New - Wind reports!

[Eran Steiner]

Wind reports are now live!

Here is an example:




As always, there is no need to re-upload any of the logs.

Cheers,
Eran

[Eran Steiner]

Please note that there may be an issue with Litchi beta 2.4.x reporting incorrect high values - this is being investigated. The official Litchi version (2.2.2) should work correctly.

[Tomas Lesik]

WOW! I love this...  You now make me wish DJI would put this information on the GO app screen. I would even pay for it as an in-app purchase...

Anyone from DJI in this group???  Take note...

[Eran Steiner]

Thanks, I'm glad you like it! I am not sure if there is anyone from DJI in this forum... your best bet may be trying a new thread on the DJI forum.

The wind issue with Litchi beta 2.4.x is now resolved. There is a bug in this beta build and the Litchi team will fix it in the next version. In the meantime the data is fixed on the site so all should be good now. Please let me know if you see any issues.

Thanks!

[JK JK]

Great, nice work.

BTW, can add rolling function to the log list? In case of many logs, the list is very long and need to roll up/down whole page. Also button First/ 

Last , Prev/Next are wellcomed.

[Jim Wilkinson]

Fantastic Eran! The wind speeds appear to be spot on! Thank you.

[Geoff Drysdale]

WOW this is awesome.. How do you do it?

[Colin Bramall]

Earn this fantastic and a really useful addition to our understanding of how we are flying the drones. I do wonder whether the reported wind speeds are a little higher than they actually were. On 26th September my flight at 7.26am was done in very still conditions, as you can see by the low lying mist. However the reported speeds including a gust of 18 mph seem excessive

[Eran Steiner]

Thanks.

I just checked your flight (in the UK, right?) - I see that the strong gusts are at 400 feet. Please note that it is common for wind to be different at altitude, so it could be perfectly calm at ground level but a lot more windy at 400 feet.
One thing you could try next time: when taking off or landing, let the Phantom hover for 5-10 seconds every 50-100 feet. This way you could see progress of wind over altitude change.

[Alpha]

Wind is calculated using the angle of the aircraft, the speed of the aircraft and the status of the remote control knobs.

Does is detect exp values correctly? The remote control knobs will have different speed at certain positions depending on which exp values 0,2-0,8 are set.

[Kevin Watkins]

This is ace, of any log reports this shows me things that really matters. 

I'm not sure this is the right forum, but I've been looking at the JNI and interfaces, now jumping into the native libraries (and UDT). Basically, I'm trying to figure out if the controller supports turning off all the lights. At times there isn't a need for it, and it's just a drain on the battery. If during litchi dev they figured out the controller cmd's that would be great, otherwise I'll continue jumping into the fray...

On Saturday, October 10, 2015 at 1:51:54 PM UTC-7, Eran Steiner wrote:

[Eran Steiner]

Yes, it will only calculate wind speed at either maximum pitch or hover in place - and the exp value would not matter at this point because it uses the pitch to determine the expected speed.

[Eran Steiner]

Thanks.
Your best bet would probably be to post this question on the Litchi forum.

[Alpha]

Does it calculate correctly including that at maximum pitch speed of the Inspire1 is

18 m/s in Atti

15 m/s in GPS

+ 2/ms for some seconds if pitch forward+up are put simultaneousely

before the firmware 1.3 speed downgrade happened the values were 22 m/s Atti und 18 m/s GPS

or does this not matter at all? I just thought to calculate the wind it must also know the maximum speed to be expected at maximum pitch or did I understand it wrong?

[Alpha]

Have to add that even this is not 100% correct because the maximum pitch speed IS affected by expo settings in GPS its 15 m/s at setting 0,2 and 18 m/s at setting 0,8

[Colin Bramall]

Eran thank you. I remain sceptical about the wind speeds show. I put the drone up for a few minutes at 655pm on 11 October and there was no wind at all, yet the readings show 5.8 and 8.7 mph and the p3a stayed close to the ground. Am I just insensitive to a breeze or is there a small bias? I am sorry to be a little persistent on this point because I think you've developed a great tool.
Best wishes
Colin

[Eran Steiner]

Yes, it's calculating based on the pitch angle, which will be higher at a higher maximum speed.

[Eran Steiner]

What do you mean by "p3a stayed close to the ground" - how close? Keep in mind that 5-8 mph is just a slight breeze (average walking speed is 3.1 mph) so 8mph is not that much.

Otherwise, you may want to calibrate the IMUs on your aircraft - as out-of-calibration IMUs will cause that. Also, in a perfectly calm day, try to hover 6 or so feet off the ground to eliminate the guessing. Think of a helium balloon being released - most chances it will never go straight up even in a calm day. Winds on the ground are typically different than even 50 feet up. Last, keep in mind that this is not exact science and its limited to what the aircraft reports. If you have a bad propeller, for example, your aircraft will shift as well, although in this case the wind will look like its come from a different direction on the map, depending on the aircraft heading.

[Frank Adams]

I'm not with dji but I am a dji product tester. From me experiences with dji, they would not incorporate someone else work into their's. They are known to hire individuals or buy the rights of some things though. Example, the video's you see from time to time beginning and ending with dji info, have been paid for by dji. This is one of if not the main reason you see a buy option with your video's and pic's on skypixel.com. They also encourage other companies to look around on skypixel.com to possible find something that said company may be working on. If you upload any of your video's on skypixel you should consider selecting the option to sell them if you are not that attached to them.

I have 2 questions about wind speed. I was curious how you come to the Data that is displayed? I use the GO app and have one of the Data Readouts showing I was flying in 35.1mph wind. And I know for a fact that I was flying in exactly 1/2 of that wind speed.

The other wind speed question is... Do you include any wind speed in the formula you use for the battery readings? I'm asking that only because such stuff as wind speed. horizontal and vertical speed, and height of coarse can play a role into figuring out the overall health of the battery during a tested flight.

-Frank  

[Frank Adams]

I forgot to mention....You have done a excellent job thus far here on healthydrones.

-Frank

[Eran Steiner]

On Tuesday, October 13, 2015 at 1:04:21 PM UTC-7, Frank Adams wrote:

I'm not with dji but I am a dji product tester. From me experiences with dji, they would not incorporate someone else work into their's. They are known to hire individuals or buy the rights of some things though. Example, the video's you see from time to time beginning and ending with dji info, have been paid for by dji. This is one of if not the main reason you see a buy option with your video's and pic's on skypixel.com. They also encourage other companies to look around on skypixel.com to possible find something that said company may be working on. If you upload any of your video's on skypixel you should consider selecting the option to sell them if you are not that attached to them.

I have 2 questions about wind speed. I was curious how you come to the Data that is displayed? I use the GO app and have one of the Data Readouts showing I was flying in 35.1mph wind. And I know for a fact that I was flying in exactly 1/2 of that wind speed.

How do you know for a fact that it was 1/2 that speed? please keep in mind that it reports the strongest gust over a period 30 seconds. So it could be that on average it was half, but for a few seconds there was a stronger gust.
 

The other wind speed question is... Do you include any wind speed in the formula you use for the battery readings? I'm asking that only because such stuff as wind speed. horizontal and vertical speed, and height of coarse can play a role into figuring out the overall health of the battery during a tested flight.

No - battery reading is not using the wind speed, but that's an interesting idea for an advanced report.

Thanks,
Eran

[Eran Steiner]

Thank you! :-)

[Nathan Hoover]

Looks great. I looked at some of my windier flights. It says it was a 36.6mph wind where I flew 65+ mph - seems about right to me (although that was in ATTI mode and you say it doesn't support ATTI). I also checked some more normal flights - one had 2.0 mpg average wind speed which is right for that day. Great stuff!!

[Tracy Garrison]

Eran great!  I was waiting for your Wind button to come out of construction because one of my first flights the wind was around 14 MPH at the ground.  I took off with prop guards on and when I got out about 1000 feet and 110 feet up I was hovering to get ready to take a photo and all the sudden my bird started heading a different direction.  Not sure what was going on I tried to make subtle corrections then tried giving it full speed.  I saw the prop guards dip way into view fighting the wind on the way back.  That was my second flight with prop guards and my last flight with them.  I wonder if you could check something out.  You show the gust at the exact time I was having the problem but you are showing the arrow in the wrong position on the Wind Map portion.  You show the 25 mph gust at 2:16 into the flight but you have it labeled as B (a 17mph) gust at 3:06 on my way back in.  Both these spots I remember having fits with and they are spot on but I think they are reverse labeled.  Flight was on Aug 21, at 2:47 Pm if you can check it.   Arrows are even pointed in direction it was being pushed and when I watch the flight on my device I can see my small tries at getting it to move forward until I give it all to get it to move back toward me.   I think you have these wind speed pretty close!

[Taylor B]

Hi Eran, I am very excited about this feature, as my Phantom primarily films kiteboarding and hydrofoil racing.  A note about degrees.  You are not displaying the proper values, or the log files are not being parsed properly...I am unsure which, as I am not a data scientist... please review the following.  The answer should become obvious: http://coolclassroom.org/pdfs/understandingwinddirection.pdf

On Saturday, October 10, 2015 at 1:51:54 PM UTC-7, Eran Steiner wrote:

[Eran Steiner]

Great!
If you switched to GPS during the flight and hovered (or flew straight) - it should use that to calculate the wind at that point. Or are you saying it calculated the wind at sections where you were only in ATTI mode?

[Eran Steiner]

Thanks Tracy. Can you please email me directly the flight link of this flight? (the URL from the address bar as you view the flight). I will take a look.

[Eran Steiner]

Thanks for the link Taylor. The site is using the same system - 0 for north, 90 for east, 180 south, 270 for west. Can you please clarify what part is not correct on the site?

Thanks,
Eran

[Frank Adams]

<<How do you know for a fact that it was 1/2 that speed? please keep in mind that it reports the strongest gust over a period 30 seconds. So it could be that on average it was half, but for a few seconds there was a stronger gust.>>

Reading flight logs and high speed flights are my actual hobby with my P3P. I have a decent weather station setup over the internet that actually sends data to our local tv news station. It has a very nice anemometer. Of coarse I can only use it for monitoring wind speed at surface level. I know what I need to know about wind speed at different altitudes as well as wind & jet stream's. If you look at the flight I referred to it was at higher then usual, or at least for those doing high speed flights like me. I don't remember exactly, but it was likely around 350ft where you deal more with constant wind. Just prior to my fastest flight, I measured the the overall wind speed for over a 1/4 mile and came up with between 17-18mph. That is how and why I knew what the air speed was for that day, as well as the wind gust. The numbers displayed by healthydrones was exactly double of what took place that day. Btw, believe me I wish I was wrong and would have had the wind speed the site showed that I did. With 35.1mph wind gust I would have flown over 75+mph.

<<No - battery reading is not using the wind speed, but that's an interesting idea for an advanced report>>

It would probably be more trouble then what it's worth. Too many variables would be attached to that type of info.

[Eran Steiner]

Thanks for the details, Frank. I don't think you sent out a specific flight information though. Could you please share so I can take a look?

Thanks.

--
Eran Steiner
HealthyDrones.com - Innovative flight data analysis that matters
Follow us on Twitter  /  Like us on Facebook

[Clive Flint]

Hi Colin, have you tried it in atti mode when the wind is calm? I'm always surprised how much my drone moves when it's not keeping a gps lock.

[Clive Flint]

This is a great addition to the site. Nice one!

[JK JK]

Today I fly. I let my bird for few seconds in ATTI mode, to see see where the wind blows. I can again  confirm the wind direction is detected correctly, unfortunately I forgot to  check horizontal speed on the screen of my phone. Next time.

Dne sobota 17. října 2015 8:15:42 UTC+2 Clive Flint napsal(a):

[Alpha]

I got this in the log file of a PC SIMULATOR flight - there is no wind setup!

Flight timeAltitudeHome DistanceWind DirectionWind SpeedA02m 44s562.1 m      66 m123°  0.9 km/hB04m 09s571.3 m      125 m313°  1.3 km/hC07m 53s571.3 m      338 m243°  19.5 km/hD08m 09s571.3 m      185 m242°  19.0 km/h

[Eran Steiner]

I can't really tell much from this, as the copy/paste in the forum doesn't work well. For the future - it's best to put a shared link to the flight. You could share only the wind page if you wanted to - check the Share button at the top right of the flight.

Either way, I am not really familiar with the simulator or do any testing with it. I know other simulator flights on HD show 0 or very close to 0 wind, so something in this particular setup is off. For example, if the simulator thinks the airplane is constantly in a pitch or roll then this could happen. Again, I am not familiar with the simulator so can't talk much about it, but perhaps the aircraft is sitting in an angle when you use the simulator? or perhaps the IMUs are not calibrated properly.

Eran

[Taylor B]

Sorry for the late reply, the system is backwards.  Wind which blows from the West to the East is said to be blowing at 270 degrees.  I flew today at a beach which ranges in direction between 210-240 and my logs all state the direction is ~60 degrees.  Does this make sense?  My attachment should help illustrate.  

[Eran Steiner]

Thanks Taylor for the clarification. I think I better understand now - you are basically saying that the over angle is correct, but the representation in the degrees is 180' flipped. I think you have a good point there - but I'll have to double check before I change the site.

Thanks again and cheers,
Eran

[Taylor B]

Correct.  Still showing that way as of today but I'm sure you know that.  No complaints, just trying to help.  Love this project.  Keep it up.

WeatherFlow (windalert, ikitesurf, iwindsurf, sailflow, etc) is an excellent source of data pertaining to the wind.