Guest Column: Getting Started with Drone Surveying
With several dozen surveyors using our drone mapping system daily, Aerotas has keen insight into what it takes to use drones profitably and what gets in the way. A critical mistake that we see many surveyors make is simple: they start by buying a drone.
This mistake starts at the beginning. Most surveyors begin the process of developing their drone program by immersing themselves in every technical spec they can find on every drone they can find. They spend months researching, only to end up buying a drone that often ends up being a poor match for their actual needs — and is often more expensive than was necessary.
The truth is that the drone is the easy part. The right drone for nearly every survey application is going to be inexpensive and easy to operate in the field. Therefore, rather than starting a drone program by focusing on what drone to buy, the most successful surveyors begin instead by focusing on exactly what they need the drone to do for them.
By starting at the end of the process and working backwards, they develop a complete drone program that is sure to be as efficient and effective as possible. Starting with spelling out the deliverable production workflow informs what software is needed and how the drone data needs to be collected. This then informs field operations and procedures, and finally leads to defining what the right drone is for the job.
What will the Drone Do?
The answer to “what do I need the drone to get me?” for most surveyors is going to be a topographic survey, including a surface, contours, breaklines and features. Though this may seem obvious, it is surprising how many surveyors don’t think this through before buying their drone. Instead, they end up with an expensive drone and complicated software that only enables them to produce a 2D aerial orthophoto.
While orthophotos certainly provide some benefit, they clearly do not enable a surveyor to produce a topographic map, which requires elevation data.
Other surveyors manage to produce 3D models from their drone data, but only through an unnecessarily labor-intensive process involving point cloud management and/or trying to load massive 3D models into their finish software (e.g., Civil3D). While finish software programs can technically handle 3D data, they are not optimized for the rich and large drone models and so are extremely slow and cumbersome to use.
The much more efficient deliverable-production workflow involves software specifically designed for extracting spot elevations and features directly from a 3D model. This software intakes the 3D model that is produced by stitching together the photos from the drone, meaning that the image stitching process needs to be optimized for use in that software. Getting high-quality results from the image stitching process requires experience, particularly in quality assurance and control procedures.
Once a surveyor has scoped out the deliverable production workflow, the next step is to define the field operating procedures. Well-defined field procedures are critical for ensuring that the right data is collected, and for making sure that the full operation is as safe, efficient and profitable as possible — as we explored in depth in a previous article.
The best operating procedures are easy to train staff on and easy to follow, while covering every important element of safety, legal compliance and quality assurance. A great deal of research, development and testing is required to create a set of procedures that meet these requirements.
Finally, the last step in the process of developing a drone mapping program is to choose a drone. At this point, this should be a straightforward process of identifying the most inexpensive and easy-to-operate drone that matches the field operation and deliverable production needs that have been defined. Complement this with a regulatory compliance strategy, good insurance and a staff-training plan based on the defined workflow, and a successful drone mapping program is born.
We have seen this systematic, deliverable-focused approach succeed countless times.