Comment

Comment by Luc van Helfteren on July 25, 2012 at 9:58am

My 433 Mhz 3dr telemetry units are not working as expected (details here: http://diydrones.com/forum/topics/433-mhz-radios-will-bind-and-conf...) essentially they bind and config utility works, but I am getting a heartbeat error with missionplanner. Anyone encountered this one before?

Comment by malcolm churn on July 21, 2012 at 4:19am

Hi Dean hopefully tomorrow, just picked up shiny new GPS, care of Martin, very happy!! I think Martin is going to try and make it down to my club field. Give him a shout. I'll guest him, I can get one of the other guys to guest you if your up for it.

Moderator Comment by Dean Wynton on July 21, 2012 at 2:12am

Hey Malcolm, when and where you flying? PM me if you prefer?

Distributor Comment by Martint BuildYourOwnDrone.co.uk on July 20, 2012 at 10:47am

Malcolm, the 3DR GPS LEA-6 units have landed! :) 

 

Regards

 

Martin.

Distributor Comment by UnmannedTechShop.co.uk on July 20, 2012 at 9:54am

Hope we get some decent weather to be able to fly this weekend!

Distributor Comment by Martint BuildYourOwnDrone.co.uk on July 20, 2012 at 3:32am

I will see what the boss has lined up for me! I have a few things I need to get done too.... I should have the new GPS here today I hope the FedEx man delivers, so I might be able to let you have one on the wekeend!

Regards

Martin.

Comment by malcolm churn on July 20, 2012 at 2:56am

flying this weekend if anyone is up for it :)

Comment by malcolm churn on July 18, 2012 at 2:03am

I always do an auto trim on CH7 every time i fly, the problem is it dosent hold it everytime. when it does hold the trim it Loiters pretty well. Just need to find out why, i think its a vibration.

Comment by Luc van Helfteren on July 17, 2012 at 3:01am

Theoretically, you should be able to calculate ideal PID settings from an impulse response of a system. The problem is, without simulation, this is hard to perform. A better approach I would have thought is using a log analyser to estimate the error in your PID settings. PID tuning is an iterative process, but given the similarity of most quad copters (in that they are mainly square and have similar weight distributions) I can't see why an "ideal" PID calculator could not be produced when taking into account parameters such as weight, drag coefficient, moment of inertia and power output.

Developer Comment by Randy on July 17, 2012 at 2:23am

@Luc re auto PID tuning...Adam's part of the dev team now..and we talked about what he was working on...i think it turned out that it only worked in a very limited number of situation.  It really would be great to hear/understand the theory behind how this could be done.

My guess is that if you have a decent guess as to the delay between when the controller sends the message and when you see the result, you can kind of work up over time an idea as to whether it's reacting as your PID values tell you it should...and then adjust them slightly up or down..

..so you have some kind of little watcher sitting there..and over a long time it keeps quietly adjusting the pids behidn the scenes..it does it a lot at the beginning when it's very unsure if it's got them right..but then slows down over time as more history builds up...

• ‹ Previous
• 1
• …
• 3
• 4
• 5
• 6
• 7
• …
• 22
• Next ›
• Page