One thing is absolutely certain: higher price does not always mean higher value.
We’re taking a magnifying glass to the professional surveying industry to see how drones are making a big impact on the bottom line.
This case study goes a bit deep into surveying jargon and mission-specific requirements but don’t let that deter you. The extra post-processing done to the DroneDeploy data in this study would be necessary only if your operation requires extremely high accuracy (in the case below, an ‘absolute’ accuracy of 3.6 in/9.14 cm was achieved).
The information below was given to us during an interview with drone operator Madison D. from Landpoint, a surveying company in Louisiana. Madison is researching the potential applications and benefits of unmanned technology in the surveying industry. Madison uses a Stratos Aerial fixed-wing UAVwithDroneDeploy (and a DJI Inspire 1on the side).
Topographic surveys are an essential part of all land development projects
In this instance, a new real estate subdivision was under development in Northern Colorado. Before homes could be constructed, an extremely accurate topographic survey was necessary for a couple of reasons:
- To ensure the initial land development (physical alteration of the land) was successful so that it allows for proper water runoff for drainage.
- To document subdivision topography in relation to the adjacent river’s floodplain for flood damage prevention and flood insurance purposes.
Subdivision development can be an expensive operation — especially if progress has fallen behind schedule
This particular project was weeks behind due to frequent inclement weather (notice the low-lying areas in the orthomosaic reference map above that are still holding rain water).
Any measure that saves time will move the project one step closer to meeting scheduled deadlines and help reduce the cost of recent downtime.
Surveying 101
Traditional topographic surveying requires the collection of GPS points (or “shots”) in a pre-determined grid. In this case, a 50 x 50 foot grid was used:
Yellow 50 x 50' grid overlaid onto the DroneDeploy orthomosaic
The GPS shots were collected every 50 feet at each intersection:
The red points were the ‘shots’, taken at every intersection on the grid
A total of 1,632 GPS shots were collected across the 85.4 acre survey area.
Without a drone, at a rate of ±20 points/hour (1 point every 3 minutes or so) the GPS shot collection would have taken approximately 82 hours.
1,632 GPS shots over 85.4 acres
82 hours of traditional surveying means that the subdivision developer would've had to expect at least 1 full week of field work before data processing and review could begin. After that, it would take another 3–4 days before the final product could be delivered.
Enter DroneDeploy
By conducting the same 85.4 acre topographic survey using UAS technology with DroneDeploy, Madison and his team were able to immediately provide the subdivision developer with a much quicker survey option.
First, there was no need to collect 1,600+ GPS points across the entire acreage. Instead they only needed traditional survey GPS point-collection on just 10 ground control targets (or GCPs), strategically placed within the survey area:
For larger projects, it can be beneficial to place GCP targets in a grid system as well but for this project, strategic placement yielded the same accuracy in much less time
10 GCP targets vs. 1,632 survey points:
In a perfect scenario, these 10 GCP targets would be the only points collected and could be completed in no more than 1–2 hours.
However, remember the standing water mentioned at the beginning of the case study?
Those familiar with photogrammetry will know that points collected on the surface of the water (homogenous imagery) are not acceptable for use in surveys like this.
So what happened next?
Having completed GPS collection of the GCP targets, they simply collected the traditional grid point locations in the areas with standing water — a combination of the two different methods above.
Final collected points:
10 GCPs and areas of standing water substituted with the traditional method of collecting grid point locations
This brings the final total of GPS points collected to 117 (10 GCP + 107 Natural Ground under water).
GPS timelines:
Ideal/Normal UAV Scenario: 10 GCP target placement + GPS collection = 1–2 hours
Actual Project Scenario: 117 GPS points (10 GCP + 107 additional GPS) at ±20 points/hour = 5.85 hours
Traditional Survey Scenario: 1,632 GPS points (collected every 50’) at ±20 points/hour = 81.6 hours
The flight
Using a Stratos Aerial fixed-wing UAV, the entire 85.4 acre survey area was completed in a single flight of approximately 22 minutes.
Total drone operations including assembly, pre-flight checks, launching, landing, disassembly, and initial DroneDeploy map stitching were completed in the field in 1 hour.
UAV Data Collection (1 hour) + GPS Survey Collection (5.8 hours) = Total time for field work: 6.8 hours
Comparison:
85.4 Acre UAV/DroneDeploy field work = 6.8 hours
85.4 Acre Traditional Survey field work = 81.6 hours
Total savings of 74.8 hours
(enough time to watch the entire Lord of the Rings trilogy… 8 times in a row)
Analyzing the data
Typically, the DroneDeploy data would’ve been good-to-go at this point. Since this is a highly specialized case, Madison needed to accomplish some additional steps to satisfy the extreme precision required for this job.
Once field work was completed, the drone’s data needed to be processed more thoroughly than traditional survey data. First the GCP targets had to be processed, finalized, and fully geo-rectified.
Next, the geo-rectified points (.las file) had to be exported to generate a foundation for the topographic data. However, the sheer number of points in the .las file meant the initial topographic contours would first come out quite rough:
Raw, rough contours before processing (pre-processed)
The contours must be smoothed in order to generate a consistent line without sacrificing accuracy or the survey data is unusable for this case.
After 2 days of additional processing, the resulting topographic contours were clean and accurate within 3.6 inches both horizontally (X,Y) and vertically (Z):
Clean contours after post-processing (post-processed)
Pre-processed vs. Post-processed Contour Data
Total Project Timelines
DroneDeploy
6.8 Hours Field Work + 24 Hours Data Processing =
30.8 Hours to Completion (approx. 4 days)
81.6 Hours Field Work + 24 Hours Data Processing =
105.6 Hours to Completion (approx. 13 days)
*For the sake of comparison, data processing for each survey is considered as 24 hours (3 working days) — this is an efficient timeframe for completion of this project using either survey method.
Savings
By using DroneDeploy’s technology, the subdivision developer received the final topographic survey approximately 75 hours sooner than a traditional survey would have allowed.
That’s 75 hours closer to being back on schedule and a hefty cost savings to offset the cost of the inclement weather down-time that was experienced earlier in the project.
Insights revealed
- Additional land development is needed to allow for proper run-off drainage in low lying areas where water is being held.
- Preliminary subdivision topography is now documented in relation to the adjacent river’s floodplain. A final “asbuilt” topographic survey will be needed when land development is complete for Flood Damage Prevention and Flood Insurance purposes.
- The subdivision developer now knows the exact amount of effort remaining and can efficiently forecast and schedule construction dates for roads, homes, etc. — moving the project closer to being back on schedule.
- The subdivision developer also learned of time-efficient, cost-effective UAV surveying and plans to utilize this method again for the final “asbuilt” topographic survey in the coming weeks.