Continuous collision detection for rotated bodies
Started by Volgogradetzzz, Apr 15 2011 05:18 AM
9 replies to this topic
Ad:
#4 Banned - Reputation: 356
Posted 15 April 2011 - 07:06 AM
Thanks, owl, but the question was not aboute SAT. For exaple I have 2 static bodies, i.e. their velocities is 0. But they can rotate and they can collide at some time. So how can I get this time?
http://www.gamedev.net/topic/586690-collision-detection-for-rotating-objects/
the first reply pretty much addresses it.
I like the Walrus best.
#5 Members - Reputation: 319
Posted 16 April 2011 - 01:21 AM
Thanks. Sadly - but see the only way - decrease time steps.
Also I can do the following - given two line segments I can predict their TOI in current frame even if they moving with rotation. But math that I use - is toooooooooooo complicated - there're tons of sin/cos, so for real time sim with many objects this method is not suitable.
Also I can do the following - given two line segments I can predict their TOI in current frame even if they moving with rotation. But math that I use - is toooooooooooo complicated - there're tons of sin/cos, so for real time sim with many objects this method is not suitable.
#7 Members - Reputation: 381
Posted 16 April 2011 - 04:29 AM
I've read some moderator talking about radial swept-volumes (yet denoted as quite complicated). You might want to ask about that.. and post me back lol
You can use conservative advancement, or cheat and use closet points combined with speculative contacts - about half way through the article I talk about rotation
#9 Members - Reputation: 381
Posted 20 April 2011 - 07:55 AM
Yeah, I've read your article, but I'm true dummy and I didn't understand a half. I hope I'll dig it someday. Thanks.
I've just published a new one on collision detection which covers continuous rotating bodies:
http://www.wildbunny.co.uk/blog/2011/04/20/collision-detection-for-dummies/
Hope its useful!
Cheers, Paul.






