Site mapping with 6" accuracy

[Darcy]

I'm new to this forum, and I'm posting this in the hopes of getting some valuable feedback.

As a landscaping company, we’re wanting to start using a drone to measure sites for estimating purposes. We’d like to maintain an accuracy of at least 6”. I’ve heard someone suggest laying an object of a known length on the site and make sure it shows up in the imagery to use as a reference. Is this a reliable solution? If not, what solution(s) do people recommend?

We’re not looking for sub-inch accuracy, so I see no need to invest $5-10k+ on RTK/PPK equipment unless there really is no step between this and 1-3ft accuracy which you might get with a good onboard GPS.

Together with this, I’d like to hear some suggestions for models of drones to consider for this purpose. We currently use a Yuneec Mantis Q for taking overview pictures of sites, but I’m thinking this unit may not be suitable for accurate mapping due to limited flight planning features, sub-par photo quality, etc. I understand the DJI Phantom 4 Pro is often considered the gold-standard for drone mapping, but are there any other more economical options?

 

[Dave Pitman]

Welcome Darcy.

A P4P is about the best drone for imagery meant for SFM processing when you don't want to mess with heavy lift multirotors carrying mirrorless high res cameras. This is due to the 20 Mp sensor and mechanical shutter lacking on lower priced options.

As far a what you want to do with it. If all you are concerned about is being able to reasonably measure on the resulting ortho image (relative accuracy), and don't need to tie into a local coordinate system like surveyors do (global accuracy). Then if you produce a good ortho, you don't need any other input. Laying out something on the ground that you can see in the images, like 3 points of a right triangle, for example, with known measurements will give you a great way to verify that your ortho is reasonably accurate, relative to itself.

Hopefully this makes sense!

 

[R.Perry]

I do mapping with an Inspire 2, and get within 4 to 8 inches both horizontally and vertically. That has bee verified by the surveyors on the job site. That is using Drone Deploy, but it's service is expensive.

 

[Darcy]

I do mapping with an Inspire 2, and get within 4 to 8 inches both horizontally and vertically. That has bee verified by the surveyors on the job site. That is using Drone Deploy, but it's service is expensive.

Are you using GCPs? If yes, what GPS equipment are you using to determine their location? If not using GCPs, are you getting that 4-8" accuracy with the drone's onboard GPS alone?

 

[R.Perry]

I'm using GCP supplied by the surveyors. I check them with the pads they have on site and verify the differences. a couple of the pads elevations have changed and the surveyors have had no need or reason or need to correct them.

 

[R Martin]

Are you using GCPs? If yes, what GPS equipment are you using to determine their location? If not using GCPs, are you getting that 4-8" accuracy with the drone's onboard GPS alone?

The least costly solution if you are going to use a GPS receiver would IMO be Aeropoints. An accurate GPS receiver with a tripod/bipod and antenna is going to set you back $10K+. I am not aware of any UAS without an RTK/PPK setup that gets 4-8" accuracy out of the box. Most onboard prosumer UAS GPS are good to around 3-5 feet of accuracy.

 

[Darcy]

I'm using GCP supplied by the surveyors. I check them with the pads they have on site and verify the differences. a couple of the pads elevations have changed and the surveyors have had no need or reason or need to correct them.

An important difference between our application and yours is that we'll be visiting a site rarely more than once, and they're not necessarily new construction sites that have recently surveyed GCPs. I'll have to research Aeropoints more to determine how that would work for us.

R.Perry, do you feel that DroneDeploy plays an important role in obtaining your 4-8" accuracy? If so, what aspect of the software makes the most difference? Like you said, it's a very expensive software so it would be great to find a more economical solution.

 

[R.Perry]

We rarely use Drone Deploy to it's full extent. Our client depends on the surveyors and the mapping is for them to look at and say gee that's neat. As for DD accuracy without GCP data, not so good. For instance you will get negative vertical elevation. When you do mapping they start with the elevation of your start point, so if you don't start from the lowest point on your project you end up with negative elevations, to me this isn't good, of course when you plug in the GCP data it comes out pretty close. I showed the superintendent one, it was off Horizonally 1' 9" and vertically 5.5 inches, his comment was that's a country mile. So they are paying a lot of money for maps that they use to monitor construction progress.
The company does have an RTK equipped drone, but I don't use it.
If I were you I would talk to DD their input, but like you, I think it is too expensive at minimum 4 grand a year you got to make a lot of maps.

 

[Kristina Fowler]

Maps Made Easy lets you use existing (readily visible; Example: Parking lot lines ) GoogleEarth GCP's to align with the corresponding GCP's in your recently captured images. They call it "Manual Referencing". I'm assuming that since you're a landscaper that your missions are relatively small and therefore free per the MME pricing. I think the maximum number of images allowed for free maps is 68. MME also has onboard measuring tools. Even if you have slightly more than 68 images, it can be safely said that it would amount to "chump change" to get them processed. Just a thought....

 

[Gr8Basin]

Darcy - for what you are looking for, it’s pretty tough to have confidence that your results would be regularly sub foot once processed, especially without GCP’s. The aeropoints are a good option, but in an ideal setting need to be out for at least an hour. If you’re strictly looking for a rough ortho model of the job site, your general flight sans GCP will suffice for measurements. You’re likely already adding in a percentage of extra anyway, but if you’re looking to reduce the percentage of extra, you really should consider the gps equipment And proper ground control.

You can find some very reasonable options to get into an RTK setup that could benefit your business by seeking out some used RTK setups if you really need to get sub foot. And if you are seeking sub foot, the RTK will get you better than that with the proper setup. You can also get setup with a sub-decimeter equipment as well.
The P4P is an excellent choice with its mechanical shutter. It’ll be more efficient in the flight process vs a rolling shutter setup. And for this type of application, check out Pix4D fields if the other mentioned processors don’t suffice.

 

[HAWK-i.us]

The least costly solution if you are going to use a GPS receiver would IMO be Aeropoints. An accurate GPS receiver with a tripod/bipod and antenna is going to set you back $10K+. I am not aware of any UAS without an RTK/PPK setup that gets 4-8" accuracy out of the box. Most onboard prosumer UAS GPS are good to around 3-5 feet of accuracy.

I get excellent results with Emlid Reach RS2 at ~$2000. Can reproduce land surveyed monumentation very closely using state CORS network for corrections, subscription required at about $600/yr

 

[Earthman]

Based on my experience, the accuracy of the point cloud obtained from an photogrammetric aerial survey using a UAV depends on the following:

1. The resolution of the camera,
2. How high you fly,
3. How fast you fly,
4. How smooth the air is,
5. The contrast of visible points on the surface being measured, and
6. Lighting conditions.

You can get 6” accuracy from a low altitude flight with a 12MP camera. I’ve done it with a DJI Mavic Pro. Any of the DJI UAVs that have a 12MP or better gimbal-mounted camera that can look straight down will work for what you are doing. I suggest you stick with DJI products since most of the available mission planning software supports them, but select and check the mission planning software first to make sure it supports the UAV before buying the UAV.

if you simply want to do surveys that are not tied to any standard reference frame or datum and you can live with some inaccuracies, all you need to do is post process the photos from the flight and this can give you an orthophoto of the site, a point cloud, DigItal Elevation Model (DEM), and contour map from which you can take measurements. Such a survey may have distortions in the orthophoto and resulting data since the photographic process isn’t perfect.

If needed, you can help the post processing software correct the distortions and inaccuracies by providing Ground Control Points (GCP) or targets in the photos. These are points of known (surveyed) location for which you identify and define the location of in the orthophoto. The post processing software will distort the orthophoto to fit these known points, which helps to correct the inaccuracies. I use white 5-gal bucket lids with a 3/8” hole Drilled at the center as GCPs. I put a large spike thru the holes into the ground to keep the lids from moving. We also paint targets on pavement when needed. It helps to number the GCPs to facilitate documenting their physical locations and corresponding coordinates but the numbers don’t have to be visible in the photos as long as you can enter the GCP coordinates correctly into the post processing software.

If you want to tie your aerial data into a standard reference frame or datum, you will need at least surveyed Ground Control Points (GCP) using the required reference frame and datum. Remember, GCPs are also needed to help correct distortions in the orthophoto.

Real Time Kinematic (RTK) GPS receiver(s) can improve the accuracy of the survey but at additional cost and you should still use GCPs if you are relying on photogrammetry.

The process described above will “map“ leaves in trees and tall weeds, which will distort your ground surface model, so the process only works well for open ground and/or when there are no leaves/weeds. If you need to map under the tree canopy and thru weeds, you need a bigger UAV, LIDAR, and the appropriate supporting software, which is a much more expensive process.

However you collect your aerial data, will not have an “aerial map” that includes surface features such as edges of pavement, sidewalks, mulched areas, landscape walls, etc. You will have to trace and add these manually in CADD using the orthophoto, etc. This is time consuming and expensive. So if you need this level of detail, you will need to become a mapmaker.

Making actual maps, including aerial maps using UAVs, for commercial purposes is usually restricted to professional engineers and professional surveyors by most states. I don’t know if this applies to simple landscaping, but you may want to check. Also, using a UAV for commercial purposes in the USA requires an FAA commercial drone license.

Consider using AirMap or Maps Made Easy as entry points for mission planning and post processing aerial photos. The apps are free and I think both services will post process smaller sites for free, saving you the cost of investing in higher end software and services to get orthophotos, point clouds, DEMs, and contours.

The bottom line is, working data collected by UAVs into your workflow can be benificial. It helps if you have the interest to stick with it and the skill to figure it out. It’s a process.

Good luck, and fly safely!

 

[LUIS MARTINEZ]

Based on my experience, the accuracy of the point cloud obtained from an photogrammetric aerial survey using a UAV depends on the following:

1. The resolution of the camera,
2. How high you fly,
3. How fast you fly,
4. How smooth the air is,
5. The contrast of visible points on the surface being measured, and
6. Lighting conditions.

You can get 6” accuracy from a low altitude flight with a 12MP camera. I’ve done it with a DJI Mavic Pro. Any of the DJI UAVs that have a 12MP or better gimbal-mounted camera that can look straight down will work for what you are doing. I suggest you stick with DJI products since most of the available mission planning software supports them, but select and check the mission planning software first to make sure it supports the UAV before buying the UAV.

if you simply want to do surveys that are not tied to any standard reference frame or datum and you can live with some inaccuracies, all you need to do is post process the photos from the flight and this can give you an orthophoto of the site, a point cloud, DigItal Elevation Model (DEM), and contour map from which you can take measurements. Such a survey may have distortions in the orthophoto and resulting data since the photographic process isn’t perfect.

If needed, you can help the post processing software correct the distortions and inaccuracies by providing Ground Control Points (GCP) or targets in the photos. These are points of known (surveyed) location for which you identify and define the location of in the orthophoto. The post processing software will distort the orthophoto to fit these known points, which helps to correct the inaccuracies. I use white 5-gal bucket lids with a 3/8” hole Drilled at the center as GCPs. I put a large spike thru the holes into the ground to keep the lids from moving. We also paint targets on pavement when needed. It helps to number the GCPs to facilitate documenting their physical locations and corresponding coordinates but the numbers don’t have to be visible in the photos as long as you can enter the GCP coordinates correctly into the post processing software.

If you want to tie your aerial data into a standard reference frame or datum, you will need at least surveyed Ground Control Points (GCP) using the required reference frame and datum. Remember, GCPs are also needed to help correct distortions in the orthophoto.

Real Time Kinematic (RTK) GPS receiver(s) can improve the accuracy of the survey but at additional cost and you should still use GCPs if you are relying on photogrammetry.

The process described above will “map“ leaves in trees and tall weeds, which will distort your ground surface model, so the process only works well for open ground and/or when there are no leaves/weeds. If you need to map under the tree canopy and thru weeds, you need a bigger UAV, LIDAR, and the appropriate supporting software, which is a much more expensive process.

However you collect your aerial data, will not have an “aerial map” that includes surface features such as edges of pavement, sidewalks, mulched areas, landscape walls, etc. You will have to trace and add these manually in CADD using the orthophoto, etc. This is time consuming and expensive. So if you need this level of detail, you will need to become a mapmaker.

Making actual maps, including aerial maps using UAVs, for commercial purposes is usually restricted to professional engineers and professional surveyors by most states. I don’t know if this applies to simple landscaping, but you may want to check. Also, using a UAV for commercial purposes in the USA requires an FAA commercial drone license.

Consider using AirMap or Maps Made Easy as entry points for mission planning and post processing aerial photos. The apps are free and I think both services will post process smaller sites for free, saving you the cost of investing in higher end software and services to get orthophotos, point clouds, DEMs, and contours.

The bottom line is, working data collected by UAVs into your workflow can be benificial. It helps if you have the interest to stick with it and the skill to figure it out. It’s a process.

Good luck, and fly safely!

"Making actual maps, including aerial maps using UAVs, for commercial purposes is usually restricted to professional engineers and professional surveyors by most states. "

Er, not sure this statement is accurate. Anyone can engage a UAS operator for purposes of creating a geomap, without state interference. It's done everyday. However, if you are referring to making a map and calling it a "survey" for property lines purposes or claiming one can create maps of "survey accuracy" then you are right; you'd be asking for trouble.

For what the OP is asking for (nothing of survey quality or tied to a datum point) any 'ol P4P with the right processing software, and following your guidelines, will be fine.

 

[Ajkm]

I’m not sure that I see much advantage in using a drone (with all the complexity that comes with mapping and post processing) over using Google Earth Pro and the polygon tool in the example/situation described.

Here’s a scientific research report on GEP accuracy - at one point it describes a difference between GEP and a hand-used measurement wheel of just 0.13 ft.

https://arcca.com/wp-content/uploads/2018/01/SAE-RESEARCH-PAPER-HARRINGTON-et-al.pdf

I should add that, if a 3D mapped image is what is required rather than just measuring a site (which is the OP’s description of the need), then I could see a use for aerial imaging.

 

[Dave Pitman]

I’m not sure that I see much advantage in using a drone (with all the complexity that comes with mapping and post processing) over using Google Earth Pro and the polygon tool in the example/situation described.

GE is great. But I anticipate the OP wants to do a bit more than measure the site, but who knows. I have provided orthos to landscapers and home owners employing landscapers. Before and after orthos, orthos when the irrigation is trenched and installed but not yet covered is a favorite on a property of any acreage.

 

[Dave Pitman]

For what the OP is asking for (nothing of survey quality or tied to a datum point) any 'ol P4P with the right processing software, and following your guidelines, will be fine.

Yup. ?

 

[Ajkm]

GE is great. But I anticipate the OP wants to do a bit more than measure the site, but who knows. I have provided orthos to landscapers and home owners employing landscapers. Before and after orthos, orthos when the irrigation is trenched and installed but not yet covered is a favorite on a property of any acreage.

I was working from what the OP posted "using a drone to measure sites for estimating purposes" and when combined with "laying an object of a known length on the site", there seemed to be no indication of anything that would need an RPA/UAS. I wondered if it might be a question of looking for a reason to use a drone rather than one actually being needed to do the job as described, but there may be more information that hasn't been shared. GE Pro is extremely accurate, and 0.2 of a foot is some 2.4", so far more accurate than required in the OP's proposed use. Personally, I would always use GE Pro over any mapping process if I don't need orthographic (3D) information about the site; why add to the workload to get information that is easily available from an armchair (well, desk chair!)?

 

[Dave Pitman]

I was working from what the OP posted "using a drone to measure sites for estimating purposes" and when combined with "laying an object of a known length on the site", there seemed to be no indication of anything that would need an RPA/UAS. I wondered if it might be a question of looking for a reason to use a drone rather than one actually being needed to do the job as described, but there may be more information that hasn't been shared. GE Pro is extremely accurate, and 0.2 of a foot is some 2.4", so far more accurate than required in the OP's proposed use. Personally, I would always use GE Pro over any mapping process if I don't need orthographic (3D) information about the site; why add to the workload to get information that is easily available from an armchair (well, desk chair!)?

No argument from me. I can see the value in aerial imagery for a landscape contractor beyond Google Earth satelite imagery. New construction is never present for months or years. I really don't see any need for anything approaching stampable accuracy though. I think the OP just doesn't yet know any better. Give them time.

 

[Earthman]

"Making actual maps, including aerial maps using UAVs, for commercial purposes is usually restricted to professional engineers and professional surveyors by most states. "

Er, not sure this statement is accurate. Anyone can engage a UAS operator for purposes of creating a geomap, without state interference. It's done everyday. However, if you are referring to making a map and calling it a "survey" for property lines purposes or claiming one can create maps of "survey accuracy" then you are right; you'd be asking for trouble.

For what the OP is asking for (nothing of survey quality or tied to a datum point) any 'ol P4P with the right processing software, and following your guidelines, will be fine.

Yes, it depends on what the OP intends to do with the data. Read the following statement form the PA Society of Land Surveyors for their interpretation of the laws of PA.

https://www.psls.org/resources/Documents/Statement_on_Mapping_and_Positioning_Services_2014.pdf

 

[LUIS MARTINEZ]

Yes, it depends on what the OP intends to do with the data. Read the following statement form the PA Society of Land Surveyors for their interpretation of the laws of PA.

https://www.psls.org/resources/Documents/Statement_on_Mapping_and_Positioning_Services_2014.pdf

Exactly what I stated earlier; stay away from the term "survey" and don't claim survey level accuracy. The PSLA is just protecting its turf...