Today i have done a ground test and it shown good results on ground navigation test it sensed the location i changed code little bit for navigation so that it shows the waypoint achieved with the buzzer beep. And it was succesfuly alarmed me the waypoint captured . The next step i did i load 2.0 version and i have fma xy sensor, and no telemetry so used buzzer for ground test navigation. Well i put the original code of jordi of 2.0 ver and i just did the settings as prescribed in 2.0 user manual.(nose down the plane and sensed it for the calibration.) But here was the problem in stabilization , the elevator movements was not exactly according to the manual. i connected fma correctly. And about weather conditions it was bit cloudy today will it be the reason. Sorry i m quite behind u guys but soon will come as i wil get some budget to upgrade it . And i m using twin star plane model and using elevator and aileron combination..... Please suggest me what could be the problem and any changes required to made in code. Well i will test it again tomorrow morning hope i wil get good valuable suggestions from all of you thanks in advance guys.

Views: 122


Admin
Comment by Thomas J Coyle III on August 13, 2009 at 4:25am
viky,

Why are you using an elevator and aileron combination? Shouldn't it be a rudder and aileron combination to yaw the aircraft?

Just a thought.

Regards,
TCIII
Comment by Vikram on August 13, 2009 at 4:29am
aileron can do the rudders task easily and its specied in manual also
Comment by Bryan Cuervo on August 13, 2009 at 4:49am
You mentioned the elevator stabilization movements were not exactly according to the manual. I'm guessing you are referring to this part,
"So if you put your hand in front of the XY sensor (which looks to the autopilot like the plane is pitching up), the elevator should go down. If you put your hand behind the sensor (simulating pitching down), the elevator should should go up."
If I understand the system right, your warm hand would mimic the warm earth so placing your hand in front of the sensor would mimick the plane diving and the elevator should deflect up. An easier test while you are at the field is to simply point the plane's nose down. This should cause the elevators to deflect up. If you roll the plane to the right, the right aileron should deflect down to lift the lower wing back up.
Hope this helps.

Moderator
Comment by Gary Mortimer on August 13, 2009 at 5:00am
Where did you place the sensor on the Twinstar?? That might be an issue.
Comment by Vikram on August 13, 2009 at 5:27am
i had placed on top such that nothing bocks the sensors visibility . how to increase the senstivity of sensor that may be the problem
Comment by Vikram on August 13, 2009 at 6:03am
So How to increase the sensitivity of sensor from code itself any idea
Comment by Vikram on August 13, 2009 at 6:32am
thanks bryan this is waht i m expecting but can you tell me if i want to test it inside my laboratory how should i calibrate it ...... just for testing because i want to work on it on night so i have no other option has to test it inside room and make it work ... other suggestions or links are most welcome
Comment by Vikram on August 13, 2009 at 6:33am
cant wait till tommorow morning thats why i m in quite hurry and has to test it on air again tommorow morning ..
Comment by Vikram on August 13, 2009 at 6:44am
Well i have run some sensor test code also it s showing changes with the hand movements also in serial port values. Let me to check it out by fitting it into plane but i would like to ask is there any factor to increase its sensitivity......

3D Robotics
Comment by Chris Anderson on August 13, 2009 at 8:02am
Viky,

Yes, you'll have to tweak the PID gains in the code. Because you're using the old code that's not supported anymore (I don't even have a copy of that anymore) I can't tell you which, but try looking for the pitch proportional and roll proportional constants and increase them for more sensitivity.

Comment

You need to be a member of DIY Drones to add comments!

Join DIY Drones

© 2019   Created by Chris Anderson.   Powered by

Badges  |  Report an Issue  |  Terms of Service