Scanning for Construction – Is It in the Right Place?
I met Andrew Evans, Product Manager at Topcon who offered this blog post on Scanning for Construction, plus a look at the GLS 2000 scanner with its ability to set up over a point and find a target for a backsight, meaning you can know the coordinates of the instrument, as well as the instrument height.
The role of the surveyor in construction. It’s time to claw back some recognition for applied mathematicians. Now, from the very start of this piece I should stake my professional interests.
Confession time: I, like many others in my industry segment, am a professional geospatial trained consultant.
A what you may ask?
Let me re-phrase that and be a little more specific. I know how to apply maths to make sure that the stuff we build today, gets built in the right place in the right orientation and to the intended design. Members of my trade also know how to quantify and validate the materials moved around and installed during a construction project.
The current trend in my industry segment revolves around mass data collection, mass data by definition implies millions of points that have coordinate information as a root element but also enables us to communicate existing conditions and make this highly accurate data look like a pretty picture.
That’s the beauty of mass data, there’s more to it than meets the eye.
Let’s cut to the point (‘scuse the pun). As one of the mass data collection tools available to us today and due to the associated software it is now possible to cost effectively employ laser scanners on a construction site not only for existing conditions, or as-built documentation, but with the right tooling and workflows it can be used to achieve a continuous update of reality a.k.a continuous representation of reality, continuous as-built. Whatever, let’s not get caught up in naming discussions, we’ve seen that confusion elsewhere, discussing the suitability of BIM as an acronym that correctly describes a process . http://youtu.be/Wac3aGn5twc
The bottom line is, with lidar based instruments we have a rich, informative, base data set that when combined with smart software gives us the potential to report daily site progress as pretty, compelling pictures as well as numbers and spreadsheets.
Those awe inspiring point clouds offer much much more than just an incredibly compelling visual communication of complex 3d structure and infrastructure.
There’s a couple of gotchas though. I’ll cover two here in this blog, but I’m conscious I only have your attention span for a shirt wheel. (Still paying attention?!)
1) Repeatable, accurate, and precise control is critical to enable this concept and minimise poor reporting – ensure your applied mathematicians are compensated well for their expertise, always challenge their coordinate deviation reports and what they mean and how they define both the accuracy AND the precision.
2) Lidar data can “mislead” point cloud analysts. An easy fix is to collect and utilise the geo-referenced images as a default to ensure correct data interpretation. However , all measurement tools have some known system uncertainty in the resulting measurement. The beauty of scanning is that you have so many points that the best fit of say a wall position will typically be much more precise than what the system specification leads you to think at first. For example a high quality laser scanner such as the Topcon GLS-2000 reports 3.5mm range error on the specification, but this is representative of measuring a single point many many times. Actually we should be more interested in the statistics of a surface fit on the points to indicate the quality of the scan data.
But to back up the statements on “misleading”, the nature of the measurement mechanism of a scanner means that you can see stray points at edge interfaces, reflected points from glass and water, no points from dark surfaces and highly shiny oblique surfaces, refracted points when the laser has measured through glass. All of these points can exist in a data set and quite happily mess up your coordination of the final point cloud, and further more can be accidentally selected during point cloud analysis. I’m not suggesting your point cloud should be clear of all noise points before accepting it, however, as achieving this can add significant cost to a project, you just need to be aware of sources and likely locations of erroneous points as there are many awesome tools that can achieve great analysis on “dirty” data without introducing a bias to the numbers. As I’m sure you will all appreciate: If you know where the error can creep in, typically it is already dealt with.
It’s clear that mass data collection is going to become simpler, quicker and an even richer data source than it is already. For the construction industry there is also now a relatively clear path highlighting that it is almost painless to implement scanning technology and reap the benefits in time and money on site.
If you want you want to discuss this further. Get in touch.
You can find @TopconImaging on Twitter or via Andy Evans on Linked In.
For more on the GLS 2000 click here.
Editor’s Note – Thanks Andy
Note – If you liked this post click here to stay informed of all of the 3D laser scanning, geomatics, UAS, autonomous vehicle and Lidar News. If you have an informative 3D video that you would like us to promote, please forward to firstname.lastname@example.org.