Is there a Vehicle v Pedestrian simulation program?

Subjects related to the reconstruction and simulation of Occupants in vehicles and Pedestrians struck by vehicles, ATB & ATB clones #ATB #MADYMO #MultiBody
brian
Posts: 499
Joined: Tue Jul 14, 2009 10:52 am

Is there a Vehicle v Pedestrian simulation program?

Post by brian »

Q? Is there a program that can simulate a vehicle v ped collision, specifically the dynamics of the ped.

A: PC Crash (and now also Virtual Crash) include in their software packages a multi-body system.
The multi-body system are basically a stripped down/simplified version of the Articulated Total Body (ATB) Program.
  • We recommend using extreme caution when using an occupant simulation model to simulate anything other than gross occupant kinematics. Please see our papers on limitations of occupant simulation models. Some of the modeling limitations of the ATB program (or any occupant/pedestrian simulation program like #MADYMO, #pc-crash, #VirtualCrash) make it extremely sensitive to the many subjective inputs required for the programs. ATB and MADYMO are great research programs but extreme caution and care MUST be exercised if they are used in individual case reconstructions.
For example
  • the inputs for restitution.
    • We have seen instances where the ATB was used to simulate a pedestrian impact and the pedestrian achieved restitution greater than 1! (it launched from the impact at a speed more than twice the impact velocity of the vehicle).
      • Be sure to do a careful analysis of the pedestian crash being analyzed before utilizing simulation (as you should do with ANY simulation program).
AND
  • Effective Restitution, Launch Angle and Friction
    • Be sure to look at the inputs and resulting effective restitution, launch angle and friction of any pedestrian (or occupant) reconstruction especially if it includes a simulation:
      • Restitution: Humans are like sangbags when struck by a vehicle, they do NOT have large amounts of restitution so the speed of separation from the vehicle should be at or below the impact speed.
      • Launch Angle: Also look at the inputs and resulting effective launch angle since it can affect the overall trajectory (air time vs ground sliding time)
      • Coefficient of Friction: And look at the inputs and resulting effective friction between the pedestrian and the vehicle and the pedestrian and the ground.
    All can dramatically change the results of any 'pedestrian simulation'
The PC Crash folks have published an SAE paper: Here is the ABSTRACT from SAE paper 1999-01-0445:
  • In 3D simulation of vehicle accidents the vehicles are typically treated as one rigid body. This simplification can be used for vehicle to vehicle collisions as well as in impacts with solid objects. To get a realistic movement of the pedestrian in the impact simulation, the pedestrian has to be modeled as a system of rigid bodies interconnected by joints.
    This paper describes in detail the extension of the discrete kinetic time forward simulation program PC-Crash to deal with multi-body-systems
    As one application of this model pedestrians have been modeled. During the generation of this pedestrian model special interest has been focused on a realistic movement of the pedestrian and a short calculation time. The pedestrian model shows to be a good tool to analyze the movement of the pedestrian regarding its post impact kinematics and contact locations on the vehicles and the ground. In addition specific contact locations on the vehicle can be correlated with the movement of the pedestrian and damage locations on the vehicle.
    For the validation of this model several crash tests have been performed and will be done. The second part of this paper describes the results of some crash tests and a comparison of the simulation results. A good agreement between crash-tests and simulation results could be found. For these crash-tests different typical shapes of the front hood of modern cars have been used to study the influence on the movement of the pedestrian during and after the impact.
    The model also involves the capability to deal with different kinds of pedestrians (size and mass) under various initial conditions (standing, walking, running).
COMMENT
  • The best it can do is make a pretty illustration of what you would get from simplified occupant trajectory equations.
    Sure when you can tweak the inputs to match a full-scale test and it looks great...however... how do you determine the actual inputs required for a real world crash with so many unknowns?

This topic has 2 more posts with additional information

To Read more, Please login and/or register. 2024 NOTE: Soon ALL Technical Sections will be ONLY for registered users. Optionally you can email us forum@mchenrysoftware.com your Name, Company, Location, a Username, and a Password (which you can change) and we will register you and send you a confirmation email.


Register Login