This paper describes a system for detecting and estimating the properties of multiple travel lanes in an urban road network from calibrated video imagery and laser range data acquired by a moving vehicle. The system operates in several stages on multiple processors, fusing detected road markings, obstacles, and curbs into a stable non-parametric estimate of nearby travel lanes. The system incorporates elements of a provided piecewise-linear road network as a weak prior. The authors' method is notable in several respects: it estimates multiple travel lanes; it fuses asynchronous, heterogeneous sensor streams; it handles high-curvature roads; and it makes no assumption about the position or orientation of the vehicle with respect to the road. They analyze the system's performance in the context of the 2007 DARPA Urban Challenge. With five cameras and thirteen lidars, it was incorporated into a closed-loop controller to successfully guide an autonomous vehicle through a 90 km urban course at speeds up to 40 km/h amidst moving traffic. The primary contributions of this paper are: A method for estimating multiple lanes of travel in a typical urban road network using only information from local sensors; - A method for fusing these estimates with a weak prior, such as that derived from a topological road map with sparse metrical information; - Methods for using monocular cameras to detect road markings; - Multi-sensor fusion algorithms combining information from video and lidar sensors. The authors also describe their method's failure modes, and possible directions for future work.
Multi-sensor lane finding in urban road networks
Multisensor-Spurauffinden im Stadtstraßennetz
2009
8 Seiten, 10 Bilder, 1 Tabelle, 18 Quellen
Aufsatz (Konferenz)
Englisch
Adaptive lane finding in road traffic image analysis
IET Digital Library Archive | 1994
|Adaptive lane finding in road traffic image analysis
British Library Conference Proceedings | 1994
|Finding multiple lanes in urban road networks with vision and lidar
British Library Online Contents | 2009
|