GNSS/INS-assisted Multi-camera Mobile Mapping: System Architecture, Modeling, Calibration, and Enhanced Navigation

Magdy Elbahnasawy, Purdue University

Abstract

To date, due to some financial and technical constraints, camera-based mobile mapping systems with GNSS/INS unit onboard the platform are still the optimal option for most mapping applications. Therefore, a great deal of research has been conducted on developing an operational GNSS/INS-assisted single/multi-camera mapping system. However, to satisfy the needs of various mapping applications, there are still challenges that have to be addressed. Such challenges start with the selection and integration framework of sensors (cameras and GNSS/INS units) onboard different types of mapping platforms (i.e., UAV and wheel-based platforms). Another challenge is modeling and calibrating the GNSS/INS-assisted single/multi-camera mapping systems. In addition, maintaining the accuracy of trajectory information in areas with intermittent access to GNSS signals is a key point to have an accurate geo-referenced mapping output. In this regard, this research work focuses on developing a generic framework for integrated GNSS/INS-assisted single/multi-camera mobile mapping platforms, starting from system hardware integration, passing through system modeling, calibration, and ending with extending the system operating range to areas with intermittent access to GNSS signals. More specically, system architecture for hardware integration, synchronization, and testing of GNSS/INS-assisted single/multi-camera system are developed. Such development provides transparency to facilitate: (i) manipulation of sensors placement, (ii) adding and/or removing sensors, (iii) handling different data collection rates, and (iv) using off-the-shelf sensors. Also, a comparative analysis of several GNSS/INS-assisted multi-camera system mathematical models is introduced. Such analysis endorses the selection of the most appropriate model that represents the imaging geometry of GNSS/INS-assisted multi-camera systems. Furthermore, a point-pairing-based calibration strategy for multi-camera mobile mapping systems is devised. The proposed strategy uses control linear and planar features for calibration, which makes this strategy a more flexible calibration approach for indoor, outdoor, airborne, and terrestrial mapping systems. Finally, an image-assisted GNSS/INS navigation technique is proposed for UAV-Based MMS to enhance the trajectory information in areas with intermittent GNSS signals accessibility. The experimental results show a significant enhancement in the position and orientation trajectory information during GNSS outages using the proposed image-assisted GNSS/INS navigation.

Degree

Ph.D.

Advisors

Habib, Purdue University.

Subject Area

Civil engineering

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server
.

Share

COinS