Flight

I've just recently added the calculation of the normal vectors to groups of cells. What this means is that each surface cell is not only characterised by a position and velocity but also by an orientation. The first thing this allowed me to do was to create better water collision physics and effects. But it has also allowed me to elaborate the basic drag model into a model of lift and drag which will allow flight. Note that the plane in the promotional video was recorded before I had this tech in place and is exhibiting purely ballistic motion.

I've created very simple flight models for games and simulations before but never anything like this where every little element of the vehicle affects its flight characteristics. That is where the flight of the vehicle emerges from the action and interaction of all of its parts rather than treating the vehicle as a point mass.

What's really surprising to me is that its stable. I guess its floppiness means that it behaves part way between a plane and a parachute. I can make it more rigid by increasing the number of physics time steps thus allowing stronger stable bonds and damping between cells but that also slows the action down. Still, once I've got the multicore optimisations going that's a possibility. Eventually I might add control surfaces and combustion and rocket engines or a powered flapping motion.

For now though I think floppy gliders operating in low gravity environments will be fun for experiments, games and eventually maybe the evolution of avian creatures.

Ok, I'm off to improve the flight model a little...

[UPDATE] To see the flight model in action download Flight.dat from the "Share your creations" section of the forum.