Version 2.5 of the ArduCopter code is now available in the AP Mission Planner and in the downloads area!

 

Enhancements:

  1. The first ArduCopter code release optimized for the APM 2.  Leans and drift should be much reduced or even eliminated for most users.  This was accomplished through a number of core improvements to the DCM implementation by Tridge like this one and this one.
  2. Loiter and waypoint following should be improved due to a D term bug fix, some tuning and the improved DCM performance mentioned above.
  3. On start-up, the yaw heading is updated with first successful mag read (so you should no longer see the slow rotation from north to your actual heading).
  4. Increased output rate to ESCs to 490hz.  This update rate is also user selectable using the new RC_SPEED parameter.
  5. hexa copter stability patch bug fix (should resolve slight flattening when pitching forward and accelerating very rapidly).
  6. improved baro filtering
  7. fix to dataflash logging of Mag heading
  8. addition of H1 swash plate type and bug fix for proactive yaw compensation for collective pitch changes for TradHelis.

 

Tuning:

The default PIDs are optimized for a 3DR/Jdrones quad with 850 motors and 10" props. If you're using more powerful motors/props, start by turning down Rate Roll P in 25% steps.

While some testers have reported very good flights with the default PIDs, some have reported that this release is a little "sharper" due to the DCM improvements and have found they needed to:

     reduce stabilize P by 10% (i.e. 4.5 -> 4.2)

     reduce stabilize D by 30% (i.e. 0.15 -> 0.10)

     increase rate D from 0 to 0.001

Tuning loiter can be tricky.  Refer to the discussions which will appear below for more community feedback on what parameters work best.

 

Please post your feedback in this discussion.  For enhancement requests  and bug report, please add them to the arducopter issues list.  When possible please include logs (tlog and/or dataflash) and tell us whether you're using APM1 or APM2 and what version of the software you're using (presumably 2.5 but tell us anyway!).

 

Thanks for this release go to the developers (both in the core team but also those who have provided bug fixes through the issues list) and also the community members who participated in the previous release thread and provided some great detailed information in the form of issue reports and logs which allowed us to nail some bugs!

 

Tags: ACM, APM1, APM2, Heli, Quad

Views: 115172

Reply to This

Replies to This Discussion

Do you changed any PID Settings?!? @Lars

wow, for the first time, is very stable. I think more than the flight, assembly and calibrationof equipment Congratulations, I'm learning to arducopter, I have some experience flyingwith the old vti Draganfly.

I have a question about Simple Mode.

Last week I had a "hard landing" because I kind of lost orientation in Simple Mode, de copter was not reacting like I thought it would react.

Only now I realized other multicopter systems use 2 different kind of Simple Modes, see picture below.

What kind of Simple Mode does AC use?  "In Course lock" or "In Home lock"? Or can you actually choose in MP?

I always thought that pulling the stick back would bring it back home, no matter what direction the copter is in.  It didn't, it reacted in a different way resulting in a hard landing. 

I believe you have the choice of either with the latest firmware. The top diagram represents normal Simple Mode and the bottom represents 'Super Simple Mode' (see the descriptions for each here: http://code.google.com/p/arducopter/wiki/AC2_Simple )

Thanks for your answer.

I'll have a look at the wiki.

I'm using 2.5.5, hopefully it's already integrated in that version...

I just had a look at it.  When I enable super simple mode, is it active also when I'm flying very close?  Or does enabling super simple mode mean that super simple mode is activated when I'm out of range (+20m away)?

I'm pretty sure super-simple mode was in 2.5.4 at least, and maybe even before that, so 2.5.5 should be fine.

I'm not sure exactly how it reacts when close to the home location. I've only flown "normal" simple mode. Obviously when you get very close to the recorded 'home' location there's going to be some slop depending on GPS resolution, etc. and I vaguely recall another discussion that mentioned a buffer zone around 'home' to prevent freaky behavior, but I'm not sure what it actually does in that zone.

Hmm...  So super simple mode is only used with a buffer zone because of inaccuracy of the gps.

Not really what I want out of a simple mode...

 

Simple mode for me is a mode where I can save my multicopter when I loose orientation no matter where the copter is.  In other words, I'm pulling back my pitch stick, the copter comes back to me.

 

You might want to explain simple mode better in the wiki.

 

So super simple mode is definitely only working when the copter is more than 20m away?

My understanding of super simple mode is patched together from a few other discussions, so don't take my word on how it operates. Regarding the buffer zone, no GPS is going to be perfectly accurate to a single spot on the ground so there's going need to be a buffer zone of some kind for this kind of implementation, unless you've got some kind of beacon on the ground it can find as a reference.

I just looked at the code (2.5.4 is what I've got) and the angular offset is recalculated if the craft is 10m away, not 20m as is implied in the wiki. So the relative 'orbital' position and therefore the angular offset is re-calculated constantly if you're 10m or more away from the home location. If you're closer than that, it still behaves like simple mode, it just doesn't update the angle relative to home until you fly back out to that 10m mark.

 

So for example: if you're flying 15m away and pull the stick toward you (assuming you are standing on the home position), it will come straight toward you, fly over you, continue past you until it is 10m away, then re-calculate it's angular offset. At that point it will compensate to fly straight for you again, without you ever changing the controls.

While below 10m away, it uses the angular offset it last calculated when it was > 10m away, and simple mode still applies. The only behavioral difference within that 10m circle is if you hold the stick to the left or right, it won't orbit the home location; it'll fly a straight left and right line relative to your facing direction (when it was last at 10m) until it gets out to the 10m mark. When it flies left/right 10m away, it starts re-calculating the angular offset again and would orbit.

This seems fairly reasonable to me, since the only truly disorienting behavior would be if I've got it 10m out and I'm facing it, I pull the stick down and it flies toward and past me but slowly enough that it doesn't get 10m away again behind me. I turn around, and then the controls are reversed. But only until the craft reaches 10m again, then all is well from that direction.

I suppose you could probably get this value down to the minimum trustable resolution of the GPS, maybe 5-6m? At that distance you're probably not even fully notice the buffer zone unless you're hovering slowly over your head a lot. The value is one number around line 1857 in ArduCopter.ino if you'd like to play around with it.

I think I fell into the super simple trap. See photo. After waiting for the 3D lock on my quad I armed the unit. The problem is the GPS was still about 10M away from my real home location. after flying around in stabilized mode I switched to alt hold super simple. By this time the GPS had corrected and had shifted. in super simple mode I hovered and made some 360s and then decided to do a slow rotation around myself keeping the nose away from myself. As i drew closer to the original lock point the controls became all mixed up and I lost control and crashed. When analyzing the log data it indicated to me that I drove it into the ground when relay I was trying to save it. It was not until I reviewed the kmz file at the same time as the on board video that I put it all together.

I think using super simple within 20M will cause problems like the one I had.

Lesson learned.

Attachments:

Hello Emile (another Emile on the block :P ).

I know for sure that GPS accuracy needs time before it settles to a correct location.

It needs to load the satellite's almanac and this is known as cold start.

Even though it gets a 3D lock in about one minute or so, it could take up to 10 minutes before you have a precise location and this depends on the number of satellites and sky visibility.

I would suggest to try to set a backup battery on the unit. You need just a solder and some patience but it can be done.

Use this link to get an idea on how to do this.

http://code.google.com/p/ardupirates/wiki/MTK_GPS_backup_battery?ts...

Cheers,

Emile

Thanks Emile

I am using the new APM2 board and have not seen any hack for the battery install yet. I know it's a pain at battery change to go through the GPS lock all over again...

Emile L

Hi Emile. I had noted before via kmz files that even after lock the gps needs about 20sec to correct the position. We should always wait about 1min after lock before arming, or home position will not be so accurate and may go to 10 or 20mts away, wich will not be a problem if you're in an open field but may be problematic if you have trees or houses around. I have never tryied super simple but I'm curious to try it.

RSS

Social Networking

Contests

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.

A list of all T3 contests is here

Advertisement

© 2013   Created by Chris Anderson.   Powered by

Badges  |  Report an Issue  |  Terms of Service