Lidar-aided Instantaneous GNSS Ambiguity Resolution in Challenging Environments: Theoretical Assessment and Numerical Results

Abstract

Facilitating modern driving systems requires accurate vehicle positioning, especially in densely built-up urban environments. Despite being a fundamental positioning technology, which provides globally referenced solutions, Global Navigation Satellite System (GNSS) is known for suffering from signal blockage and multipath effects in urban canyons. Moreover, high-precision GNSS positioning requires the employment of ambiguity-resolved carrier phase measurements from a modest number of satellites, which can be challenging to access in urban environments. Hence, complimentary sensors such as Light Detection and Ranging (lidar) devices are often needed in addition to enable accurate vehicle positioning. In this paper, a lidar-aided instantaneous ambiguity resolution method is proposed. Lidar measurements are generated via a learning-based key-point extraction strategy by registering point clouds to a pre-built HD map of georeferenced scans. Such measurements are tightly coupled with double-differenced GNSS observations in a mixed measurement model to obtain precise float ambiguities. Integer ambiguity resolution is then applied to achieve accurate fixed positioning solutions. Furthermore, a closed-form Ambiguity Dilution of Precision (ADOP) expression is developed to assess and predict the ambiguity resolution performance of the proposed method using the numbers and precision of GNSS and lidar measurements. Theoretical analysis and experimental results show that the proposed lidar-aided instantaneous ambiguity resolution method can significantly improve the ambiguity resolution performance and positioning accuracy comparing with the GNSS-only approach, while also making single-system single-frequency integer ambiguity resolution feasible.