Mapping speed vs. travel speed

The case where you want to go slower than what Map Pilot Pro calculates based on your overlap settings is the only time this would be a problem. Almost no one wants to do that. Usually everyone wants us to magically make it faster...

The manual setting is the maximum flight speed. Usually it is higher than the speed it will fly while mapping so it increases the speed on the approach and return legs. 

We will look at it but we really don't want to add another control for a case like this.

I tried to explain the motivation, but I'll try again.  The app will do what the documentation says. That is, it will calculate the best speed based upon agl and overlap.  And, if selected, it will slow the speed down depending on lighting as it flies to the first waypoint.  Then, it's up to luck at that point.  If conditions stay exactly the same, (or better) as when the craft flies to the first waypoint, then there is a good chance that the images will turn out sharp.

If, however, there is varying overcast during the mission (almost always here in the PNW), then the mapping speed can easily end up too fast.  So, in order to reliably end up with sharp images in these conditions, I lower the flight speed manually to ensure the image set is sharp.  When I lower the flight speed manually, the app also lowers the travel speed based upon the manual mapping speed.  That is poor logic.

I understand everyone wants to have the craft fly as quickly as possible to maximize battery.  But if you end up with some percentage of images with marginal blur, the mission can be of no value and must be flown again or live with poor results.  That is a worse scenario than using another pack to get a quality image set. 

If you don't want to add a separate speed setting, then how about a setting to reduce motion blur below 1x GSD rather than 2x GSD as is current.  This is a "Pro" tool now, correct?  :)

I get the motivation for it. I think your idea of having a setting for limiting the GSD is a good one that can easily be implemented. 

That would be cool.  I'm all for letting expressions do the work when possible. But sometimes we need different values for the expression.

Thanks Zane,

Hi Andrew.

You understand correctly.  The ability to slow the drone down during the mission portion without slowing down the transit speed. 

They did implement the ground smear setting but in a way that leaves you with no idea what the mission speed is going to be until it is actually flying the mission. 

For planning purposes, the app has the ability to input a camera (craft) type.  But no way to input an exposure value for planning purposes.

They must not do too much in-office mission planning at headquarters.  :)

Thanks for your +1.

 

Hi Dave, Just want to +1 this, if I understand it correctly.

I would like the drone to be able to fly as fast as possible between mapping/photo points. In my application, I am okay with 2x GSD motion blur, but when this requires the mapping speed to be very slow, I would still like the speed between photo points to be as fast as possible. Otherwise missions take way too long and too many batteries as it putt-putts along.

Currently I override the automatic speed adjustment or set the GSD to allow for more significant motion blur, just so I can get missions accomplished in a reasonable length of time over large properties that don't require great mapping precision.

Or maybe I misunderstand the point of your suggestion?

The flight path speed is what we consider to be a runtime variable. It depends on conditions on the site.

What is your goal in flying more slowly? Most would say it is to keep the smear from getting out of control. We have provided a mechanism to control that for varying levels of smear. It is deterministic in how much time it will take to execute the mission: no. There is no way to know how long that flight will take given that the programmed speed will be derived from sensed conditions. If you just want a slider to override the speed then you you are saying execution time is more important the the setting you picked. In this case you can turn it off or use a larger smear multiple. Basically, you either care about setting the smear or you care about setting the speed. You can't really have both. 

If you want to speed things up the best thing to do would be to take actions to shorten the exposure time as much as you can. Map during peak light hours. Don't use filters. Bump up the ISO or EV values. Etc. 

Given that you are generally collection data at some range of exposures, say 1/200 to 1/300 (hopefully), and you want your ground smear to be limited to 1X GSD you could pretty easily do some quick math and come up with the speed the system will select. Maybe we could add that range in there but this is really diving down the rabbit hole of features that add complexity to the app and VERY few users actually care about or will ever use. 

Do you have another suggestion?

Hi Zane,

What is your goal in flying more slowly? Most would say it is to keep the smear from getting out of control. We have provided a mechanism to control that for varying levels of smear. It is deterministic in how much time it will take to execute the mission: no. There is no way to know how long that flight will take given that the programmed speed will be derived from sensed conditions. 

Yes. The goal is to reduce smear to below GSD.  The original adjust for light conditions would mostly do that okay except in cases where the light diminishes during a mission after the app makes it's calculation or something. I would then see the smear warning blink into yellow/red on occasion.  The setting you added lets the user add a sort of buffer to ensure that smear stays below the GSD. Thank you.

In planning on larger sites, estimating the amount of mission time based upon the expected exposure settings allows me to maybe adjust the size of the mission scope or the height if I have latitude to do so in order to adjust the total flight time of the mission.  Without knowing what the flight speed might be makes it difficult to estimate mission flight times, especially on large sites.  Simply speeding up the mission speed is not the variable I mess with because that would make the entire goal of eliminating smear moot.  I know, based upon the time of year and short term forecast what exposure settings I'm likely going to need to use.  I do have a spreadsheet that I use to estimate what is the fastest mission speed I can use and not get smear.  But, I'm not so confident in my sheet to assume that my speeds are what the app is going to use given the new smear setting selected.  I would be happy if you shared the math that the app is going to use when it sees the shutter speed and flight height which you are calling "sensed conditions" and then I could just use that for planning.  I suppose after a few missions, I would get a sense of what the app is doing.  But you guys already know the formula that the app uses when it runs it's check on the height and shutter speed, so why should we have to guess for planning.  I'm glad for the "majority" of operators that don't have to operate with heavy overcast, deep shadows, and low sun angles where the default app setting are fine.  But some of us have more challenging environments.

The math is pretty easy. We measure the average exposure time over the first 70% of the approach leg (takeoff to green dot). We then take that value and compare it to the GSD at the currently programmed flight speed. If needed, we set the speed to slow things down to get the smear at the desired range (2X or whatever you choose). 

For example, we start out at 15 m/s flying a P4P at 60 meters and we determine the average exposure time to be 1/200s. That means that during one 5 millisecond exposure the aircraft would travel 7.5 cm (.005s * 15m/s). This is the ground smear value at the programmed speed.

At 60 meters the P4P has a GSD of 1.6 cm/px so 2X GSD would be 3.2 cm. To get the speed we need to go to limit the smear to 2X GSD we would divide the smear we want by the exposure time.

.032 m / .005s is 6.4 m/s. So, flying at 6.4 m/s would make it so with an exposure time of 1/200s a P4P flying at 60 meters would not collect data that is streaked more than 2X the GSD. 

The GSD at a given height for most aircraft is fixed. So the two inputs are exposure time and the GSD multiple.

You could use a manual exposure setting and know exactly what value the speed would get adjusted to. 

Thanks for the math, Zane.  I'll play around with it and come up with my planning solution.