The company who occupy the building are in the process of moving office and the existing tower block is due for reconstruction. The details given to us by the client were very limited but we were asked to create floor plans of each of the 7 floors. (6 office floors and plant room) plus a roof survey and 4 elevations for each façade.
My first walk round the building was a long one. I estimated each floor would consist of 60-80 two minute scans. Then the plant floor would take 100 plus scans and the roof, plus full colour elevations of the externals. It quickly dawned on me that this would be by far the largest scan project I had ever done in my years of surveying.
Normally the routine for 3D laser scanning projects is to create the point cloud, fixed in one faro scene job then move this onto survey control. With a job this size there is no chance of this happening as the processing power of my laptop is no where near fast enough. So I decided to fix each floor independently with two sided checkerboards shot in from a closed loop of stations externally. After I had stuck on all the targets to the windows I had coordinated 150 targets. These were easy enough (all be it exhausting). Trying to co-ordinate the roof with survey control was more challenging. There was a 4 meter high parapet wall enclosing the whole of the roof making shooting survey control up there impossible. The only way I could overcome this was to independently place targets on the roof and GPS the stations using the Leica GPS rover unit. Once coordinated I surveyed in line and level of the parapet wall from the ground and at roof level and I was achieving 30mm tolerance between the two. Not ideal but more than acceptable for the purpose of the survey.
After all survey control was established it was time to start scanning. I started with ground floor and working my through using the survey spheres for traversing and scanning in the coordinated checkerboards on the window. This was a good way of doing things and I quickly worked my way up the floors. After 3 tough days I had reached the roof and completed the internal part of the survey. Next I tackled the outside. Once again I traversed round using survey spheres for a target based registration for Scene 6.2 and the coordinating the survey off the checkerboards. This was quickly done with no real problems.
Then came the task of processing. This was time consuming and pretty unpleasant. Working through processing each floor as a separate project was the only was to tackle the task. The biggest problem I had was referencing the scan data to survey control which I have written a separate blog for. If you have had serious issues with referencing in scene I recommend reading it as I think you will find it helpful. After I had each floor coordinated it was simply a case of exporting as .fls File and improting it into Autodesk Recap.
A large job and I have to admit it was not a very enjoyable one but I was happy to see it over and done with and I was pleased with the overall accuracy of the project. I'm also glad I get the pleasure of handing this one over to the office for drawing floor plans and elevations as it’s a long task and not one I would like to conclude myself.
27th April 2018
2017 was an interesting year for CD Surveys Ltd. In general, the construction industry seemed to be rather quiet and with a quiet start to the year…Find out more