Kay Fürstenberg

Vehicle and object models for robust tracking in traffic scenes using laser range images

Detection and modeling of dynamic traffic scenes around a, driving passenger car is the long-term aim of the research project ARGOS at the University of Ulm. Each object close to the own car should be modeled and tracked using a specific individual dynamic model. The object classification is based on the geometric outlines and the dynamic behavior. For any sensor combinations usable to detect the environment, the velocity of the objects can be measured relatively to the movement of own vehicle. To. get the absolute velocity of the objects, the motion of the own vehicle must be measured for which the well know bicycle model is used. This ego-model is fed by sensor signals provided anyway by ABS, ASR or ESP. To eliminate the own motion from the object measurements, several coordinate transformations are required in the different stages of data processing. A proposal is given on how to solve this problem when using a laser range finder as a sensing device. Moreover, a simple object model is introduced for this task in order to save processing power. The algorithms can extended towards a multihypothesis approach which will result a more robust classification and tracking algorithm.

Object Classification exploiting High Level Maps of Intersections

An object classification system is introduced. The system observes the vehicle’s environment with a laser scanner. Preprocessing and object tracking algorithms are applied. The object classification combines a pattern classifier with rule based a priori knowledge and high level map information. The pattern classifier uses significant features to calculate membership values for each class. These membership values are verified and corrected by a priori knowledge. Furthermore, a precise position of the test vehicle is estimated. The positions of observed objects in the high level map can be determined exploiting this information. As the object position is restricted for some object classes, this knowledge can be used in the classification, which significantly improves its performance. Finally, the system is evaluated with labeled test data of several sequences at different intersections.

Fahrzeugumfelderfassung mit mehrzeiligen Laserscannern (The Use of Multi-Layer Laser Scanners for Monitoring the Environment of Driving Vehicles)

Abstract Es werden Methoden zur Fahrzeugumfelderfassung, basierend auf einem neuartigen mehrzeiligen Laserscanner als Sensor, vorgestellt. Mithilfe eines Sensormodell- und Objektmodell-gestützten Ansatzes zur Messsignalverarbeitung ist es möglich, andere Verkehrsteilnehmer wie Fußgänger, Radfahrer oder Kraftfahrzeuge zu detektieren, zu klassifizieren und in ihrem Bewegungsverhalten zu modellieren. Derartige Algorithmen bilden die Grundlage zum Aufbau eines konsistenten Fahrzeugumfeldmodells, das für zukünftige Sicherheits- und Komfortsysteme im Kraftfahrzeug benötigt wird.