Accurate Global Trajectory Alignment using Poles and Road Markings
This addresses the challenge of low GNSS precision in urban areas for digital maps, improving global map accuracy and sensor calibration, though it is incremental as it builds on existing feature-matching techniques.
The paper tackles the problem of aligning vehicle trajectories to aerial imagery for automated driving by using poles and road markings as features, achieving accurate geo-referencing as demonstrated in evaluations on data from Karlsruhe, Germany.
Currently, digital maps are indispensable for automated driving. However, due to the low precision and reliability of GNSS particularly in urban areas, fusing trajectories of independent recording sessions and different regions is a challenging task. To bypass the flaws from direct incorporation of GNSS measurements for geo-referencing, the usage of aerial imagery seems promising. Furthermore, more accurate geo-referencing improves the global map accuracy and allows to estimate the sensor calibration error. In this paper, we present a novel geo-referencing approach to align trajectories to aerial imagery using poles and road markings. To match extracted features from sensor observations to aerial imagery landmarks robustly, a RANSAC-based matching approach is applied in a sliding window. For that, we assume that the trajectories are roughly referenced to the imagery which can be achieved by rough GNSS measurements from a low-cost GNSS receiver. Finally, we align the initial trajectories precisely to the aerial imagery by minimizing a geometric cost function comprising all determined matches. Evaluations performed on data recorded in Karlsruhe, Germany show that our algorithm yields trajectories which are accurately referenced to the used aerial imagery.