Thorsten Weiss

Robust Driving Path Detection in Urban and Highway Scenarios Using a Laser Scanner and Online Occupancy Grids

Many driver assistant and safety systems depend on an accurate environmental model containing the positions of stationary objects, the states of dynamic objects and information about valid driving corridors. Therefore, a robust differentiation between moving and stationary objects is required. This is challenging for laser scanners, because these sensors are not able to measure the velocity of objects directly. Therefore, an advanced occupancy grid approach, the online map, is introduced, which enables the robust separation of moving and stationary objects. The online map is used for the robust detection of the road boundaries for the determination of driving corridors in urban and highway scenarios. An algorithm for the detection of arbitrary moving objects using the online map is proposed.

Precise ego-localization in urban areas using Laserscanner and high accuracy feature maps

Robust ego-localization is an essential technology for future intelligent vehicles and cooperative applications. In this paper a new localization algorithm based on IBEO AS Laserscanners and high accuracy digital maps is proposed. Algorithms to create accurate grid maps with Laserscanners and the extraction of static objects used as landmarks for ego-localization is introduced. The key problem in landmark navigation in urban areas is the localization of landmarks in distance profiles of a Laserscanner. A fast algorithm is presented, that associates landmarks with data of a Laserscanner which is robust against large rotational, and translational position errors. A position correction algorithm determines the vehicles ego position in WGS-84 coordinates also often used by GPS and navigation maps.

Automatic Detection of Traffic Infrastructure Objects for the Rapid Generation of Detailed Digital Maps using Laser Scanners

Detailed digital maps are of benefit for navigation systems and also for future driver assistant and safety applications. The generation of detailed digital maps is an expensive and time-consuming process as there is much manual rework. In this work, algorithms for the automatic detection of traffic infrastructure objects are proposed. The accurate positions of lane markings, sidewalks, reflection posts and guardrails are determined automatically by a vertically and a horizontally mounted automotive laser scanner. The high position accuracy of the mapped traffic infrastructure objects allows for the rapid generation of up to date, accurate and detailed digital infrastructure maps with a cost efficient sensor setup and a low demand for manual rework.

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.

Automatic detection and classification of safety barriers in road construction sites using a laser scanner

Lateral control of trucks is a challenging task, especially in demanding environments such as road construction sites as the lane width is reduced. Detailed and highly accurate maps, which contain traffic infrastructure objects, will provide important information for a lane keeping assistant for heavy duty vehicles. This paper presents a new approach for the automatic detection and classification of safety barriers in road construction sites using a laser scanner. This data then is used for the generation of a detailed road work map.

Improved object classification of laserscanner measurements at intersections using precise high level maps

This paper deals with real-time object classification at intersection scenarios. Objects are observed using a multilayer laserscanner. The classification is performed using well-known methods of statistical learning. The statistical classification is corrected by rule based a priori knowledge. Precise high level maps provide the possibility to additionally improve the classification by using infrastructure information and the position of the objects in the scene. Classification results of several neural networks and support vector machines are described. Finally, the improvement by high level maps and the final system performance are presented.

A Scalable Sensor Fusion Framework for the Localization of a Vehicle on Detailed Digital Maps Using Laserscanners

Robust ego-localization is an essential technology for future intelligent vehicles and cooperative systems. In this paper a multi sensor fusion framework for the precise estimation of the position, the orientation as well as the velocity and the yaw rate of a vehicle is proposed. Therefore, the data of different sensor systems is combined, such as laser scanners, high accuracy digital maps, GPS and the on-board sensors of the vehicle. In order to improve the accuracy of the position and the orientation estimation, stationary objects, such as posts of traffic lights and traffic signs as well as house walls are registered in detailed digital maps and are used as landmarks, which are detected by an automotive laser scanner. With known positions of the landmarks in the laser scan, the position and orientation of the vehicle are precisely estimated. The scalable architecture also supports any other position measurement systems

Automatic Generation of High Precision Feature Maps of Road Construction Sites

Road construction sites on highways are a demanding environment for drivers as the lane width is reduced. Especially for trucks, lateral control is a challenge. If the driver slightly drives over the lane markings, other vehicles cannot use the neighbouring lane. This leads to a reduction of the road capacity, thus causing traffic jams. A driver assistance system which supports the driver in the task of lateral control highly benefits from the use of accurate infrastructure (feature) maps. This paper presents an approach for the automatic detection, classification, and mapping of specific elements of road works like guard rails, safety barriers, traffic pylons, and beacons with laser scanners.

Positionierung eines Fahrzeugs in unbekannten Gebieten mit Hilfe von Laserscannern Localization in Unknown Environments Using Laser Scanner

Zusammenfassung Viele Komfort- und Sicherheitsapplikationen benötigen eine präzise Information über die Eigenbewegung des Fahrzeugs. Aus den Daten von serienmäßig in heutigen Fahrzeugen verbauten Sensoren kann die Eigenbewegung bestimmt werden. Im Rahmen dieser Arbeit werden Algorithmen vorgestellt, die die Genauigkeit der Eigenbewegungsbestimmung mit Hilfe eines Laserscanners verbessern. Ein rasterbasierter und ein merkmalsbasierter SLAM Algorithmus für den robusten Einsatz im realen Straßenverkehr erlaubt die präzise Eigenbewegungsbestimmung in Innenstadt-, Landstraßen und Autobahnszenarien und auch in extremen Fahrmanövern wie Schleudern oder Drift.