1. European Geosciences Union General Assembly 2016 Vienna,Austria 17–22 April 2016 Abu Dhabi Base-map Update Using the LiDAR Mobile Mapping Technology Omar Alshaiba 1, M. Amparo Núñez-Andrés 2, and Nieves Lantada 2 Abu Dhabi Municipal System, Department of Civil and Environmental Engineering, Universitat Politècnica de Catalunya Abstract Mobile LiDAR system provides a new technology which can be used to update geospatial information by direct and rapid data collection. This technology is faster than the traditional survey ways and has lower cost. Abu Dhabi Municipal System aims to update its geospatial system frequently as the government entities have invested heavily in GIS technology and geospatial data to meet the repaid growth in the infrastructure and construction projects in recent years. The Emirate of Abu Dhabi has witnessed a huge growth in infrastructure and construction projects in recent years. Therefore, it is necessary to develop and update its basemap system frequently to meet their own organizational needs. Currently, the traditional ways are used to update base-map system such as human surveyors, GPS receivers and controller (GPS assigned computer). Then the surveyed data are downloaded, edited and reviewed manually before it is merged to the base-map system. Traditional surveying ways may not be applicable in some conditions such as; bad weather, difficult topographic area and boundary area. The study presents a proposed methodology which uses the Mobile LiDAR system to update basemap in Abu Dhabi by using daily transactions services. It aims to use and integrate the mobile LiDAR technology into the municipality’s daily workflow such that it becomes the new standard cost efficiency operating procedure for updating the base-map in Abu Dhabi Municipal System. On another note, the paper will demonstrate the results of the innovated workflow for the base-map update using the mobile LiDAR point cloud and few processing algorithms. Study Area and Mission Analysis Al Hili area (district in Al Ain City Municipality, Abu Dhabi Emirate, UAE) was selected to be the area under studying since it is a border area and it has variety of different ground features, and has different visibility for the sky i.e. obstacles, trees, and high buildings so that the accuracy will be fluctuating accordingly. it was confirmed that the area will meet all requirements such as; having enough information about this area that required to obtain the project deliverables which needed for this research, data acquisition process can be done within the research time frame, size of project is suitable. It is necessary that all information about this area being available before the survey such as; (referenced geodetic system, utility line location, as-built drawings of constructions inside the area (that has the advantage of excluding or using less land surveying or repeating the survey in the site). The study Proposed Workflow The drawback of the manual digitization is the longtime of work to digitize even small patches of data and it requires a lot of team members to digitize all the required layers i.e. roads, light poles, walls, buildings etc. as a response of this drawback to save the time and cost of the process, this study will present an automated feature extraction techniques to map all the street furniture, assets, and roads with relevant high accuracy and success rate. It is very obvious that any dataset shall involve tricky texture of surface which the algorithm could fail to digitize it properly and that’s one of the automatic feature extraction especially in a 3D environment like the point- cloud. Data-Processing output and Base-map update Examples on the feature extraction algorithm: Points Classifications: this step is intended to classify the point cloud into multiple types i.e. ground, vegetation, buildings…etc. Multivariate-Filters: The ground level points are applied to multiple filters consequently for example: apply band bath filter on intensity, then apply filter on the scan angle, then on the elevation for the points. The result of the multivariate filtering could be the required feature to be extracted. For more clarification, if the intensity of the road paint is known we can extract the all the road paint, and the using the scan angle and the elevation filter we might cut the unimportant features. Post-Processing Trajectory Accuracy Analysis Fortunately, the solution status is almost fixed at all the valid mission and that means after the running the post processing procedure, all observations have been recovered with no issues. Minimum number of satellite is acceptable at all missions except a short duration in mission 5 due to the same reason of high buildings and obstacles. All the baselines of the missions are acceptable from the best practices records which recommend 10 km maximum for the baseline length over the entire mission. On another note, the single base station post-processing improved all the RMSEs for all the positions and orientations. The horizontal accuracy i.e. Easting and Northing have been improved from 2 meters to 1 cm, as for the vertical accuracy has been improved from 5 meters and above to less than 10 cm. On the level of orientation, the roll and pitch angles accuracy have been improved from 5 arc- minutes to less than 1 arc-minute; the heading was slightly improved by 10 to 20 cm because it’s already adjusted using two GPS antennas on the vehicle. Table 6 summarizes the trajectory solutions for the four missions. Real time errorPost processing error Area of Interest (Unclassified point cloud) Road Edges extraction Object Detection Output (Walls, Poles, and street signs) Vector data ready for base-map update Challenges and Difficulties 1.Collecting permission to collect the data 2.Harsh environment in terms of (temperature/ humidity/ others) 3.Big data size 4.Repairing Trajectory Data in the post processing 5.Shadowing nature (Unseen data) Conclusion This study has been tested am automated method of using mobile LiDAR system for updating the base-map. After data analysis and processing, it was clear that Mobile LiDAR is much more effective than the traditional survey ways. The Relative Accuracy is incredibly high (Millimeters). The Absolute Accuracy from the mobile LiDAR is 5-6 cm depending on the quality of the base stations data. Author Contacts Omar Al Shaiba – email: omar.alshaiba@hotmail.com – cell: +97150 448 6616omar.alshaiba@hotmail.com