How to apply force to individual rigid bodies based on impact?

[karsnen]

Hello,

SCENARIO :
I am a game developer and I am playing around with rigidbodies. I have a piece of wall which might get impact by an external force. The wall has in turn several other broken pieces (hidden for the moment) put up together to resemble this huge wall. Upon impact I switch one whole piece of wall for the individual pieces and would like to simulate that the wall is broken and subsequently forces are applied to the individual pieces. Thus creating a realistic visual appeal.

PICTURE ORIENTATION :
In the picture, the (a) is a whole piece which would get impact and then upon impact it would switch to (b) which in turn have to be applied the respective forces.

RESOURCES :
Now I have information from the force on impact and also the information from rigidbody of the whole piece at any point of time, till the switch.

REQUIREMENT :
What I am not sure is how should apply forces to the individual pieces to create the sort of realistic simulation? How should I approach it?

I hope I was clear with my scenario, resources I have and my requirement. If I am not clear at any point, please I would like to detail it.

Thank you for your time and highly appreciate it.

Karsnen.


INFORMATION I HAVE :

The External Force :

• Source of the force
• Position of the source of the force

The Whole Piece of Rigidbodiy:

• mass
• velocity
• angular Drag
• angular Velocity
• center Of Mass
• inertia Tensor
• inertia Tensor Rotation
• position - The position of the rigidbody.
• rotation

 

[elegysix]

Well, a lot of it will depend on how exactly the impact happens. As a good general rule, I would keep conservation of momentum in mind, and don't forget to immediately start accelerating your pieces downward due to gravity.

Depending on where the impact occurs in relation to where the breaks in the wall are, you should be able to estimate the angular momentum of the pieces as well.
This will draw heavily on your intuition if you don't want to get swamped in the math.
But like I said, if the object has momentum mV before the collision, then the sum of all momenta immediately after the crash must be mV.

I would not worry about making the physics perfect, as I think it is likely that the locations and designs of the breaks will not correspond perfectly to reality either.

Hope this helps

 

[CWatters]

Rather than write one from scratch .. Is there a an open source physics engine you could use?

I don't know but perhaps have a look at some of the physics simulation programs such as..

http://physion.net/
http://www.algodoo.com/

They seem to do a reasonable job of simulating blocks being knocked over.