TR2023-074
Bayesian Sensor Fusion for Joint Vehicle Localization and Road Mapping Using Onboard Sensors
-
- "Bayesian Sensor Fusion for Joint Vehicle Localization and Road Mapping Using Onboard Sensors", International Conference on Information Fusion (FUSION), DOI: 10.23919/FUSION52260.2023.10224204, June 2023, pp. 1-8.BibTeX TR2023-074 PDF
- @inproceedings{Berntorp2023jun,
- author = {Berntorp, Karl and Greiff, Marcus and Di Cairano, Stefano and Miraldo, Pedro},
- title = {Bayesian Sensor Fusion for Joint Vehicle Localization and Road Mapping Using Onboard Sensors},
- booktitle = {International Conference on Information Fusion (FUSION)},
- year = 2023,
- pages = {1--8},
- month = jun,
- publisher = {IEEE},
- doi = {10.23919/FUSION52260.2023.10224204},
- isbn = {979-8-89034-485-4},
- url = {https://www.merl.com/publications/TR2023-074}
- }
,
- "Bayesian Sensor Fusion for Joint Vehicle Localization and Road Mapping Using Onboard Sensors", International Conference on Information Fusion (FUSION), DOI: 10.23919/FUSION52260.2023.10224204, June 2023, pp. 1-8.
-
MERL Contacts:
-
Research Areas:
Abstract:
We propose a method for joint estimation of a host vehicle state and a map of the road based on global navigation satellite system (GNSS) and camera measurements. We model the road using a spline representation described by a parameter vector having a Gaussian prior representing the uncertainty of the prior map. Both GNSS and camera measurements, such as lane-mark measurements, have noise characteristics that vary in time. To adapt to the changing noise levels and hence improve positioning performance, we combine the sensor information in an interacting multiple-model (IMM) setting to choose the best combination of the estimators with the vehicle state and the parameter vector of the map as the state vector. In a simulation study, we compare vehicle models with varying complexity, and on a real road segment we show that the proposed method can accurately adjust to changing noise conditions and correct for errors in the prior map.