In 2015, the construction of the Autopista al Mar 1 motorway was awarded to a Consortium of construction companies – STRABAG/Sacyr and Concay. The project comprises of designing, building, financing and operating a 176-km motorway to will link the capital city of Medellin with the Caribbean Sea. The project includes construction of a 75-km new motorway, reconstruction of 35 bridges and tunnels with single length of up to 300 meter, and modernization of a 65-km section of the existing motorway. The complex part of the project included the extension of a mountain road with an 11% steep grade – from 2 to 4 lanes. The total length of expansion included 33-km of the motorway, and 5-km tunnel construction. Core risks associated with the project includes earthquake and tense geological conditions.
Traditional surveying tools such as total stations were found to be time-consuming, and expensive to map the existing infrastructure. Using the traditional tools, it was impossible for the organization to measure the complex terrain of the project especially the points under the bridge, column of the bridge or the beam of the bridge. Therefore, to map the environment (including foundations, pillars, and topographical terrain), the consortium of construction companies turned to drones and software (DJI Inspire 1 Drone and Pix4D software – Pix4Dengine, and Pix4Dcapture). The drone was used to fly over the existing infrastructure – bridges, tunnels, roads, undersides and terrains – and capture 2,000 images per location with a resolution of GSD ~ 1.2 pixels per centimeter. Using photogrammetric mapping, drones were able to easily fly around the existing infrastructure and precarious locations and map inaccessible and high-risk areas with ease and high accuracy.
The consortium used the Pix4Dcapture app to map the bridge foundations and pillars in a free-flight mode. Simultaneously, they placed the Ground Control Points (GCPs) both horizontally and vertically – on road and on the bridge pillars wherein the data from the GCPs were automatically marked. The point cloud data collected through both the sources had ‘inconvenient noises’ and they were added on to Pix4D’s rayCloud solution to edit it and remove the noises. The data was processed with the Pix4D mapping software to generate accurate and highly-detailed 3D models of the existing road infrastructure and existing bridges – georeferenced orthophotos of the surrounding photo. Using the 3D models, the team were able to visualize the construction environment and plan the project with precise measurements. Finally, using the 3D point clouds, the construction company was able to generate the Digital Terrain Models (DTMs). Using the measurements and identifying the material requirement, the bridges were constructed using precast components construction method using the cantilever method. The resulting 3D models were incorporated into CAD/BIM models for subsequent planning and calculation.
The use of drones in the project instead of traditional surveying tools help measure what would otherwise be impossible to measure, save costs and time. The use of Pix4D software streamlined the process, saved manual processing times in integrated automatic features such as automatic GCO marking and improved the productivity and accuracy of the project.