tower surveying, solo speed and picture rate

[ivegotahaddock]

sorry if this has been asked before, but I cannot seem to find this information.

I'm new to tower but not to solo. when planning a survey mission, tower seems to calculate in the background the speed of the drone and how far between pictures that it triggers. however, I cannot seem to find where those factors are, or how to adjust them if I want. I can do a series of calculations based on the length of the mission and number of pictures to be taken, but I would prefer if I could easily see how often pictures are to be taken (i.e. 1 picture every 2 seconds) and the speed of the drone. this is in part because I am using a MAPIR camera on the accessory bay and it is not triggered by tower.

is it possible to specify the time between pictures and/or the speed of the drone, or am I stuck doing a calculation and then guessing what the correct shutter time is for the MAPIR? I've run a couple of scenarios on the same survey using different heights and different overlaps, and the drone speed vs. shutter speed that tower settles on seems to be rather random.

any help is greatly appreciated.

 

[Jubalr]

sorry if this has been asked before, but I cannot seem to find this information.

I'm new to tower but not to solo. when planning a survey mission, tower seems to calculate in the background the speed of the drone and how far between pictures that it triggers. however, I cannot seem to find where those factors are, or how to adjust them if I want. I can do a series of calculations based on the length of the mission and number of pictures to be taken, but I would prefer if I could easily see how often pictures are to be taken (i.e. 1 picture every 2 seconds) and the speed of the drone. this is in part because I am using a MAPIR camera on the accessory bay and it is not triggered by tower.

is it possible to specify the time between pictures and/or the speed of the drone, or am I stuck doing a calculation and then guessing what the correct shutter time is for the MAPIR? I've run a couple of scenarios on the same survey using different heights and different overlaps, and the drone speed vs. shutter speed that tower settles on seems to be rather random.

any help is greatly appreciated.

Tower is not triggering the Mapir camera. They have come up with a PWM trigger for it, but I don't believe it is available for the Solo yet. Using the Mapir camera, you want to set it to take pictures about every 2 seconds and be sure you are using a fast SD card. They have some recommendations on their web site of tested cards. If you are using the GP4 BK, Tower will trigger the GoPro and you can see that setting after you create the survey in Tower. After the survey is created and before uploading, you will see a WP at the bottom that is the camera trigger. If you tap on that WP, it will bring up an options screen. I believe the default is to trigger around every 50-60', and that is what you can change. Remember though that you can't trigger any faster than the camera is capable of.

Using the Mapir, I set the Solo to fly at about 15mph. That is where I have found I get the best results. Easiest way to set that is to first put a WP before you want the survey to start. Then set that WP as a 'speed change'. The slider you see will be in MPH. Then draw your survey area.

Good Luck..

 

[ChrisBird]

sorry if this has been asked before, but I cannot seem to find this information.

I'm new to tower but not to solo. when planning a survey mission, tower seems to calculate in the background the speed of the drone and how far between pictures that it triggers. however, I cannot seem to find where those factors are, or how to adjust them if I want. I can do a series of calculations based on the length of the mission and number of pictures to be taken, but I would prefer if I could easily see how often pictures are to be taken (i.e. 1 picture every 2 seconds) and the speed of the drone. this is in part because I am using a MAPIR camera on the accessory bay and it is not triggered by tower.

is it possible to specify the time between pictures and/or the speed of the drone, or am I stuck doing a calculation and then guessing what the correct shutter time is for the MAPIR? I've run a couple of scenarios on the same survey using different heights and different overlaps, and the drone speed vs. shutter speed that tower settles on seems to be rather random.

any help is greatly appreciated.

Hi,

Your speed and overlap (sidelap wont but affects total flight time as it dictates the number of legs flown) affect the time / distance between shots. Use MP survey and it says the time between shots that you can then use. The solo with gp cam tho from mem triggers on distance. I'll fire it up later today to check.

Chris

Sent from my SM-G930F using Tapatalk

 

[ivegotahaddock]

thanks guys, good information and I appreciate the help. jubair: so, a speed-change waypoint before the survey will affect how tower calculates the survey? that's good to know, and makes this easier. I also typically try to use around 15 mph for surveying, but hadn't figured out how to get tower to use that as a starting point.

I'll give this a try in a day or so and let you know if I hit any hitches. thanks!

 

[Dronie]

Hey guys, are any of you having any issues calibrating your Mapir images right now with the Fiji plugin? If not, what version of the Fiji plugin are you using as well as Java? Also, if you're using different methods for calibrating the images please share.

Thanks,

 

[ivegotahaddock]

drone:
i've not yet used the fiji plugin for the mapir images, so unfortunately i cannot help you.

 

[Dronie]

What are you currently using to calibrate the images from the Mapir camer then?

 

[ivegotahaddock]

at this point I have not calibrated them. I've got a visible light model and have been using the images for point cloud generation. so far, I've just used them straight out of the camera.

 

[Dronie]

at this point I have not calibrated them. I've got a visible light model and have been using the images for point cloud generation. so far, I've just used them straight out of the camera.

On another note, are you finding that the Mapir RGB visible camera generates better point clouds than the a GoPro? just curious if you have done any comparisons because I use the GoPro Hero4 for my orthos because I get a better deliverable.

 

[ivegotahaddock]

I'm still evaluating the difference at this point. I've only flown one mission so far with the MAPIR camera, and the narrower field of view, along with the fact that it's mounted sideways on the solo, resulted in images that did not have enough overlap to complete a decent point cloud. I usually get pretty good results with the GP hero4 i use, but am hoping that the higher resolution of the MAPIR will give me a bit crisper results for the structures I'm trying to fly.

hoping to get back out soon to retest. I've modified the MAPIR mount so it is oriented with the top of the picture towards the front of the drone, and will be more careful with the amount of side overlap i fly. fingers crossed...

 

[Jubalr]

I'm still evaluating the difference at this point. I've only flown one mission so far with the MAPIR camera, and the narrower field of view, along with the fact that it's mounted sideways on the solo, resulted in images that did not have enough overlap to complete a decent point cloud. I usually get pretty good results with the GP hero4 i use, but am hoping that the higher resolution of the MAPIR will give me a bit crisper results for the structures I'm trying to fly.

hoping to get back out soon to retest. I've modified the MAPIR mount so it is oriented with the top of the picture towards the front of the drone, and will be more careful with the amount of side overlap i fly. fingers crossed...

I use the Mapir in the original 3DR static mount. Images turned out great and stitched fine. Here is a raw Ortho from the Mapir of 80 acres of soybeans. This is the flight I posted about earlier where the Solo flew for 21 minutes to get 80 acres.

 

[Dronie]

Nice, are you taking it a step farther, calibrating the images and calculating an NDVI map?

 

[Jubalr]

Nice, are you taking it a step farther, calibrating the images and calculating an NDVI map?

Yes, Here is the NDVI rendered down from the 500mb .tif to a small jpg..

 

[Dronie]

Awesome dude! very nice

 

[ivegotahaddock]

yeah, that is nice. I thought about using the static mount for the MAPIR, but at this point I've been using both the MAPIR and the GoPro at the same time, as backup.

how in the world did you get 21 minutes out of a solo? multiple flights? so far the best I can ever get is about 12-13 minutes on any single battery.

 

[Dronie]

I also used the MAPIR camera in the static mount and it worked fine, however when I added an additional MAPIR camera (now two) their dual camera mount worked better because it distributed the weight on the solo to the CG. I took the static mount off and no gimbal just the dual mount and the thing flies solid. On a good day, no wind, low humidity, moderate speed, I could see him getting close to 19,20,21 mins. This is of course taking the battery almost down to nill before landing which I would not recommend.

 

[Jubalr]

I don't recommend pushing the battery that much either, though it is still doing fine. Without the gimbal weight and power draw, you will certainly gain flight time. Here is the thread on that flight..

 

[Dronie]

I'm still evaluating the difference at this point. I've only flown one mission so far with the MAPIR camera, and the narrower field of view, along with the fact that it's mounted sideways on the solo, resulted in images that did not have enough overlap to complete a decent point cloud. I usually get pretty good results with the GP hero4 i use, but am hoping that the higher resolution of the MAPIR will give me a bit crisper results for the structures I'm trying to fly.

hoping to get back out soon to retest. I've modified the MAPIR mount so it is oriented with the top of the picture towards the front of the drone, and will be more careful with the amount of side overlap i fly. fingers crossed...

Hey do you think if you flip-flopped your overlap settings it would help? For example, 75 side and 85 frontal since the front is basically the side of the camera with this mount? Just a thought.

 

[ChrisBird]

Hey do you think if you flip-flopped your overlap settings it would help? For example, 75 side and 85 frontal since the front is basically the side of the camera with this mount? Just a thought.

Hi Dronie,

I dont think that will work. What I reckon is needed to do is add the mapir camera as per these instructions:
Adding the MAPIR Camera to Mission Planner and Tower

But I think he should swap the sensor width and height settings. I think that will do what he's are after.

Reason why swapping the overlap and sidelap is that it will make it better but not in the exact way that tower is trying to do it. For example the overlap setting is telling Arducopter how far it has to travel in a distance measurement before taking a shot. The sidelap affects how many tracks are done. So if you swap the two figures in towers interface your still not getting the actual overlap requested, if he swaps the two sensor values then it should work more precisely.

I havent tried it but would be good he tested it out and lets us know if it works as expected.

Chris

 

[Dronie]

I have added the Mapir camera to the Tower app per the instructions on their website.. which brings me to another question. Do you think these settings are taking into account the cameras are mounted sideways as mentioned above or will modifying the rotation of the mount really work as mentioned above..?