Does RTH work with Compass Error?

[jonebk12]

I don't see why the quad would have to rotate at all as it does the trial and error GPS method to go home - these things can fly in any direction.

 

[Mimoid]

Pulling the throttle stick backwards will result in an eventual landing.

Yep, I meant the right stick!

//Tom

 

[Jsun]

One observation with the Phantom 4: I was flying together with a colleague of mine. He had NEVER EVER calibrated the compass. None the less, he had experience no issue with failsafe RTH. The app itself was yelling Compass error, calibration needed etc. No IMU calibration has ever been performed. As he told me he had updated it and figured out how RTH worked, and that's it

Yes, I calibrated the compass for him, and strongly informed him to RTFM and calibrate the IMU.

 

[soulphantomflyer]

I might be wrong and I usually am
But I thought GPS location is based on set of global sat triangulation with your GPS receiver. While GPS tracking is using the same set of data but over time as your receiver move to a different point of location
As far as I know. GPS have nothing to do with compass at all


Sent from my iPhone using PhantomPilots mobile app

 

[ianwood]

I swear some of you really don't read any of the posts, do you?

Here's a summary: it doesn't work!

Why? Go back and read what's already posted!

 

[Mark The Droner]

I don't see why the quad would have to rotate at all as it does the trial and error GPS method to go home - these things can fly in any direction.

It doesn't have to rotate, but it will rotate, because there's NOTHING to keep it from rotating. It's not going to magically stay fixed in one position. It's going to spin around. And because of that, there's no way to drive it in any particular direction.

I swear some of you really don't read any of the posts, do you?

Here's a summary: it doesn't work!

Precisely.

/thread

 

[Mimoid]

And that is the answer I got from a DJI representative as well.

If you encounter a compass malfunction in midflight, you should immediately switch to Atti mode and treat the Phantom as a fully manual RC aircraft. There is no need to panic - although the drone might start drifting with the wind, it will not crash, and you are still in control - so just use FPV to fly home and land. However, all GPS-based functions should be treated as unreliable and should not be used. This also includes RTH.

Do not engage any of the automatic nav guidance functions (such as RTH or Home Lock) in a compass error scenario, as they might cause the Phantom to fly in the wrong direction where you might eventually lose the control signal or the video link, and even autoland in the middle of nowhere, far from the actual home point.

Without having any statistics at hand, I would dare to say that most of the reported flyaway stories are caused by panicing and handing over the control to a temporarily malfunctioning navigation system, instead of simply regaining control and flying the darn machine home. Practising long distance flights in Atti mode + FPV is not a terribly bad idea!

//Tom

 

[BudWalker]

It doesn't have to rotate, but it will rotate, because there's NOTHING to keep it from rotating. It's not going to magically stay fixed in one position. It's going to spin around. And because of that, there's no way to drive it in any particular direction.



Precisely.

/thread

If there is no compass the P3 wouldn't start rotating uncontrollably. The P3 doesn't use the compass to maintain heading. To do that it uses the IMU, in particular the gyroscopes. The flight controller uses the compass to set the initial value of yaw, but after that it uses the compass only to correct the very small errors that can occur. Without a compass any rotations would be measured in minutes.

 

[FrequentFlyer]

If there is no compass the P3 wouldn't start rotating uncontrollably. The P3 doesn't use the compass to maintain heading. To do that it uses the IMU, in particular the gyroscopes. The flight controller uses the compass to set the initial value of yaw, but after that it uses the compass only to correct the very small errors that can occur. Without a compass any rotations would be measured in minutes.

Thanks Bud for setting us all straight on that matter. Trying to understand the complexities of the flight controller and all the sensors are quite perplexing at times, especially when we have to change our mindset on things we thought we understood.


Sent from my iPhone using PhantomPilots mobile app

 

[N017RW]

What has been set straight?

Does anyone know DJI's sensor fusion and filtering algorithms?

The MEMS sensors used in the IMU are subject to drift and vibration perturbations.
Not likely the highest quality either.

Anyone who remembers 'Operation Drunken Sparrow' knows how the compass aids flight.

I wish it to be sttled as well but it will take more than an anonymous post to do so.

 

[syotr]

The Phantom does not just start spinning without a compass. The NAZA controller is sold without a GPS module which includes the compass. It is easy to fly in ATTI mode without GPS or compass. The yaw does not drift.
The Phantom may not be programmed to return home without it, but certainly could be.

 

[WetDog]

Pulling the throttle stick backwards will result in an eventual landing.

Doing ANYTHING (or nothing at all) eventually results in a landing....

 

[jonebk12]

If a $50 toy quad without a compass or GPS can hover without spinning uncontrollably like a top, I think a Phantom can too. I tend to agree with syotr: the ingredients should theoretically be there to allow a Phantom to successfully return home without a compass. Whether DJI has implemented such a fail-safe/backup system is another story. I suspect we'll never know for sure unless we get to interview a DJI engineer.

 

[Fat City]

The Phantom's IMU has a chip that is both a three-axis accelerometer and a three-axis gyroscope, similar to what's commonly used in smartphones. One axis of the gyroscope could detect yaw and would used to stabilize the AC's orientation.

 

[BobMcKinlay]

the ingredients should theoretically be there to allow a Phantom to successfully return home without a compass.

Trust me, the ingredients are there. The compass is not needed for RTH. Without it the process is not efficient, but it will get you pretty **** close to home. Well within the automated RTH margin of error (isn't it 20 feet or yards)?

Don't listen to those in this thread that are screaming impossible without offering evidence. A GPS and the ability to manipulate the motors to achieve rotation in a commanded direction and duration are all that's needed. Bank on that.

The algorithm isn't much (any?) harder than --

1. Oh, ****. I just lost compass and I see that my pilot has configured me to wander back home.
2. First, let's make sure I'm at or above RTH altitude. If the pilot f'ed that up, oh well. Nothing anyone can do about it.
3. Let's see, I'm at coord 1:1 and home is at 100:100. That means I need to go what I'll call NE.
4. I'll fly for x seconds in my current direction which I'll assume is N for this exercise. Oh, now I'm at -1:3. Turns out that was sort of NW'ish.
5. Let me fiddle my motors so I'll rotate a bit clockwise. Maybe I'll be pointing NE home then.
6. Fly for x seconds and now I'm at 3:-1. I guess I spun a bit far. Let me rotate a smidge back counter clockwise.
7. Repeat... repeat... repeat.
8. I'm now within some margin of error from RTH. I'll start lowering myself and if I'm going to hit something, I'll trust the pilot to hand catch me. You know, just like GPS-aware RTH.

 

[N017RW]

We've seen may algorithms.
None demonstrate how compensate for drift or other ambient weather conditions.
Just goes to prove you need a compass to do this reliably.

 

[ianwood]

@N017RW is exactly right. The compass is used to determine heading. The gyro is involved too but the compass is the only sensor with absolute reference to orientation relative to the Earth.

When you're trying to go home, angular rate means nothing without an absolute reference. Gyros which means they're only good for absolute reference for about 1 second. Bottom line, anyone who thinks it's possible to go home on GPS but no compass, you've got a few big issues:

1. How do you know when the compass data is bad?
2. How do you factor wind into the difference between planned course and actual course?
3. How do you correct for gyro drift to determine your heading?

Here are some hints:

1. There's no absolute way to know. It's a guess.
   
2. You can't. You can correct for wind but as soon as it changes (which it does all the time), your correction is bad.
3. You can't. Your margin of error will grow every second.

If you want to navigate without a compass, you will need a much better IMU and an RTK ground reference signal. Expect it to be standard in the Phantom 6.

P.S. Let's not confuse "manual" with ATTI. The P3 and P4 do not have "manual".

The Phantom does not just start spinning without a compass.

It won't start spinning but it will experience yaw drift since it only has the gyro to rely on for angular rate.

 

[syotr]

@N017RW is exactly right. The compass is used to determine heading. The gyro is involved too but the compass is the only sensor with absolute reference to orientation relative to the Earth.

When you're trying to go home, angular rate means nothing without an absolute reference. Gyros which means they're only good for absolute reference for about 1 second. Bottom line, anyone who thinks it's possible to go home on GPS but no compass, you've got a few big issues:

1. How do you know when the compass data is bad?
2. How do you factor wind into the difference between planned course and actual course?
3. How do you correct for gyro drift to determine your heading?

1. The Phantom already reports compass error when it detects bad data, (out of range or rapidly changing values.)
2. You still have GPS to hold position. You correct for wind as you go. The trip may be a curve or even a spiral but your can get there unless the wind is faster than the bird can fly in.
3.You never even need to yaw the craft. Fly by pitch and roll only as if the craft was in Homelock mode. If there is yaw drift, it will be very small and you can correct for it as you go.

 

[syotr]

Here is another way to look at the problem. Lets say you discover your compass heading is not shown correctly on the app. You can see the Phantom's position on the screen and you can see where you launched from (for the sake of argument say it is North of you.) The app is showing your bird facing North but when you pull the right stick back, instead of moving south, you see it has moved west. So now you move the right stick to the right and see that the bird is coming towards you. After a bit, you see that it is drifting a bit to the east so you pull back a bit on the stick as well as continuing to push it to the right. Continue this process until you get home.
Anyone can see that this is easy to do as long as you can see where you are and where you want to go. It would be even easier for a computer to do this. There is no need to yaw the craft, just translate your position and recalculate as you go.
Just because DJI has not done something, does not mean it is impossible. Certainly it is easier with a compass, but it is absolutely not necessary.

 

[syotr]

About that yaw drift. Sorry for the tilted camera. I had the camera mounted on my cap. I guess I am not very level headed or I've been walking around these mountains so long, I have one leg shorter than the other.

To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.

Accept third party cookies