Fabian Poggenhans (M.Sc.)

Wissenschaftlicher Mitarbeiter

Werdegang

Fabian Poggenhans studierte vom Oktober 2008 bis September 2014 Maschinenbau am Karlsruher Institut für Technologie. Seit Oktober 2014 ist er am FZI als Wissenschaftlicher Mitarbeiter im Bereich Mobile Preception Systems (MPS) tätig.

Schwerpunkte seiner Arbeit sind die Gernerierung hochgenauer Karten in urbaner Umgebung und die Erkennung, Klassifikation und Lokalisierung mit Hilfe von kartierungsrelevanten Merkmalen.

Publikationen

Zeitungs- oder Zeitschriftenartikel (3)

1. Application of Line Clustering Algorithms for Improving Road Feature DetectionInfoDetails

   Fabian Poggenhans, André-Marcel Hellmund and Christoph Stiller, 2016

   Although many algorithms have been proposed for the camera-based detection of road features (such as road markings, curbstones and road borders), truly contextual or relational information between the detections is rarely used. This is all the more surprising, since a lot of potential remains unused, regarding outlier rejection or compensating detection failures, multiple detections, misclassification or fragmentation. The aim of this paper is to present an approach that is suitable for such a task in both online and offline applications as a post-processing step after the actual detection and classification step. This is achieved by adapting a perception-based line-clustering algorithm that groups the pre-classified road features based on their relations and assigns them a final class. The grouped features are then fused to form continuous lines instead of individual dashes or fragmented lines. The evaluation on a 10 km drive in both rural and urban environment, as well as an online test on a short highway driving sequence shows that this approach is very well capable to increase the performance of road feature detection at a low computational cost.

2. A Universal Approach to Detect and Classify Road Surface MarkingsInfoDetails

   Fabian Poggenhans and Markus Schreiber and Christoph Stiller, 2015

   In autonomous driving, road markings are an essential element for high-precision mapping, trajectory planning and can provide important information for localization. This paper presents an approach to detect, classify and approximate a great variety of road markings using a stereoscopic camera system. We present an algorithm that is able to classify characters and arrows as well as stop-lines, pedestrian crossings, dashed and straight lines, etc. The classification is independent of orientation, position or the exact shape. This is achieved using a histogram of the marking width as main part of the feature vector for line-shaped markings and Optical Character Recognition (OCR) for characters. Classification is done by an Artificial Neural Network (ANN). We have evaluated our approach over a 10.5 km drive through an urban area.

3. Detecting symbols on road surface for mapping and localization using OCR InfoDetails

   Markus Schreiber and Fabian Poggenhans and Christoph Stiller, 2014

   In this paper, we present a system to detect symbols on roads (e.g. arrows, speed limits, bus lanes and other pictograms) with a common monoscopic or stereoscopic camera system. No manual labeling of images is necessary since the exact definitions of the symbols in the legal instructions for road paintings are used. With those vector graphics an Optical Character Recognition (OCR) System is trained. If only a monoscopic camera is used, the vanishing point is estimated and an inverse perspective transformation is applied to obtain a distortion free top-view. In case of the stereoscopic camera setup, the 3D reconstruction is projected to a ground plane. TESSERACT, a common OCR system is used to classify the symbols. If odometry or position information is available, a spatial filtering and mapping is possible. The obtained information can be used on one side to improve localization, on the other side to provide further information for planning or generation of planning maps.

Konferenzbeitrag (2)

1. Lanelet2: A high-definition map framework for the future of automated drivingInfoDetails

   Fabian Poggenhans and Pauls, Jan-Hendrik and Janosovits, Johannes and Stefan Orf and Maximilian Naumann and Florian Kuhnt and Matthias Mayr, 2018

   Although accurate and comprehensive maps are indispensable for highly automated driving, especially in complex urban scenarios, there are hardly any publications in which requirements for these maps are discussed. In our opinion, such maps must meet high demands in terms of accuracy, completeness, verifiability and extensibility, so that the resulting complexity can only be handled by an enclosing, carefully designed software framework. In this paper we therefore introduce the open-source map framework Lanelet2 implemented in C++ and explain the underlying concept. The goal of Lanelet2 is not only to be usable for typical, isolated applications such as localization or motion planning, but for various potential applications of maps for highly automated driving. On the basis of both abstract and real examples we show the concrete structure of Lanelet2 maps and its use for automated driving.

2. Precise Localization in High-Definition Road Maps for Urban RegionsInfoDetails

   Poggenhans, Fabian and Salscheider, Niels and Stiller, Christoph, 2018

   The future of automated driving in urban areas will most probably rely on highly accurate road maps. However, the necessary precision of a localization in such maps has so far only been reached using extra, sensor specific feature layers for localization. In this paper we want to show that it is possible to achieve sufficient accuracy without a separate localization layer. Instead, elements are used that are already contained in high-resolution road maps, such as markings and road borders. For this, we introduce a modular approach in which detections from different detection algorithms are associated with elements in the map and then fused to an absolute pose using an Unscented Kalman Filter. We evaluate our approach using a sensor setup that employs a stereo camera, vehicle odometry and a low-cost GNSS module on a 5km test route covering both narrow urban roads and multi-lane main roads under varying weather conditions. The results show that this approach is capable to be used for highly automated driving, showing an accuracy of 0.08m in typical road scenarios and a is available 98% of the time.

Export Suchergebnis .bib

Kontakt

Telefon: +49 721 9654-260
E-Mail: Poggenhans@dont-want-spam.fzi.de

Zurück