R/C flight sim programming

A better way might be to contact any of the existing flight sim devs, especially the free/shareware ones and see if they let you port the code so it compiles with SDL and/or OpenGL. The tricky part is IMO the physics engine, thus an existing sim might be a better way to start.

Jen

Would you consider adapting a full scale Linux flight simulator to R/C?

If so, Flight Gear would be a great starting point. FG has documentation on several flight dynamics engines that are included in its code - through not very good for the beginner.

I had a go at a project like this a few years ago, managed to interface buddy cable from R/C Tx into Linux kernel 2.4 series so that it appeared like a pair of analog joysticks to the FG interface. Then I had a go at hacking the parameter files to represent R/C trainer based on the default Cessna but it was all taking too much time and had to give up.

Branko, It occurred to me that the delta between power to weight ratios between rider scale and models might make that type of adaptation not work very well. C152 has approx 110 HP and a gross of around 1600 lb while many have models weighing in at 5 and 6 pounds or LESS and 1 to 1.25 HP on the nose. The physics engine might get clobbered ..

You are spot on. I got stuck trying to scale down several attributes in regular increments but beyond a certain point it ceased to scale properly. I still reckon that FG as a basis for R/C simulator can be made to work but not with full scale equations in the code.

| You are spot on. I got stuck trying to scale down several attributes in | regular increments but beyond a certain point it ceased to scale | properly. I still reckon that FG as a basis for R/C simulator can be | made to work but not with full scale equations in the code. ...

| > It occurred to me that the delta between power to weight ratios between | > rider scale and models might make that type of adaptation not work very | > well. C152 has approx 110 HP and a gross of around 1600 lb while many have | > models weighing in at 5 and 6 pounds or LESS and 1 to 1.25 HP on the nose. | > The physics engine might get clobbered ..

Models and full scale planes obey the same laws of physics (and I won't get into Reynolds numbers here, but if you want to know why models fly differently than larger planes, google for that.) If a flight simulator models everything about a plane, then it will do the right thing when the plane is scaled down. If the flight simulator uses lookup tables to calculate things, then that might fall apart when the plane changes.

As an example, as I understand it, X-Plane models everything, and Microsoft Flight Simulator (at least the 2004 and earlier versions) used lookup tables. No idea about Flight Gear.

In any event, modifying an existing flight simulator for R/C use should be MUCH simpler than writing one with similar features from scratch. If the physics are wrong, then you simply create new lookup tables.

I gave a list of flight simulators for Linux in another post a few minutes ago -- most were open source and two were R/C simulators rather than full scale simulators. Most have Windows ports, if that's what you're into. They might make better starting points.

I forgot to mention another flight simulator for Linux -- YS Flight.

Start with an existing open source sim, and either branch it or coordinate with the developer to do your own work.

They're out there for the googling.