If anyone has any issues or logs to analyze please post them here. I've also updated the Wiki to detail out all of the AP mission scripting commands. If any command doesn't work as expected please post that here as well.
You can also post issues to the Bug list here: http://code.google.com/p/arducopter/issues/list
BTW, I just updated the Firmware version so settings will be cleared. You'll have to redo the level, radio, compass, etc for this version. This is to ensure no bad EEPROM values cause conflicts in flight.
Jason
Tags:
Permalink Reply by Greg on September 26, 2011 at 3:13am Jeff: I started testing with 2.0.44 on the weekend and then moved to 2.0.45 after Chris mentioned some fixes posted. In the 2.0.44 post I read the following comment from Jasonshort:
"Comment by jasonshort on Saturday Alt Hold needs to be enabled in the config.h file. It's still experimental, but Randy has flown it and it works. Right now it blends with GPS."
I assumed this meant I could not select Alt Hold from the mission planner using the modes menu under setup at the moment. Whilst I did start initially on this project using the Ardunio IDE an compiling the code manually, I have found the MP very convenient. Of course I have had to run MP in Windows rather than on my Apple OSX laptop. Anyway I ended up not testing Alt Hold. I see in further posts today people are testing ALT Hold enabled from the missions planner so this must have been a 2.0.44 specific thing or I have misinterrupted Jason's comment.
I here you about the weather. Its was awfull in Canberra on Sunday. In the end I did a high wind test on 2.0.45. Even in stable mode I had to fight hard for my quad not to be blown away. Hope the rain clears for you soon.
Permalink Reply by Jose Angel on September 26, 2011 at 4:28am Greg,
Jason comment was regarding the altitud hold using "optical flow" sensor, which is absolutely experimental and needs to be specifically enabled in the code.
The version tested by users in this thread, use sonar/barometer for altitude hold and gps for loiter, so it is safe to test.
Good luck!
Permalink Reply by Greg on September 26, 2011 at 1:03pm I tested v45 this afternoon in stable no simple, alt hold w simple, and loiter w simple.
each mode worked great. Also found that going from stable w no simple, to alt hold w simple
keeps the compass forward holding with controls:-).
This is great because in future I plan to do some FPV flying while changing modes.
Two motors have bearing slack. Replaced one motor exept one on the right rear wiggles ~1mm up/down.
Because of the bad bearings it tends to trim drift to the right over time.
New parts will arrive next week.
This build has been a enlightening experience
Permalink Reply by Greg on September 26, 2011 at 3:24am Martin: Just curious about your Loiter test. Did you have to feed in much throttle to maintain altitude i.e did the copter tend to mosts descend or stay within a range?
The copter held level using the sonar with a range of 5 meters and less.
I did not notice any creeping. I have noticed with all prior versions that if the
throttle is below the hover point in stable along with the battery depleting it lowers altitude over time.
The latest versions have better altitude governing with the throttle position.
Older versions the throttle had to be at maximum for a many seconds before a response was seen.
The current version has a tighter response or less deadband.
I have to be careful when yawing to not move the throttle.
New bearings will come in either tomorrow or the next day.
Am going to use graphite along with synthetic motor oil to hopefully get more life out of them.
Permalink Reply by Greg on September 27, 2011 at 4:40am Thanks Martin for the feedback. I am just curious though I thought loiter did not have auto throttle only interactive. In other words you need to manually control the throttle in loiter to maintain altitude. With alt hold the baro and sonar are used but there is no GPS position held like loiter. Seems it would be good to have loiter include altitude hold or have I missed something in the mode descriptions.
Permalink Reply by Jeff E on September 26, 2011 at 7:28pm Finally had a small window to do test, stopped raining but still gloomy weather now its raining again.
- Alt_hold worked really well both sonar and baro but baro seems more aggressive or maybe because there's a slight random breeze
- Loiter might be a bit aggressive but maybe because I left a small rate_i will try at 0
Here is a short on-board clip
- its on alt_hold (start of vid) but it stays in place (wish that's how it is on loiter)
- (0:16) Notice again a drop on throttle switching alt_hold to stab this time I was on 50% stick input luckily I was really trying to climb fast
- I must admit the fast climb was stable and just shoot straight up
- (0:30) Triggered loiter after that but had to bail out after 10sec as it was creeping towards the tree on the left (favorite for my quads)
- not included on the vid I had a alt_hold about 40-50ft and it was well and did a pirouette and it held with in a 5foot radius which was like on v42

I'm pretty sure that was your Baro sensor drifting since you started. Try re-arming it again after it warms up.
That will recalibrate the sensor.
Jason
Permalink Reply by John Hanson on September 27, 2011 at 2:51pm Just had fun with .45. Stable is great, Alt hold under sonar is excellent. Loiter did very well, out of several minutes it only wandered a couple of times.
I ran a mission (middle of log) of 5 waypoints in a circle - it came close, but I stopped it when it went past the start point.
In general, unit works well but there is one big problem I have duplicated twice (at end of this log). I set it in simple mode with alt hold. Initially it seems fine. I then yawed to left 90 degrees and it was a little shaky but seemed to stablize and went in the correct directions. I then yawed another 90 so back arm was away from me. It immediately flew really fast to the right, moving stick to left no help. I then cut throttle down and it kept going! Finally the motors stopped when I flipped back to stabilize and luckily it was low, only lost one prop. This is the second time I have had this exact behavior.
This sure is getting fun to play with. Thanks to Jason and the crew !
Season Two of the Trust Time Trial (T3) Contest has now begun. The fourth round is an accuracy round for multicopters, which requires contestants to fly a cube. The deadline is April 14th.1280 members
16 members
182 members
86 members
51 members
© 2013 Created by Chris Anderson.
Powered by
