Drones have become an essential tool throughout the lifecycle of a construction project. However, one area where they have sometimes fallen short is in conducting as-built surveys. This is for a simple reason: drones have been designed to effectively capture imagery and data from the top down, but not from the side. Surveying a groomed site for pre-construction, for example, is far easier than trying to capture taller, finished structures—the flight is more complex, and requires the correct altitudes and gimbal angle in order to ensure accuracy.
We cracked that code with perimeter scan, our newest flight mode. Perimeter scan is a first for the commercial drone industry: it enables automated scans of vertical structures, collecting far richer data than ever before. We’re excited to share how Argyle, one of our customers, has already put perimeter scan into use on one of their projects, saving significant time on-site and creating an impressive, BIM-ready deliverable.
The goal: fast, accurate as-built survey of mixed-use property
Argyle Asia is an integrated firm based in Southeast Asia that does real estate investment, property and infrastructure development, and design and construction delivery. James Hadden, Director of Property Development & Construction Management Services, led this project and set out to perform the as-built survey at a group of mixed-use buildings in Kuala Lumpur.
Argyle’s goal was to fast-track the collection of as-built data that could be used in the design development of the property assets. Since the client had limited historical records and had very little architectural drawings, Argyle needed to collect an accurate as-built for design.
The site itself also presented a number of challenges when performing a traditional survey. For example, it was densely populated with large trees that were in close proximity to the structure, making it difficult to capture the points on-the-ground. Also, the architectural features and alignment of the buildings required additional time to survey. “There’s also the disruption to commercial and residential tenants to consider,” James said, “in addition to the operational safety risks of a manual survey.”
Because of these factors, James thought that capturing the site with a drone would be a great alternative. However, Argyle needed a drone solution that would make it possible to scan the façade of the buildings without any gaps or major inaccuracies.
The solution: detailed 3D models using perimeter scan
Argyle decided to use the perimeter scan flight mode on Site Scan to capture the property and create a photorealistic model. He set 5 survey control points to lock in the survey data with the orthomosaic and point cloud, and then performed the flight. “In terms of workflow, it took minutes to set up the UAV and run the survey—it was fast, reliable, efficient and the registration of the scan was great.”
Given the complexities of the site, it was vital to capture the parts of the building that would be difficult to capture on the ground. “The drone worked well with getting into the confined spaces that I didn’t think would be possible to capture,” James said. “It picked up faces on the building that I thought would be obscured by the trees or the shadows.”
5X production of georeferenced CAD files
For James, the saved time and reduced manpower required on-site is one of the biggest benefits of perimeter scan. “A traditional as-built survey for this project could take 3-4 weeks with additional manpower, and that’s just for the survey,” he said. “Then, it may take 2 ½ weeks to interpolate the survey notes and prepare the CAD as-built survey and Revit model. In total, it would take approximately 7 weeks—49 days—to deliver Revit and AutoCad drawings.”
“I think perimeter scan is the first of its kind. No other drone solution provides the ability to autonomously scan vertical structures at this level of detail.” — James Hadden, Argyle
By collecting drone data, James said, “we are now in a position to provide that deliverable much quicker. In this project, we scanned the site, processed the data into a geotagged orthomosaic and point cloud, and converted them into CAD-friendly files in under 3 days. Then, about a week later, we finished the georeferenced CAD and Revit drawings, taking a total of just 10 days.”
This time difference—going from approximately 49 days to just 10—represents a near 5X increase in time to deliverable on this project. “While each as-built survey is different, and has its own unique time constraints to consider, there were considerable time savings by using a drone on this project,” James said.
BIM-ready 3D models
Traditional as-built data is usually kept in a spreadsheet, making it difficult to visualize and integrate with BIM workflows. As BIM continues to make its way into the AEC industry, this lack of 3D as-built data is starting to present some difficulties.
Case in point: In Europe, as-built BIM models are soon to be required in all construction projects. However, according to a survey in the JB Knowledge 2015 Construction Technology Report, less than half of U.S. companies would be able to comply with this as it stands.
Collecting drone data—especially with flight modes that are designed specifically for scanning vertical structures, not just topographical surveys—makes this easy. James was able to create rich 3D point cloud and meshes, and could use them to easily identify the infrastructure features of the building, which could then be incorporated in the 3D model for BIM integration. “We were able to run that survey, register the scans, and start modelling and taking the sections within hours of having done the scan.”
“With the point cloud tools in Revit,” James continued, “I could take sections of the building, establish the profiles, and start drawing perimeter walls and detailing windows. The beautiful thing about a project like this, for example, is that the windows are fairly standardized. This means I can create them in Revit as a family in the library, then easily use that for the other windows.”
Now that this kind of data collection is possible, James is looking at different ways in which it can be used to drive value in his projects across Asia. For example, he would like to collect indoor laser scan data of the site, and integrate it with the point cloud in order to get a complete model of existing conditions.
“I think perimeter scan is the first of its kind,” James said, “no other commercial drone solution provides the ability to autonomously scan vertical structures at this level of detail. I look forward to continue on even bigger, more complex projects in the future.”