It WORKS...!! Fully automated parachute drop from quadcopter.

Thanks to the Dev team for getting ACv2.9 to us, especially as it fixes a lot of features that have been broken for a while. This one in particular is one that I had been trying to get to work - http://diydrones.com/forum/topics/automated-parachute-drop.

The mission was to takeoff to 2m altitude, fly to WP2 @ 25m alt, drop the chute, fly to WP4 to a) stay out of the chutes way and b) not yaw towards home hopefully to capture the chute while it fell with the 808 #16v2 keychain camera attached to one 'copter foot, then descend to WP5 @ 5m and land.

Some of the flight was a little wobbly as I had adjusted a few PID settings yesterday while experimenting with something else. Below is how the mission looked. The setup was followed as per the camera control wiki page > shutter configuration - http://code.google.com/p/arducopter/wiki/AC2_Camera. A simple 5 gram servo released a rubber band holding the chute underneath the 'copter.

I had to change the nav speed to 2m/s because the waypoint was very close to the takeoff point and that was still too fast as the 'copter had reached the waypoint before the 25m altitude had been reached.

Waypoint radius was set to 1m and I had turned AUTO_VELZ_MAX to 150cm/s from 125cm/s but that also made the takeoff very/too rapid.

Views: 6629


Developer
Comment by Randy on January 20, 2013 at 5:04am

Congrats!

This line below makes me think that we have an issue where a waypoint is considered complete even if you haven't reached the target altitude specified in the waypoint.  Is that true?  I thought we'd resolved that with this release.  If not, it'll be done in 2.9.1.

I had to change the nav speed to 2m/s because the waypoint was very close to the takeoff point and that was still too fast as the 'copter had reached the waypoint before the 25m altitude had been reached.


Moderator
Comment by Graham Dyer on January 20, 2013 at 5:11am
Thanks Randy, and thanks again for the hard work put into this.
Yes, I had a couple of premature drops as the altitude had not yet been reached.
The only other thing I noticed was the takeoff is pretty aggressive even on the defaults.
Comment by Rob_Lefebvre on January 20, 2013 at 6:38am

Pretty cool Graham.  I was looking at trying to drop something from mine (a first aid pack) to simulate a mission.  But was told this is illegal?  I dunno...


Moderator
Comment by Graham Dyer on January 20, 2013 at 9:14am

Randy, the descent to 5m from 25m also reached 5m before WP5 was reached (about 1:12 on the video) so it got to 5m and then continued about 3-4m to WP5. I have the logs if they are of interest.

Comment by Peter Chaffe on January 20, 2013 at 11:45am

Well done Graham, got there at last eh!
I went a different route. External MCU to control the parachute servo, this was to be triggered by the relay-on code (either pulsed or set to on, didn't matter). Was deciding on a port to use. was going to use A9, and alter the servo lib file to use a connected pin on th MCU. But then Xmas and other things got in the way, now things seem to have been resolved! At least with using the camera trigger command.

Randy, I can only speak for v2.81 which I tested extensively. As you say, as soon as a WP radius is reached, the next Wp is executed (current height ignored), however it will still climb after that wp. depending on the next wp height.
I explained it in a pdf file (if I can find a way to upload it). Not through code but through observation. All done on video too.

Graham, Had you used auto take off, then it would have reached specified height before heading off to first WP

Randy, Just for your information, Servo code still not tested, Relay code cannot work as is, due to relay.lib addressing an unconected pin on APM2 & 2.5. MCU. Yaw to a compass position not working, (this was with gimbal disabled) Think the yaw code is now only reached if gimbal is enabled. ROI not working...

-----------


Moderator
Comment by Graham Dyer on January 20, 2013 at 12:34pm

Hi Peter, thanks. BTW this WAS using auto-takeoff, but I started the motors first and actually did the lift off myself before switching to Auto otherwise the takeoff is very aggressive. The takeoff altitude was set to 2m so it went to that height before heading off to the WP. In one test the WP radius was inadvertedly set to 30m which caused the chute to release about 2m from the takeoff point at 2m altitude so WP altitude definately has no priority.

Comment by Joshua Johnson on January 21, 2013 at 7:30am

awesome!!!!!!! Keep up the good work! Please keep posting more videos.

Comment by John C. on March 13, 2013 at 3:34pm

dang guys....I've been smacking my head on this for a plane.

I can't get do_set_servo or do_digicam_control to work on a APM 2.5+

Anyone had it working yet?

Thanks

JC


Developer
Comment by Randy on March 13, 2013 at 4:01pm

do_digicam was never implemented for planes but do_set_servo should work.  I've created an issue for it here so Tridge or I (or maybe someone else from the community) will hopefully get a chance to look at it.

Getting do_set_servo working shouldn't be hard.  I think it's just a small bug somewhere.

Comment by İlhan Başçuhadar on April 7, 2013 at 11:48am

Hi Randy,

Is there any progress in the issue you created? Will we, Arduplane users be able to use do_set_servo command soon?

For a contest FFD2013 we need to release locking pin of 0.5lt water box holder during a guided flight. 3 boxes shall be dropped, so we need to activate ch5, ch6 and ch7 of APM by do_set_servo command. On Arduplane V2.71 I couldn't make this command to work, to drive a servo automatically. Maybe someone could do this.

 

Comment

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

Join DIY Drones

Groups

Season Two of the Trust Time Trial (T3) Contest 
A list of all T3 contests is here. The current round, the Vertical Horizontal one, is here

© 2019   Created by Chris Anderson.   Powered by

Badges  |  Report an Issue  |  Terms of Service