Why cannot a live estimate of wind speed and direction be displayed?

[dirkclod]

Too many first time flyers with something that has more range for the weight for inexperience flyers.
Imo not a good drone to learn to fly with.

 

[old man mavic]

there is a fine line between having lots of flight info to look at at any given time and keeping VLOS on the drone and the more experience you have when flying your drone then it is easier to switch between the screen and the aircraft the MM app at the moment is an attempt to simplify the screen to some extent for new fliers

 

[R2TheTroll]

I seems that the drones have good groundspeed data, but no airspeed input. They don’t have a pitot tube. So any wind data is an estimate.

 

[sar104]

I seems that the drones have good groundspeed data, but no airspeed input. They don’t have a pitot tube. So any wind data is an estimate.

Not entirely true, because for a quad in flight its tilt is a direct function of airspeed and direction.

 

[SqFt]

As mentioned, you need an airspeed input to accurately compute wind direction and speed. Using tilt would be nearly useless. If you used the pitch and roll angle it would only be useful under stable conditions, like holding a hover, or moving in a steady direction and speed. And even then, this method would require some kind of calibration equating pitch/roll with a lateral vector. Not an accurate method. These quads are constantly twitching and correcting and accelerating/decelerating and that would make this type of calculation useless. A pitot type airspeed probe is simple in concept, but difficult to execute properly. So unfortunately, this feature is not likely to happen soon. I would love it if it did tho.

 

[sar104]

As mentioned, you need an airspeed input to accurately compute wind direction and speed. Using tilt would be nearly useless. If you used the pitch and roll angle it would only be useful under stable conditions, like holding a hover, or moving in a steady direction and speed. And even then, this method would require some kind of calibration equating pitch/roll with a lateral vector. Not an accurate method. These quads are constantly twitching and correcting and accelerating/decelerating and that would make this type of calculation useless. A pitot type airspeed probe is simple in concept, but difficult to execute properly. So unfortunately, this feature is not likely to happen soon. I would love it if it did tho.

You are correct that it is best computed in stable flight or hover, but those conditions are common during flight, and it's quite accurate. The calibration is trivial, since it is easy for each model to measure speed as a function of tilt in the absence of wind.

 

[dronelapaz]

calculating windspeed is very simple:
SOG with the wind minus(-) SOG against the wind divided by 2 = true windspeed

 

[sar104]

calculating windspeed is very simple:
SOG with the wind minus(-) SOG against the wind divided by 2 = true windspeed

That's fine to make a deliberate measurement, but it doesn't give you wind speed continuously in real time during a flight. And you need to know wind direction for that method, which may not be the same as on the ground.

 

[surfbird]

From reading a lot of threads on here, it would appear that flying in excessive wind is a major cause of drone losses.

Why cannot a live estimate of wind speed and direction be displayed on the controller/app? It would allow you to fly up to a height and make a much better assessment of actual conditions.

Surely if a drone has a good GPS fix, it will know it's "ground speed" and from that can calculate the what the wind is doing from the remaining vectors? I accept it wouldn't be perfect but it would be a very decent indication.

The fact that there seems to be 2 levels of warning, "fly with caution" and "land immediately" on the GO4 app, indicates that the calculation is being made.

I am not an experienced drone pilot and have never flown a "real" aircraft but I have been a professional mariner for 30 years and it seems like a no-brainer to me.

Am I missing something?

I have done that in my own app, just for testing purposes. No changes in firmware is needed. Pretty accurate.

The sdk gives you ground speed x,y and attitude x,y.
For it to be accurate you have to consider:
* Kind of model, since speed differs for the same attitude angles.
* True windspeed direction , which also differs between models
AC need more tilt to go sideways that forward for the same speed in no wind, since it shows bigger are against the wind.
* calibrate of x,y attitude when hovering in no wind is needed, since it always is some degrees off.
* Take the mean value over couple of seconds.

I also tried to calculate W/km/h. Could be used for similar things.

All numbers needed are in the SDK. I don't think DJI just bothered to do it. Might be confusing for some users. Hard to keep track of all models is maybe another. The high wind warning today seems to be little random at least.

I think there are a lot of functions that could be added to the app. I wish I had more time to do it myself

 

[Maviac]

the drone uses the data from the processor to estimate the amount of force it requires to try and keep it in one place, if it is struggling to hold position, it will put up a wind warning in the app, it has no way of knowing exactly what the speed of the wind is, just the effect it is having on its ability to fly those parameters will be preset by the firmware

Actually, if I’m not mistaken, it can use that data to get a pretty accurate estimate of wind speed. It may not be exact but it should be pretty close. AirData can do it, and it can’t be a very complex task. The math required is not quantum physics. It would be a welcome addition.

edit: sorry, I see my post is repetitive and others have confirmed this.