3D Aerial Survey Technologies: Which tool to use, and when to use it
By John Leipper, Insitu Solutions Architect
Surveyors have no shortage of tools at their disposal to aid their land survey activities, such as construction layout and control, earthwork and volume survey, utility mapping, and topographic mapping for engineering design, to name a few. By leveraging advances in GPS and laser scanning, terrestrial-based survey equipment has improved not only the positional accuracy of measurements, but also the sheer number of measurements obtained.
While ground equipment has improved the efficiency of surveyors to survey more area in less time, larger projects still benefit from airborne 3D survey technologies, such as photogrammetry and Light Detection and Ranging (LiDAR). Insitu has compared the benefits of these airborne 3D survey technologies and their applications in traditional survey use cases. Today, I’ll share insight on how to determine which tool will provide the most value in specific use cases. For a more comprehensive look at our observations and conclusions on this topic, I invite you to follow the link at the bottom of this article to read the complete white paper.
Use Cases and Trade-offs
Admittedly, there are benefits and shortcomings when it comes to LiDAR and photogrammetry that need to be considered when choosing a technology for a particular use case. Through our extensive experience performing 3D aerial survey for our customers, we’ve established some general guidelines on when to use one technology over another.
- Better for fine details, especially linear surveying, such as powerlines and train tracks.
- Environmental operational considerations are mainly tied to the reflectivity of the surface being scanned and visibility on the day of operation. LiDAR can operate day or night, over water, in partly cloudy conditions, and in feature-free environments. Smog and highly matte finishes (e.g., dry snow) adversely affect LiDAR performance.
- Better for broad areas and scalability due to relatively low amount of data created. One platform can survey many sites in a single mission.
- Faster processing time. Raw point cloud available in near real time.
- Accuracy with minimal calibration overhead
- Generally heavier and more expensive, requiring a larger platform (~ 5.5kg) for LiDAR ranges which are compatible with fixed-wing platforms. Smaller 2 kg systems are only suitable on multirotor platforms with current technology.
- Higher cost, but lower overheads (like computing power, size of data involved, processing time)
- Altitude capped on LiDAR range (currently typically to <2,000ft AGL on a LiDAR small enough for a 55 lb. UAV).
- Needs airborne camera alongside to “colorize”
- Better for open, “bare earth” survey. Reflections and trees adversely affect result and number of artifacts.
- Environmental operational considerations include shadows (especially changing shadows through-out a mission), amount of sunlight, shadows from clouds, and reflective or feature-free (e.g., bodies of water) inclusions in data.
- Better for smaller specific areas (around 5-7 square km) to keep post-processing time to within 24 hours and amount of data to transfer acceptable (e.g., transfer of data to cloud or customer).
- Better point cloud density if you don’t need the data in a rapid fashion. Also able to choose a desired density when rendering a result.
- High accuracy (<1.0m) comes with logistics overhead (as large number of Ground Control Points are required). >1 meter accuracy without GCPs, or RTK GPS
- Generally lighter, can be carried on more platforms (~2 kg)
- Lower cost but higher overhead
- Larger operational envelope with fewer restrictions on altitude and ability to size cameras and lenses to the appropriate capture geometry.
- Output model inherently colorized
Of course the trades discussed above aren’t exhaustive, but provide a glimpse of the trade-offs we consider when choosing a sensor and platform combination that answers a customer’s need. Additional aspects that we consider when evaluating a use case include:
- How quickly is the data needed? (For example, for live dig-planning, emergency response, etc.).
- Are there environmental considerations that favor one payload over the other for a given site? (Are there large bodies of water? Is it a smoggy environment?
- Do mid-day operations suit the customer? (This is a scalability challenge for photogrammetry.)
- Is a high-bandwidth internet connection or powerful PC available on site?
- Do you have high accuracy needs? (If the use case involves volume measurements, then photogrammetry might be “good enough.”)
- Will the customer allow GCPs within their site? (Many sites in Australia are moving away from people in the site and they do not accept GCPs as standard practice, favoring LiDAR.)
- Is radio Line of Sight (LOS) an issue? Many sites can have large sand hills surrounding deep mines, meaning LOS can be challenging at LiDAR altitudes; thus photogrammetry survey from higher altitudes is favored.
- Will the aerial survey “scale up” over the life of site operations? (LiDAR will scale more readily than photogrammetry due to sunlight, shadow growth and big data constraints.)
From our experience, our customers tend to want high accuracy for their use cases, and will not tolerate GCPs on their sites. This starts to make the case for LiDAR favorable, yet there are many customer use cases for which this won’t be the case. Through collaborations with our customers, we help identify the proper 3D tool that is appropriately sized for the job at hand.
We at Insitu understand the benefit of 3D data collections at a large scale and are constantly working to advance the state-of-the art of planning, collection, analysis, and distribution of aerial data. Soon we’ll unveil details of our wide-area survey capability, which will help users gain insight into physical assets covering large geographic regions. Together with our customers, we’re working to improve the efficiency of existing asset management workflows and wide-area 3D survey is one tool in our aerial survey toolbox.
Get the full white paper here