My most recent goal has been to build a FPV Racing Quadcopter as folows:
FPV Racing Quadcopter
My intent with this project is to develop and build a 250mm
FPV Racing Quad copter. I’ll put resources and links as I find them available.
Initial challenges with the MINDS-i system will be making the frame small and
lightweight enough for 5-6in propellers to provide ample thrust to weight
ratio.
As you can tell my original design was to use the MINDS-i pieces. The main problem is plastic flex, will the frame bend too much? With a little inspiration from Tested I had a good idea of what I needed to do. So I went and built this:
Here it is! I used the colored beams to mimic the flight lights on modern aircraft. (I didn't have any green beams, so I used blue.) Evenually, I'd like to use some LED's but let's just get this thing tuned and flying first. The reason for this color coordination is so that I will always know which way my quad is facing. On this I'm using the 10x4.7 props, so it's actually pretty overpowered for a small copter. It's about 12in wide and long so it is much bigger than the 250mm class quad but it's a good place to start.
PID TUNING - 12/14/15
Unsuccessful attempts at PID Tuning the quad. Changing
multiple values appears to not make any effects. In most cases it flies
normally until the error ramps up, (about 5 to 10 secs), the quad will continue
to shake back and forth uncontrollably. Changing the D seems to make the most
difference controlling how fast the quad shakes the higher the value the
faster. (D= 0.001 to 0.008. D=0.004
seems to be the best) However, no setting appears to work. We also tried the
Auto Tune function but it would deactivate the moment it was armed.
Thoughts?
I may need to set the P value below 0.08.
I would like to try auto tune again, see if it will work.
The flex in the frame may be causing it to shake.
After doing a little bit of research on these sites I returned to tuning the next day.
Helpful Links:
12/15/15
The tuning seems to be going better. I think I can negate
the up and down wobble from the string and focus on getting the “shakes” out.
It seems to almost be flyable.
From DIYDrones I’ve found the following excerpt helpful.
Starting with D.
“I’ve found the quickest way to tune the arducopter is to
start with rate_D, (but you simply can’t do this unless you can hold your beast
in hand or in a jig as the copter won’t fly without any rate_P or I) so a more
universal way to start, and actually a more informative way is to start with
rate_P.
STEP TWO – Tuning rate_D
1.
You have found a nice rate_P value
(or so you think), so leave that fixed for now. For example 0.110
2.
Set a range of rate_D from 0.000 to
0.025.
3.
Throttle up in hand with rate_D at
0.000, things should be as you left them at the end of step one.
4.
Now start turning up rate_D, again
move the sticks, shake the beast about, you will notice that at a fairly
precise point you start to see very fast oscillations. This is your maximum
rate_D, I’d set it just under where you can induce these fast oscillations by
moving the sticks fast of shaking a leg.
5.
Now you can go back to rate_P and
tune that again, you should be able to increase it considerably!!! So to step
three.
NOTE: once you have
found your rate_D you can try something interesting - you will have to hold
onto your bird for this obviously. Set rate_P to zero, then crank up your
rate_D as you just have. You will discover that these fast oscillations occur
at pretty much the same level of rate_D, no matter what you have set rate_P to.
This is why I think it’s more than possible to start
tuning rate_D then move on to P. But that’s a shortcut, and
this is not what this guide is all about.”
What I did:
Tuned D to 0.003
Tuned P to 0.087
After a few test flights, it wasn't working at all, it would either not arm at all or when did it was very under-powered. To fix the problem I went back and redid all of the calibration and parameters.
12/16/15
Yesterday: After resetting all of the parameters, re
installing the firmware and re calibrating the quad it flew very nice. I was
able to fly it most of the day.
Originally the error were that it would not get off of the
ground, like the throttle was limited. It also had the problem of it arming and
then disarming.
After I got it flying I took it out several times. There
were several crashes. It would fly well for about five to seven minutes but
then it would twist and fall out of the sky. It think it has something to do
with the low voltage failsafe not being set correctly but it’s hard to tell.
The PID seems to be set fairly well. Its flight
is very smooth. It seems to wander a bit and it does tend to shake a little in
the wind, but it is able to maintain flight and I feel like I have very good
control over the air craft. I think I’m done tuning the PID for now. I can
focus on fixing the failsafe.
I was able to fly it for most of the day on the 15th and the 16th testing how it flew and handled. However that night I took it home to practice flying it a little more and disaster struck. I write about my experience the next day:
12/17/15
Well I lost it. It last saw it floating off in the distance
just to the north east of my house. I searched hours for it but no luck. Time
to think about GPS Tracking.
To add a little bit more to the entry, I remember it flying it in circles, trying to get used to using the yaw in different directions. Flying it towards me and away from me. All the time it was in stabilize mode or in altitude mode. After flying a flew loops it slowly got higher and higher. I tried to bring it lower but there were several trees I had to avoid. At one point it started to drift and because I was not sure which way was the front when I went to correct it's course it drifted in the wrong direction until it was out of sight and drifted behind a tree. At that point I went to run to catch up to it and by the time I got there it had either drifted far from that location or crashed. I searched for hours after that but was unable to find it.
I continued to search for the quad in the days to follow. Talking to many neighbors and passing out my business cards and contact information to anyone I could talk to. I even mapped out a search area and highlighted the area I visited as well as the red areas where I could not get to, I then went and checked all the red areas the next day, but with no luck. That weekend there was some good snowfall so wherever it went was likely under snow, I continued to search through the weekend with no luck.
Since then I've started to practice more on the flight simulator Phoenix R/C version 5.5.i. I've also moved to a smaller model. This little micro quad I got for Christmas. Thank you Merrill and Rachel!
Between the simulator and this micro quad I've tried to fly every single day since the accident. I've also taken some tips from Flite Test to start in Acrobatic mode, or Agile mode. I've started that on the flight simulator and seems much more manageable now. Since then I've been following these tips in this video to be helpful.
That's my story, in the days to follow I've continued to practice flying. I've also thought about what my new quad is going to look like and laid out an initial design. I try to update this post regularly so I can keep track of my progress and research.
My newest project:
Levi came up with this design. After flying it around and testing it for a few weeks he'd like to make a few changes and I'd like to build a new quad so with that extra carbon fiber sheet we'll have to make another quad like this.
The future design will have shorter stand off's so the quad height will be lower. Levi noticed that there was plenty of space in the middle so we'd like to chop off another 0.4 in off the height. We'll have to make a few custom length stand off's instead of using the MINDS-i threaded collars but that should be simple enough. Another change is to cut out as much weight as possible while still maintaining the strength.
I'll need to get a few things to complete this quad, namely:
| The multiwii flip32+ Deluxe v5.0 |
No comments:
Post a Comment