The performance analysis of a portable mobile mapping system with different GNSS processing strategies

Over the years, land vehicular based Mobile Mapping Systems (MMSs) have demonstrated that accuracies suitable for all but the most demanding cadastral and engineering applications can be achieved. Mobile mapping refers to a means of collecting geospatial data using mapping sensors that are mounted on a mobile platform. This result, combined with a reduction in both the time and cost of data collection, made MMS a very interesting technology potentially able to meet the demand of spatial information system operators for rapid spatial data acquisition. The idea of mobile mapping is basically executed by producing more than one image that includes the same object from different positions, and then the 3D positions of the same object with respect to the mapping frame can be measured. Multi-platform and multi-sensor integrated mapping technology has clearly established a trend towards fast geospatial data acquisition. Sensors can be mounted on a variety of platforms, such as satellites, aircraft, helicopters, terrestrial vehicles, water-based vessels, and even people. As a result, mapping has become mobile, and dynamic. Since the early nineties, advances in satellite and inertial technology made it possible to think about mobile mapping in a new way. Instead of using ground control points (GCP) as references for orienting the images in space, the trajectory and orientation of the imager platform can now be determined directly. Cameras, along with positioning and orientation sensors, are integrated and mounted on a land vehicle for mapping purposes. Objects of interest can be directly measured and mapped from images that have been georeferenced using positioning and orientation sensors. The development of land vehicular based mobile mapping systems was initiated by two research groups in North America, The Center for Mapping at Ohio State University, USA, and the Department of Geomatics Engineering at the University of Calgary, Canada to fulfill the need for highway infrastructure mapping and transportation corridor inventories. In the early 2000s, numerous land vehicular based mobile mapping systems have been utilized in commercial applications. Over the last five years, Google has adopted the technology on a large scale, introducing substantial fleets of mobile mapping vehicles for their imaging and mapping operations then producing fruitful spatial information content for added value applications. This has resulted in the further rapid development of the technology which can now be regarded as being well established and proven. However, the high costs involved in the development of such systems limits their growth in the market, so that land vehicular based MMSs are still mainly operated by the companies or institutions that build them. In addition, the primary limitation of such land vehicular based systems in terms of operation flexibility is the dependence of the availability and quality of road networks. Sometime streets in the metropolitan area are too narrow for land vehicle to operate. In addition, at distances greater than 20 to 30 km from a reference station, the residual ppm error caused by the atmosphere delaying. Hence with traditional processing, it is always necessary to be within 30 km of a reference station sometime during the mission in order to resolve the ambiguities. Once the correction are resolved, the PMMs can be operated to about 75 km from the nearest reference station before the magnitude of the ppm error exceeds level required for high-accuracy applications. For land-based applications a significant productivity improvement in Real-Time Kinematic (RTK) positioning has been achieved using the concept of a “Virtual Reference Station” or VRS. The objective of this study is developed a prototype of PMMS to allow a wider community of spatial data user to benefit of mobile mapping applications - in particular the lower costs and greater efficiency of data collection and validation of DG performance between based on traditional method and VRS. The proposed PMMS is composed of a tactical grade Inertial Measurement Unit (IMU), dual frequencies Global Navigation Satellite System (GNSS) receiver, two digital cameras, a centralized power supply as well as data storage unit to avoid the use of computer or laptop. Data processing modules including Position and Orientation System (POS) module, mounting calibration (boresight and lever arm) module and close range photogrammetry module are developed in this study. The proposed mounting calibration method is two-Step that averages the lever-arm and boersight including each epoch, respectively. In addition, different GNSS processing strategies applied for proposed PMMS operation including conventional differential GNSS with a physical reference station, differential GNSS with VRS and Precise Point Positioning (PPP) are investigated in terms of their Direct Georeferencing (DG) performance. Preliminary results presented in this study including two parts as calibration and DG. In the aspect of calibration, the accuracies of lever-arm and boresight are about 1-3 centimeters and less than 0.5 degree through error propagation. In the word, it will cause 4-7 cm error for 5 meters. On the other hand, by comparing DG results of checking points provided by the proposed PMMS from various scenarios with their known coordinates depicts the absolute 3D positioning accuracy can reach less than 10 centimeters when the object distance is around 20 meters based on DGNSS and VRS. With the inclusion of the PPP and VRS GNSS processing strategy, the proposed PMMS can be deployed immediately in a designated area to gather useful geo-referenced spatial information without sending ground crews for GCP surveying as well as arranging master GPS stations for DGNSS processing to gain sufficient accuracy for mapping and other rapid urgent response for disaster relief applications.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.