Applying Drones to Surveying & Engineering Today

I tested Bentley's Context Capture, Topcon now supports it also. It has most of the features as Pix4d. Best feature that stands out is render farm. You can purchase the engine for multiple PC's on your network and share resources for very large projects..

leegreen, post: 419055, member: 2332 wrote: I tested Bentley's Context Capture, Topcon now supports it also. It has most of the features as Pix4d. Best feature that stands out is render farm. You can purchase the engine for multiple PC's on your network and share resources for very large projects..

Their Descartes software looks pretty good for 3D feature extraction from point clouds. What are the other software options for 3D feature extraction? I have heard of TopoDOT that runs with Bentley Microstation. Is Autodesk into this at all? Any other standalone software products?

Carlson does have a point cloud software, not sure if is on AutoCAD or their own CAD engine.

It does seem that most of the cloud based extraction software is geared towards Bentley. Probably because their software has always been well suited to handle large data sets, as most DOT's use it.

Bentley's latest Connect version of Microstation is a 64-bit application. The benefit of a 64-bit application is the ability to use more memory, which will enable huge models (or models having a large number of references) to be handled. .

Thanks for posting! this is exactly the question I was asking the other day.

That's really odd that they did not include the GCP's. Since the GCP's need to be applied first, then process the point cloud, then extract data. They wasted many hours, that will need to be repeated.

leegreen, post: 419094, member: 2332 wrote: That's really odd that they did not include the GCP's. Since the GCP's need to be applied first, then process the point cloud, then extract data. They wasted many hours, that will need to be repeated.

He said that the data "is not ortho-rectified per se" and concerning the ground control points they are trying to "work it into their workflow a little easier than it is now". I know that they are using onboard post-processed GNSS to obtain the precise camera focal point positions, but I wonder how they are getting precise orientations? Also, I thought onboard GPS and IMU in the Photogrammetry world meant you just needed less ground control, as opposed to no ground control at all.

Back in the 2000s, I worked on a Low Altitude Mapping Photogrammetry (LAMP) project on a portion of I-75 in North-Florida and we had ground control points every 500 feet on both sides of the road on the shoulders. Although, I am not sure if they had onboard positioning or not. It was taken at low altitude from a helicopter at 1,000 feet I think. We did QA/QC on the road surface with reflectorless total stations. As I remember, I think it was plus or minus 0.1'. I heard that the FDOT project managers didn't think the LAMP results were good enough, so they used our QA/QC shots instead.

Hello all! I am Ryan MacNeille, the Data Operations Manager from this video. I appreciate the conversation here about this webinar, this was our first attempt to share some of our projects and partnerships so please bare with us as we work to make these better.

I would be more than happy to answer any questions if anyone is interested in more information about the project. Indeed we did not use ground control for the initial processing of this data, which is normally a standard procedure for us. The reason for this is due to the time difference between the traditional survey and the drone survey; there were no surveyed targets on site when we deployed our team. Although the final comparative results seemed less than desired during KLJ's presentation, we both mutually discussed the many variables biasing this data comparison.

We have conducted many traditional vs. UAS tests in the past using metric sensors, PPK GPS, and ground control and have received much better results than what was shown in this webinar to verified centimeter-level accuracy. However, we felt this was an interesting project to discuss in terms of workflow influence and wanted to show the real-world logistics of integrating drones into well-adopted industry software. We are continuing to work with KLJ and other surveying firms on future projects so I would say not to take this example as definitive accuracy potential.

I will be doing another webinar soon using the Autodesk Civil3D software, with which there is much potential for easy feature extraction and other forms of data analysis. Again, I am happy to answer any questions, we do not want to be a secretive company 🙂

Ryan MacNeille, post: 419199, member: 12579 wrote:
I will be doing another webinar soon using the Autodesk Civil3D software, with which there is much potential for easy feature extraction and other forms of data analysis. Again, I am happy to answer any questions, we do not want to be a secretive company 🙂

Ryan,

Will you let us know when the C3D webinar comes out?

Gregg

GMPLS, post: 419211, member: 8404 wrote: Ryan,

Will you let us know when the C3D webinar comes out?

Gregg

Hi Gregg, I certainly will.

Ryan MacNeille, post: 419199, member: 12579 wrote: Hello all! I am Ryan MacNeille, the Data Operations Manager from this video. I appreciate the conversation here about this webinar, this was our first attempt to share some of our projects and partnerships so please bare with us as we work to make these better.

I would be more than happy to answer any questions if anyone is interested in more information about the project. Indeed we did not use ground control for the initial processing of this data, which is normally a standard procedure for us. The reason for this is due to the time difference between the traditional survey and the drone survey; there were no surveyed targets on site when we deployed our team. Although the final comparative results seemed less than desired during KLJ's presentation, we both mutually discussed the many variables biasing this data comparison.

We have conducted many traditional vs. UAS tests in the past using metric sensors, PPK GPS, and ground control and have received much better results than what was shown in this webinar to verified centimeter-level accuracy. However, we felt this was an interesting project to discuss in terms of workflow influence and wanted to show the real-world logistics of integrating drones into well-adopted industry software. We are continuing to work with KLJ and other surveying firms on future projects so I would say not to take this example as definitive accuracy potential.

I will be doing another webinar soon using the Autodesk Civil3D software, with which there is much potential for easy feature extraction and other forms of data analysis. Again, I am happy to answer any questions, we do not want to be a secretive company 🙂

Ryan, it definitely was an honest video presentation of the project that didn't gloss over the difficulties you had, which I respect. Many of us that are starting to get into this would like to know the realistic potential of drone photogrammetry and what the recipe is for good results. I will look forward to more from you guys.

Very informative. Thanks for sharing.

I've been doing some experimenting with drones. Another company does the flying for me and delivers a pointcloud. I give the control points with rtk-gps.

For x and Y it's ok and usable. For z-value only for rough terrain work, volumes,... Not for as-builts. Over here they want cm-accuracy in height (manholes etc.) and that is not possible with my findings. Some values are spot on but other are up to 5cm off.

For an overall view and materials used it's perfect and a good addition to a surveying business.

I've got a running project were i scanned the interior of a church and the same drone company is flying the exterior. When it's ready i'll present some results.

beuckie, post: 419298, member: 2245 wrote:
For x and Y it's ok and usable. For z-value only for rough terrain work, volumes,... Not for as-builts. Over here they want cm-accuracy in height (manholes etc.) and that is not possible with my findings. Some values are spot on but other are up to 5cm off.

Well, you have to think that the vertical from the onboard GPS is going to have at least a tenth of uncertainty from the get-go. Then you have to stack on top of that whatever error you have in the vertical of the mapping system you are using. It all adds up in the final result. To me it speaks to the need for precise ground control.

Bow Tie Surveyor, post: 419056, member: 6939 wrote: Their Descartes software looks pretty good for 3D feature extraction from point clouds. What are the other software options for 3D feature extraction? I have heard of TopoDOT that runs with Bentley Microstation. Is Autodesk into this at all? Any other standalone software products?

I use LP360 and love it. It works as stand alone and inside arcmap. It allows programming of automated point classification and feature extraction. We also use terrascan and it appears to be very similar but works on top of microstation.

First, let me say we are a reseller for both 3DR and Autodesk. Drones and photogrammetry can be very useful with or without ground control. It all depends on what final deliverable you need.

Think back before technology added so much to the AEC industry. Aerial flights were needed to capture large sites with supplemental survey for the detail desired. That scenario is no different here. Except that the drone can be used to address small to medium site capture and return the RAW data (point cloud, ortho-mosaic) very quickly.

Drones will not replace aerial flights nor will it make total stations obsolete. It's just another tool to use in capturing a site.

Given current technologies, beuckie is right on point. But would we use the elevation from an aerial flight for a manhole? Not normally, we would go survey that feature traditionally.

That said, this aspect of the surveying industry is transforming on a monthly basis. Between 3DR, DJI, Autodesk and Bentley, there is a value to quickly obtaining data for construction, for design and for inspection. Its very clear that technology in the AEC market is moving at a faster pace today than it ever has. It's even hard for me to keep up with everything that's changing. So don't feel overwhelmed, just ask questions when the time is right...

I just wanted to add a little calm to the tech storm that we're all squarely centered in at this time.

Here is a follow up video that Altavian just put out on the test project:

[MEDIA=youtube]zWZLlLWVen8[/MEDIA]

I just found this. They give you a 1 year subscription to try it out for free. https://www.icaros.us/onebutton/

My question about drone surveying/mapping is "Should the volume, location, topographic renderings and alignment of the Aerial Services be certified by a Licensed Land Surveyor?"

I have been doing research. We are dabbling in it right now and the more research I do the more I am noticing volumetric and topographic applications by non-surveyors.

The more research I am doing on these micro projects, the more I am convinced we are watching the ship sale by, similar to what we did with GIS.

http://www.asprs.org/a/society/divisions/ppd/State_Licensure_Map/State_Licensure_Map.html