Altitude errors in Maps Made Easy

Joined
Jan 27, 2016
Messages
2,409
Reaction score
893
Location
Ohio
Has anyone else using Maps Made Easy experience significant errors in the published flight altitude and 3D heights in the completed maps. Just ran two missions, one at 365 ft and one at around 230 ft and both show average height as 132 ft AGL! looking at the data, buildings that are >25 ft high are only showing as 10-12 ft. etc. Really screws up the volume estimates and DEMs. Didnt have this problem on mission that I flew about 3 weeks ago. And have made no changes to software or firmware. I've got a ticket into MME, but no feedback yet.
 
I've not used MME lately but the last time I did I didn't notice huge errors like that.

Keep us posted.
 
Just got this curious reply
"Richard,
Sorry to hear that you are not happy with your results. It appears that your flight pattern was oriented along the path of the prevailing winds which will adversely affect the quality of the outputs. Flying downwind you were likely going multiple mph faster than up wind and that discrepancy can cause big issues when it comes to the accuracy of heights.

If you are concerned with the relative accuracy of your models is important to use lots of overlap and collect all images at the same speed.

Thanks,
Zane"
Not sure why airspeed would have that effect. I'll try re-flying the mission cross wind as soon as the weather cooperates.
 
  • Like
Reactions: BigAl07
I've never even really considered wind direction a factor when setting up my flying grid.
 
I've never even really considered wind direction a factor when setting up my flying grid.
Neither have I and I guess Im still not convinced that was the problem. Snowing here now, but as soon as I can get some decent weather, I plan on flying some tests missions. I'll set them up to cover same area and at the same height. I'l make one of them co- and the other cross-wind for comparison. Figure I'll fly the bird up to altitude and turn on atti mode to see which way to blows to set up the mission. IF this really is something that has to be factored, MME needs to include some discussion about it in their tutorials. Im just glad that these were training missions and small enough to get processed for free. Hate to have spent points and have a customer waiting and then find out that this problem may exist. Really deceptive because the maps looked good. I was only when I started looking at the 3D and doing some volume measurements that the problem surfaced. Send by, I'll post more info as I get it.
 
Really deceptive because the maps looked good. I was only when I started looking at the 3D and doing some volume measurements that the problem surfaced. Send by, I'll post more info as I get it.

Do you shoot Obliques as well? Say 45degree camera angle?
 
No, I wasn't interested in developing full 3D, 2.5D was good enough for me to start. That works for things like volume estimates and slope and watershed study. IF/when a customer requests full 3D i'll run some - takes more points and I don't want to buy points while I'm still practicing.
 
  • Like
Reactions: BigAl07
MME app will take a photo when on the ground (at end of flight once systems all warmed up) from which it can calculate the above ground (from take off point) effective height, so you need to make sure you upload this if processing online with their cloud services - this was found on one of their support postings. They also have a tag fixer to correct images - see Maps Made Easy - DJI EXIF Tag Fixer. Please let me know how you get on.. I should add this didn't work with my P4P JPG images.
 
Last edited:
MME app will take a photo when on the ground (at end of flight once systems all warmed up) from which it can calculate the above ground (from take off point) effective height, so you need to make sure you upload this if processing online with their cloud services - this was found on one of their support postings. They also have a tag fixer to correct images - see Maps Made Easy - DJI EXIF Tag Fixer. Please let me know how you get on.. I should add this didn't work with my P4P JPG images.
Yes, I know about the extra photo required. They started doing that when DJI switched from recording the heigh AGL to GPS altitude in the exif files. I always include that pic in my uploads. Just did another set yesterday. First one parallel to the wind and the second cross wind. I figured which way the wind was blowing by climbing the bird to altitude manually and turning off GPS mode. Just watched which way it drifted. Flew both mission at an average height of 200 ft. Got the results back and for the one parallel to the wind, they reported an average height AGL of only 50 ft! Heights of structures and tress were similarly compressed. The one flown cross wind was better, reported height of 120 ft and heights in that map better, but still off considerably. I just finished uploading the same data sets to Drone Deploy. I want to compare their results with those from MME. I really want MME to work, the pay-as-you-go and 0 cost for small maps is really attractive. At least to start, I would be hard pressed to generate enough work to justify the $99/month for the Pro version of DD. Should have the results back tomorrow. I'll post an update when I get them
 
  • Like
Reactions: BigAl07
I'm surprised what they said about wind affecting height/accuracy seeming so much - have you tried reviewing the recorded flight data in more detail with healthydrones (or rather airdata now)? If the drone hasn't enough power or agility to maintain the correct height when it's very windy, then can understand if the height dropped, but sounds quite extreme behaviour to me.
 
I'm surprised what they said about wind affecting height/accuracy seeming so much - have you tried reviewing the recorded flight data in more detail with healthydrones (or rather airdata now)? If the drone hasn't enough power or agility to maintain the correct height when it's very windy, then can understand if the height dropped, but sounds quite extreme behaviour to me.
Haven't reviewed the flight logs, but watching the screen during the flight, it reported airspeed and altitude right at the programmed levels. 8-10 MPH winds aren't going to pull the bird down! And watching it, it flew nice straight lines right where it was supposed to! Just sent a query to MME about this latest set of data (Co- and Cross- wind sets) to see what they have to say now!
 
Latest reply - am told that the problem may be in the amount of overlap (used the default of 75/75%). I missed in the previous message that they suggested that I needed 80/80% or higher. New set of mission planned with overlaps of 80, 85 and 90% in both directions. hope to fly them this weekend.
 
That's a LOT of overlap.

Can you fly the exact same mission again (in similar wind conditions) with the increased overlap?
 
I agree, 75% seemed to be plenty, but not according the the response that I got from MME. Difficult to believe that going from 75 to their recommended 80% is going to 'fix' the issue. But,until I get the data, I have nothing to base that on. Repeating the same mission isn't difficult, but can't repel ate wind conditions. So I will probably fly 4 missions - 75, 80, 85 and 90% for completeness.
Also, loaded the same data into Drone Deploy with similar results! I've asked both if there may be some other setting (other than overlap) that I don't have set correctly. Neither have suggested anything. Standby for further updates
 
  • Like
Reactions: BigAl07
Well, weather cooperated this afternoon and I was able to fly my 3 missions. 90% overlap is still processing, but I got similar results for the 80 & 85% flight. Building is still 6 feet tall and my SUV only 3 ft tall! And it tells me I was flying at 82 ft agl (actual alt was 200). Of course, all volume measurements are way off too. Something isn't right and I'm not getting any positive help from MME.
 
Every time I 'correct' my missions in response to what MME is telling me a still get poor results, they come up with a new set of reasons why the elevation data is way off - here is their latest response "It might just be too small of an area for it to turn out properly. It looks like you are doing everything properly. There are limits to what can be done with the trial processing size. Absent us reviewing your log files, the only thing left to try is to slow down the maximum speed of the flight." I've sent them my logs and asked them to recommend what a mission should look like to map an area like what I have been doing in order to get accurate elevation info. Waiting for a response.
 
Does anyone have a MME map in which they got good elevation results? Would you mind sharing the info about it - such as mission parameters like speed, altitude, overlap, etc. And any numbers on the relative accuracy of the elevation data.
 
Just a thought - have you tried pumping the data through another processing product like Pix4d or Agisoft and looking at accuracy reports? One other option would be to pull off the flight logs and geotag photos again (elevation data only) if that's an option for you?
 
  • Like
Reactions: BigAl07
I have, read post 14. I've read the logs and the exif data on the photos. All the data is good and the geolocation data on the map is good too. I even checked a top map to confirm the local elevation and the exif altitudes are within 1-2 meters of the top.
 
Ok, at this point, I'm ready to give up on Maps made Easy! Just ran another mission following all of the suggestion/recommendation that MME had given me in response to my questions about this problem. Elevation data is still worthless - showing a height AGL of 123 ft when 200 ft was the mission altitude, 5-6 foot buildings, 4 ft tall SUV, etc. The maps that I am getting are totally useless for analyzing things like volume or slope. Has anyone else come across this type of issue with MME? had anyone else gotten good elevation maps from them?
 
  • Like
Reactions: BigAl07

Members online

Forum statistics

Threads
143,066
Messages
1,467,354
Members
104,933
Latest member
mactechnic