So here is the thing. Quad follows the waypoints as fast as possible and when it comes to the dropoff point it just cant go the exact location and drops it like 1m away. But in some flights (2/10) it really drops to the right location. I really did not understand the cause of this problem. It is trying to correct its position error but it can not. Also when we tried putting 3 more waypoints on the dropoff point it could actually correct its position a little bit better. We were actually flying with 20~ satellites and 0.5~HDOP.Could you guys please help. It is urgent. Thanks.
Also As you can see in the video there are 4 bricks making a big square that is the dropoff point
1m GPS error is pretty reasonable IMO.
If it’s urgent you better run out and buy hardware for RTK. What you are describing is typical accuracy I would be happy with.
Another TEKNOFEST participant dropping water bottles?
And copter 4.1 will fix some of that issue with S-Curves, allowing the copter to slow proportionally to meet the waypoint. The GPS accuracy… see Daves comment.
I thought some people may be interested in the progress we have made with the development of S-Curve navigation.
We have got the first proof of concept working and so far it is living up to all my expectations.
Here is a video showing the performance of 4.0 navigation. This video highlights the problems with the current approach when we use low accelerations and high speeds. The first half of the video looks pretty good because the acceleration is only 1/3 the speed. When the acceler…
This is a simple comparison of the path that Copter-4.0’s “leash” based navigation takes vs Copter-4.1’s “s-curve” navigation when the target speed is high (i.e. WPNAV_SPEED = 1500) and target maximum acceleration is left low (i.e. WPNAV_ACCEL = 100).
You’ll hopefully notice these three improvements:
the vehicle does not cut corners as much
the path in and out the corners is symmetrical
the vehicle’s heading more closely matches the path the vehicle is following
This enhancement should also…
I’m just dropping by to show an example of a real flight with Copter v4.1. In this case, it is the difference between approaching a waypoint between trees with and without the S-Curve.
In this case, the vertical approach is much safer, moving the quad away from the trees in a clearing:
Blank is the planned trajectory.
In red is the quad flying with version 4.0.7 WITHOUT S-Curve
In orange is the same quad flying with version 4.1 WITH S-Cu…
Yes thanks for your answer. So that being said we can not actually buy a RTK GPS device and is there anything possible to make our GPS more accurate ?
last test’s GPS Datas by the way I’m using Neo m9n dual compass module
Also I want to state that our quad’s speed and acceleration is way over the limits. We are actually thinking the problem is there. We had much worse GPS’s like m6n and way more slower devices they were doing their waypoint missions perfectly I would say
I wonder if
choosing the day and moment (smaller HDOP) would make any difference.
I dont think I will achieve better HDOP rates than 0.4 HDOP
Note that HDOP varies along the day and is predictable; you can relate HDOP with your results quality.
I don’t think will much matter. I fly Missions >30 m/s and GPS is fine (have to use at least Beta7). Standard M8N. But I’m not looking for 1m accuracy.
M9N Position accuracy:
CEP-Circular Error Probable, meaning it could be better or worse.
while I was observing the flights I noticed something. Quad actually goes a little bit right side when flying waypoint missions. Have you ever experienced something like that
And do you think ESC Based notch filtering would help ??
Not if the Acclerometer calibration is good and the craft has sufficient thrust for stability if the load is offset.
No I meant that waypoints are treated as splines like there was an obvious curve on the mission (As you can see in this link:
https://drive.google.com/drive/folders/1-1s-5J7hOsDKs_q20o5_UjrZZJ-xM4sa ) and even if the quad can drop into that area it cannot drop the desired location. @dkemxr Have you heard something like that @amilcarlucas
There are multiple videos in that link. If you want someone to review a video offer an explanation for ONE video. Videos don’t have much value for flight analysis.
No I just wanted you guys to see the curve (btw I updated the link) and here is the .bin file
2021-09-04 01-41-26.bin (898.1 KB)
It went out of control. It just treats the waypoints as splines like it goes right and it drops somewhere else which is not even close. Here is the link I uploaded the last flights .bin file
https://drive.google.com/drive/folders/1-1s-5J7hOsDKs_q20o5_UjrZZJ-xM4sa?usp=sharing @dkemxr @amilcarlucas @Leonardthall Please I beg you to take a look at this
Why are you running 4.0.6? Update to 4.1.0-beta8. There are enhancements and additional parameters useful to you.
will my PID Values come with me and will it be safe to fly