Many driver assistant and safety systems depend on a correct information about the dynamic states of the host vehicle. The translation and the change of the yaw angle of the host vehicle between two times can be determined using integrated serial sensors, such as yaw rate, steering angle and wheel speed sensors. However, in extreme situations, such as skidding or wheel spin, the integrated sensors are not able to measure the vehicle's real translation and rotation. In order to provide a valid position and orientation information even in these extreme situations, stationary objects, which are used as landmarks, are detected online by an automotive laser scanner using a robust algorithm based on advanced occupancy grids. In this paper a localization approach for skidding situations is proposed. With the help of the online map, stationary objects used as landmarks are determined. The robust TrAss algorithm allows for the association of landmarks to the data of the laser scanner. With the help of the position estimation, several safety and assistant systems can be supported by a valid information about the vehicles movement.
Robust host-localisation in skidding situations using laser scanners
Robuste Fahrzeuglokalisiation in Schleudersituationen mit Hilfe der Laserabtastung
2001
6 Seiten, 7 Bilder, 8 Quellen
Conference paper
English
Landmark Navigation for Robust Object Tracking in Skidding Maneuvers Using Laser Scanners
British Library Conference Proceedings | 2007
|Robust control prevents car skidding
Tema Archive | 1997
|NTIS | 1966
NTIS | 1970