Hello, i just received my first apm2 board and am trying to tune it to fly in acro mode for fpv. My problem is when i send the quad in any direction it will start to pitch back the other way shortly after i gave it the command. If i send it forward it will pitch back about 10 to 15 feet and not continue forward like it should. I have tuned the rate p gains in pitch and roll so it gives a nice crisp response without much bounce back, still get the same reactions. Also it does not want to fly straight, this might be because my quad is a little off balance from side to side but the for aft cg is perfect. Does it need to be perfectly balanced to fly in acro mode?

Views: 728

Reply to This

Replies to This Discussion

Why on earth would you want to tune it in Acro mode for FPV, Acro mode is ACROBATIC mode which doesn't use the accelerometers, just the gyro's so it'll easily rotate past 45 deg and is capable of doing a 360 deg flip.

Use only Stabilize mode!

To get go acro performance you must use rate_i terms for roll and pitch. Otherwise an imperfect CG or other factors will tip the quad.
Jason

stabilize mode is not how i am used to flying, i currently fly a naza on my other quad and i fly in manual ( acro ) mode, its alot more like flying my helis. Flying in stabilize mode is boring for fpv, i like to do loops and rolls and fly around kind crazy, maybe the apm2 is not really the right board for fvp, i just wanted to try out something besides naza that has rth gps.

I completely understand why you want to fly FPV using acro mode. Stabilize great for hovering and cruising around slowly but it quickly becomes limiting as the speed increases or you want to be a little more creative with the way you fly.

Unfortunately Acro is still a baby as far as development goes on the arm2. It is currently based only on the rate sensors with no feedback to maintain the attitude if you don't touch the sticks. This means that aerodynamic effects can rotate the copter in pitch and roll.

When I used acro (before I realized it's limitations) I found that the copter would quickly rotate back to a horizontal attitude when flying into the wind.  This would happen if I was hit with a sudden gust of wind also. I came to the conclusion that this was the dihedral effect of the propellers. Basically the copter is trying to maintain zero airspeed.

I am currently putting together a fix for this problem but it is my first attempt at contributing to the code base so it won't be quick.

I have found something in the code that might partially fix the problem though. If you set AXIS_ENABLE to 1 the code does do basic attitude feedback for ACRO. I have not tried this out though so I would make sure that you are at altitude and have STABALISE handy.

On other word of warning. You can do 360 degree in roll but not in pitch. As the copter goes through vertical the yaw will kick in 100% to rotate the copter 180 degrees. This should be fixed relatively quickly though.

Jason is correct about the tuning. This should only improve things if there is a balance problem. I use this as a good starting point:

http://diydrones.com/profiles/blogs/arducopter-changing-from-pi-to-...

Basically these are problems that will get fixed but it takes time. I think the development team has focused on the UAV implementation (and they have done an excellent job) rather than the fly by wire aspect.

thanks for the informative reply. I am going to switch my apm2 to a f450 frame and work on tuning. i will have to use another board for my fpv, thanks again for the reply.

Hi!

I am also one of the acromode - lovers and i was a little disappointed with my apm2. Finally i cranked up the rate I to 0,35 and increased rate D a little. It still flies not so good in acro like my Multiwii quad or even my Blade mqx :). Still doesn't do a solid hold of the steered angle. Anyway it's ok now.

So long

Kraut Rob

Hi Rob,

i am also diasappointed with the acromode. The mutlwii is much better ;-) But it´s gone up in smoke :-(

Did you also tested some imax settings?

I increased i to 0.300 and Imax to 90. So the APM  will hold the angle much better for my small quad in Acro.

regards

Florian

No, i haven't tried that yet. Instead i fly my multiwii :))

So long

Rob

hi, hay i might be wrong here so please correctt me if i am , but if u dont have acro mode tuned right stabilise wont fly right

That's what i thought, but from my findings it does not apply to the apm logic - only to a limited extend.

You can have a good stable mode but a bad acro mode at the same time. I just put the acromode away from my flightmodeswitch :( .

I think it's genetic :) . The apm tries to be a flying robot wich can be manually controlled. The multiwii was made for piloting and tries now to adapt the apm gps functionality.

Just my thoughts on this.

So long

Rob

Hi All,

My advice for great ACRO mode for FPV:

1. Set AXIS_ENABLE to 1. (this is a rate controlled Stabilized mode so no flips, max angle is 45 degrees)

2. Set the stabilized I terms to zero and tune using the Rate I terms. This is critical.

I would be very interested if people would try this out and give me feedback because I want to improve this further. We are making some foundational changes int the APM code to get ready for a much better and unrestricted ACRO.

The default ACRO mode is just a rate feedback only mode and that is why the aerodynamic forces tend to level the copter all the time. I believe that this is also responsible for the flip after forward flight problem (I believe this is made worse by using the stabilize I term because it is slower).

Sorry for the necro bump.  I'm in the ACRO flying camp and what I can say after swimming in that scene is that you're going to see a steady march of acro flyers coming over to the APM camp.  The current thinking is that MWII is a great place to start for new flyers, but to get the real McCoy, you need to jump into the APM camp.  

This team has done amazing things with the platform, and it's only getting better with time, case in point, the work being done for inertial compensation in position hold modes.  Absolutely fascinating stuff.

In any event, there's been a slow but steady push for a multirotor racing league, and what we're seeing is something as simple as a KK5 board paired up with a skilled pilot is enough to do those sub 45 second pylon runs.  I'm smiling as I write this because in 6 months time, those times will be cut in half.

I'm also seeing development in the ESC world that's effectively doubled the input resolution to 1khz, increased stick resolution from SimonK's 800 steps to 2000 steps, and added in regenerative braking (freewheeling current) to the mix.  What I see Warthox do with a nanowii paired with these ESCs was just astonishing.  Some of the fastest speed changes I've seen short of CC3D controllers. 

Anyway, long story short - I'll be flying the APM colors in the races, I just wanted this team to know there's a small but growing contingent that would love to embrace APM for acrobatic flying.

Leonard - I'll go ahead and make the changes you mentioned above and report back...My acro currently flies like crap, stabilize works, but I fight it every step of the way (tendency to level in forward flight and turns etc)

STAB_P 4.0, zeros for I and D

RATE PIDs are .0950, .0600, .0030

HT-FPV, NTM 2826s, 8x4 CF Props, Skywalker Quattro 4-1 (soon to be replaced with SK flashed ESCs)

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

Groups

Advertisement

© 2013   Created by Chris Anderson.   Powered by

Badges  |  Report an Issue  |  Terms of Service