector aligns with the longitudinal axis."
Doglegs type trajectories can be a problem for simplified momentum solution procedures.
  • Obviously if you KNOW the answer you want or need it can solve the issue, however if you don't, HOW to you objectively vary the separation angles of the vehicle: what evidence do you use to justify the change?
  • That is why we recommend using a time history vehicle simulation program like msmac3D for such cases to allow an objective analysis of the damage AND trajectory evidence to evaluate what combination of speed, angles and offset project the proper damage areas and trajectory match.

i will post up some dogleg type crash trajectories, etc
i added this post because we really hadn't addressed 'dogleg' directly in any posts yet, now we have!
Quick LINKS: ->HOT TOPICS<- ->msmac3D LEASE pricing!<- ->McHenryForum Index<-
Question? Comment? Please email forum@mchenrysoftware.com
(c)McHenry Software ALL Rights Reserved.
MSI
Site Admin
Posts: 2005
Joined: Thu Jun 18, 2009 12:37 pm

Re: "Dogleg" Trajectories Which Can Confuse Trajectory/Momentum Solution Procedures

Post by MSI »

Adding some more background on the "dogleg" trajectory issue:
  • In the early development of the CRASH program, first as a preprocessor for SMAC, and then as a major tool to be used by NHTSA for statistical research (the NASS research), there was investigations of possible solutions to the "dogleg" type trajectories for the simplified single step momentum solution procedure of CRASH:
    • The first way it was addressed by Ray in 1976: was with the curved path option, end of rotation designation, etc:
      Crash Traj refinements.jpg
      Crash Traj refinements.jpg (60.91 KiB) Viewed 143 times
      • The second item explored was development of a Trajectory Simulation Model:
        • It was essentially an early iteration program of the SMAC program.
        • It was thought that simple trajectory simulations took less computer time, were quick
          • the collision interaction of SMAC uses the most computer time and resources so this would save computer costs
        • It would also importantly provide a test to see what combination of angular and linear speed would produce a match of the measured evidence
          Other research looked into this approach :
          • Jones, Ian. S., "Automated Reconstruction", SAE paper 750894 Accident
          • Moffatt, C.A., Byrd, J. Jr. :A Computer Model to Operate the SMAC Program Automatically, Highway Collision Reconstruction, Winter ASME Meeting 1980
      However due to
      • the high cost of multiple SMAC trajectory simulation runs on mainframe computers in the 1970s
      NHTSA budget considerations created the idea for our 1979-1981 project we worked on to refine CRASH3 trajectory option to try to find an more accurate "single step" for the trajectory aka simplified momentum solution procedure of CRASH.
More information will be posted soon...have a few fires to put out....
Quick LINKS: ->HOT TOPICS<- ->msmac3D LEASE pricing!<- ->McHenryForum Index<-
Question? Comment? Please email forum@mchenrysoftware.com
(c)McHenry Software ALL Rights Reserved.
Post Reply