On Sat, May 30, 2009 at 12:02 PM, Vivian Meazza wrote:
I don’t see much of a problem with a
suitable jumper. Making it a ballistic object with drag and mass would be easy.
But a more realistic FDM … Hmmm
An accurate FDM would be immensely complex considering all the possible poses a human can achieve. But perhaps something simplistic could be worked up using the arms and legs as control surfaces. My focus right now is not so much getting accurate free fall dynamics, but to get a nice jumper model and then just hack up some sort of dynamics with approximately the right lift/drag ratio for someone is a stable free fall pose.
The goal would be to get approximately the right fall rates and timings so that there is training value in solving problems and overcoming various combinations of faults with in a realistic time frame. It's still only a very partial simulation but hopefully a step better than just sitting around in a circle talking through various scenarios.
The next step would be to have a canopy that could be configured to have various problems opening up and be able to draw that somehow from the perspective of the sky diver, and perhaps have some appropriate dyanmics for partially tangled or partially inflated chutes? Obviously there's endless variabiltiy and high fidelity in all respects would be crazy to try to achieve, but it would be interesting to take a few small steps forward and see how far we can get.
Curtis Olson: http://baron.flightgear.org/~curt/