Society of Robots - Robot Forum
Software => Software => Topic started by: bboy8012 on February 23, 2009, 12:04:07 PM
-
Im trying to find the mathematics to make a 6ch Heli hover. My overall goal is to create a board that will help new Heli pilots learn to control those planes, and eventually it will become autonomous, but need to start out slow first since i am a beginner. Thanks
-
What do you mean by the "math to make a heli hover"? From simple physics we get that the sum of the forces acting on the helicopter equals 0.
Beyond that, you need to look at all the axises the helicopter can move in and keep the motion there to zero. If you are starting from a hover, just keep the acceleration to zero. Just like with the gyro you probably have for the tail rotor already.
The problem is, if you start to drift, and then keep acceleration to zero, you will be constantly drifting. So you will need some other way to measure velocity in all directions. You will need to look at maybe GPS or some other way of triangulating position, or look at using computer vision to calculation motion.
-
gps,vision and accelrometers and gyroscopes are a must,...for the uav ur designing ull need lots of funding and a gud team... and if ur a beginner then ill say byuilding a simple obstacle avoider is a big challenge in itself...so choose a simpler project..
-
Note that you will probably not see any profits from this project, there are virtual flybarless systems for $200 that can stabilize the heli and do flybarless.
Anyways follow the $50 robot tutorial first, it may seem irrelevant but you will need to get familiar with microcontrollers and how to implement sensors.
-
Thats a really really ambitious project for your first one. It may not seem that difficult to do but for your first project it will be very overwhelming!!
I suggest starting out with something much simpler for your first project such as the $50 robot. It will answer a lot of questions for you!
-
I'll try not to tell you forget it.... cause it's harsh....
First of all.... you need funding... for every rotor/ gear/ servo/ tail broken...
Secondly... you need to be good with maths....
As you correctly suggested, such flying machines are good to be based on a mathematical matrix...
that will calculate all four dimensions..... Yup!!! Four..... Cause you will surely need time for calculating the d(ifference)...... :P
you will also need time for speed calculations and so on, cause the bigger the meter of the speed, the
sharpest your correction must be.... Also, see here that you really need filtering... that just wastes more time....
After some theoretical calculations, the real world, must not differ very much.... cause..... it's gonna.... boommmmmmm..... NO GOOD.....
You need also, imagination and courage to make the mistake aka testing......
If you are good, with physics, and advance mathematics then it should be alright.... I assume you know how to code well......
May Zeus be with you..... hehehehehe :P :P :P :P :P :P :P :P
Best of luck!!!
Lefteris, Greece
-
I know that this may seem like a really simple solution but say you have a 3d accelerometer. Now a microcontroller can decode what commands are coming from the remote control. If the forward/ backward control is measuring center i.e. no movement and the accelerometer is detecting forward motion then surely it is easy to assume that the microcontroller should just adjust the rotor heading backwards a bit. The same goes for left/right cyclic. If the remote controller doesnt command any movement but the accelerometer is saying that its moving left then just adjust the rotor head to the right to counter it.
is this just too simple? or do you really need to build up a full on mathematical model?
-
the coffee machine is prone to oscillations.... unfortunately.... :P
You need at least PID to do this....