3DR IRIS+ Follow Me and Auto Mission Demo

We did some more IRIS+ testing this past weekend using the Pebble watch and "Follow Me" mode.  Here is the video-

Views: 994

Comment by Jack Crossfire on October 28, 2014 at 8:28pm

Would have been neat if the entire flight from start to finish was controlled by watch.  It's technically much easier than Google Glass (does anyone remember that heap of trash?), but expensive. 

Comment by Superwalloon on October 29, 2014 at 4:03am

Comment by Superwalloon on October 29, 2014 at 4:06am

Translations in Follow Me are looking pretty rough to me... It's like "oh so you've moved, let me catch up...zzzz... oh so you've moved, let me catch up...zzz....". Now I understand that you only walk away a few feet and translations might be smoother if you were riding on a bike for a longer distance. Maybe it's because the weather was windy. Maybe it's because of the IRIS+ being a bit of an overweighted machines and a good PID tuning is difficult to achieve.

So all in all Follow Me looks very promising but needs obviously some smoothing.

Comment by mP1 on October 31, 2014 at 12:04am

Superwalloon: Iris is not overweight, its underpowered. The frame is reasonable but the motors are weak, thats its most obvious weakness.

Comment by Superwalloon on October 31, 2014 at 12:56am

Frustrating is that the IRIS+ has not solved that problem. When you check the throttle mid it is quite high. Indeed it might be more underpowered than overweight. I just built a Sky Hero Little Spyder with an APM 2.6. The beast is heavy, but my throttle mid is right at 500 and it feels crispy and reactive - good motors and props.

Comment by Euan Ramsay on October 31, 2014 at 1:20am

Ignoring the weight issue, I think Superwalloon has a valid point - it is a bit "stop/start" in the video. Maybe a telemetry polling rate issue? It looks like an update command is only being sent/received every 5 seconds or so. It looks like a programmatic issue, rather than a physical one, so should (hopefully) be an easy fix/tweak.

Comment by Euan Ramsay on October 31, 2014 at 1:32am

Having said that - heading control looks good, going by the video; just seems to be x/y position refresh rate looks a bit slow.

Or...watching it again...it looks like the x/y speed is controlled by the normal waypoint speed. Which is slower than Andrew's walking speed, obviously. Or below a speed that WP speed can manage.


So it tries to accelerate to the preset x m/s and quickly reaches the follow me "near" x/y position tolerances because it caught Andrew up, so it stops. The "far" x/y position tolerance is quickly hit (as Andrew continues to walk away), so it tries to accelerate to the normal WP speed again. And repeat.

Somebody stop me now, if I'm talking rubbish. It's just a theory.

Comment by Superwalloon on October 31, 2014 at 2:25am

Good point on the acceleration ! I'd like to see a description of the follow me mode, ie I would expect that the roll and pitch PIDs are being soften and indeed acceleration could be lowered a bit. I don't like the idea of delegating all stabilization to a camera gimbal.

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

© 2018   Created by Chris Anderson.   Powered by

Badges  |  Report an Issue  |  Terms of Service