Geo-Localization Based on Dynamically Weighted Factor-Graph
This work addresses geo-localization for autonomous vehicles in ambiguous environments, representing an incremental improvement over existing methods.
The paper tackles the problem of poor geo-localization due to ambiguous or sparse feature representations by introducing a dynamically weighted factor graph model that adjusts weights based on LiDAR detection information and GNSS-based error estimation. It demonstrates mitigation of outliers and deviations where state-of-the-art methods fail in challenging environments with detection losses.
Feature-based geo-localization relies on associating features extracted from aerial imagery with those detected by the vehicle's sensors. This requires that the type of landmarks must be observable from both sources. This lack of variety of feature types generates poor representations that lead to outliers and deviations produced by ambiguities and lack of detections, respectively. To mitigate these drawbacks, in this paper, we present a dynamically weighted factor graph model for the vehicle's trajectory estimation. The weight adjustment in this implementation depends on information quantification in the detections performed using a LiDAR sensor. Also, a prior (GNSS-based) error estimation is included in the model. Then, when the representation becomes ambiguous or sparse, the weights are dynamically adjusted to rely on the corrected prior trajectory, mitigating outliers and deviations in this way. We compare our method against state-of-the-art geo-localization ones in a challenging and ambiguous environment, where we also cause detection losses. We demonstrate mitigation of the mentioned drawbacks where the other methods fail.