So I had a bit of a scare today.  My hexa was set to loiter mode at about 10m altitude when all of a sudden it just took off like a bat out of hell, due North from its intended loiter position.  I was able to quickly regain control, but not before it had covered a good 15 to 20m.

Looking back at the log I noticed that the ground track showed a jump in the opposite direction from where it actually traveled.  This hinted at a possible error in GPS positioning.  I found the following jump in the GPSHDOP at that same moment:

Had the hexa been flying at a lower altitude it could have easily crashed into surrounding objects.  The questions are: 

-  Is it possible to limit this kind of reaction?

-  In the case of a large shift in dilution of precision, such as above, is it possible to limit the reaction of the vehicle?  Could APM fall back to some sort of dead reckoning temporarily?

Thanks for any coments

Views: 914

Reply to This

Replies to This Discussion

It would probably be a good idea to put in some error detection in the GPS interpretation code.  But AFAIK there's no dead reckoning capability in the APM, it is only an AHRS system.  Position is entirely taken from the gps.

Jake, would you know if the AHRS GPS gain could help in any way to smooth out such a response?  According to the ArduCopter wiki: "This controls how much to use the GPS to correct the attitude"  The possible values vary between 0 and 1 but there are no explanations as to what a change in the gain does exactly.

This is a typo I think.  "Attitude" should be "altitude".  I think this is the ratio of barometer to GPS in determining the altitude.

The GPS doesn't provide a attitude, it can only provide heading (yaw), position (including altitude), and speed.

Somebody should fix this in the wiki.

Reply to Discussion

RSS

© 2019   Created by Chris Anderson.   Powered by

Badges  |  Report an Issue  |  Terms of Service