Selecting a Site
The drone is an excellent tool to have in your surveying arsenal, but it is not the right tool for every job. At the beginning of every project we decide how we are going to execute our data collection, whether it be with the drone, or traditionally. Determining the effectiveness of the drone for your specific job will be pertinent to your success and overall satisfaction with your drone. So how do we make the decision on whether or not to use our drone? For all examples here we will reference the Phantom 4 series (P4RTK and P4PV2.0) for photogrammetry mapping.
It is important to set realistic expectations for the drone, the first and most all-encompassing is that the drone can only map what it can see. It is equipped with a 20MP camera which is its data collection apparatus. So naturally, if the camera cannot see it, it cannot collect the data. The biggest consideration when deciding if it will be worthwhile to use the drone or not is the type of ground cover at the job site. If there are trees, or extensive ground cover, such as brush or shrubs, the drone will not be able to accurately map the earth's surface. For a site that has extensive ground cover, traditional land surveying methods will be more effective.
A poor site for drone photogrammetry. With all of this tree cover, the sensor will not be able to obtain enough true ground shots to stitch an accurate representation of the surface.
A good site for drone photogrammetry. With minimal tree cover over the subject area, the surface is visible and full of features for matching to derive an accurate surface.
Another factor will be airspace. Is your site located within 5-10 miles of an airport? If so, we have to check the airspace with the FAA and DJI Geofencing and deal with them accordingly. If the area we plan to fly in is within a mile of an airport, there is a chance there will be a 0-foot ceiling and it can be difficult to get authorization to fly over that site, and alternate plans should be made in case that happens.
Additionally, you need to consider what type of area you are flying over; if it is rural or industrial, that’s usually a full-on green light; and you are good to go. But if the site in question is in a heavily publicly trafficked area, such as a school, shopping center, or even neighborhood, you have to operate with discretion, and not during peak hours of business to mitigate risk to people on the ground.
There are also limitations to what the drone can discern. For example; the drone has a hard time picking up the front of curb and thus will be an ineffective tool if that is the main feature you are after.
RTK Vs. Non - RTK
At the beginning of this section we mentioned the P4 series drones. There is a delineation between what a good site is for the P4P and what a good site is for the P4RTK. Their main difference is in regards to accessibility and GCP requirements.
The P4P requires many GCPS, evenly distributed throughout a site to ensure proper anchoring of the model. Larger sites require more and more GCPS to keep data accurate.
The P4RTK only requires 3-5 GCPs per site to VERIFY accuracy of and georeference the already relatively accurate model, so distribution is not as important. Larger sites do not require more and more GCPS to keep the data accurate. Regardless of size the P4RTK only needs 3-5 GCPs per flight.
Bottom line: With non-RTK Aircraft, you need to make sure you can set control evenly distributed throughout your entire site in order to get accurate data everywhere. With an RTK enabled aircraft, you do not necessarily need to access your entire site to get accurate data on it. Keep this in mind when selecting a site for drone surveying.