Download Autonome Mobile Systeme 2009: 21. Fachgespräch Karlsruhe, by Andreas Koch, Adam Berthelot, Bernd Eckstein, Oliver PDF

By Andreas Koch, Adam Berthelot, Bernd Eckstein, Oliver Zweigle, Kai Häussermann (auth.), Rüdiger Dillmann, Jürgen Beyerer, Christoph Stiller, J. Marius Zöllner, Tobias Gindele (eds.)

Das 21. Fachgespräch Autonome cellular Systeme (AMS 2009) ist ein discussion board, das Wissenschaftlerinnen und Wissenschaftlern aus Forschung und Industrie, die auf dem Gebiet der autonomen mobilen Systeme arbeiten, eine foundation für den Gedankenaustausch bietet und wissenschaftliche Diskussionen sowie Kooperationen auf diesem Forschungsgebiet fördert bzw. initiiert. Inhaltlich finden sich ausgewählte Beiträge zu den Themen Humanoide Roboter und Flugmaschinen, Perzeption und Sensorik, Kartierung und Lokalisation, Regelung, Navigation, Lernverfahren, Systemarchitekturen sowie der Anwendung von autonomen mobilen Systemen.

Show description

Read or Download Autonome Mobile Systeme 2009: 21. Fachgespräch Karlsruhe, 3./4. Dezember 2009 PDF

Similar german_12 books

Additional info for Autonome Mobile Systeme 2009: 21. Fachgespräch Karlsruhe, 3./4. Dezember 2009

Example text

However, using our visual approach these obstacle can be detected safely. In Table 1 we try to quantify this result. For each obstacle, we have manually labeled those parts of the outline that are relevant for navigation and obstacle avoidance during the above test run using the ground truth map. edu/bundler/ Monocular Obstacle Detection for Real-World Environments 39 approach, the laser range finder and a combination of vision and laser. These results show that major parts of the above mentioned obstacles can be detected.

Hong, T. Chang, and M. Shneier, “Color model-based real-time learning for road following,” IEEE Intelligent Transportation Systems Conference, 2006. 7. J. -H. Nagel, “Texture-based segmentation of road images,” IEEE Symposium on Intelligent Vehicles, pp. 260–265, 1994. 8. E. D. Dickmanns and B. D. Mysliwetz, “Recursive 3-d road and relative ego-state recognition,” IEEE Transaction on Pattern Analysis and Machine Intelligence, vol. 14, pp. 199–213, February 1992. de/is/ Abstract. Robust time-to-contact (TTC) calculation belongs to the most desirable techniques in the field of autonomous robot navigation.

For performance reasons we use the "FAST' comer detector [11] since it is superior to SIFT or SURF in terms of computation time. The selected features are then tracked in subsequent frames while recovering their 3D positions . Davison et al. [6,7] use a single EKF for full covariance SLAM that is able to handle up to 100 features. As we require a denser reconstruction of the scene for obstacle detection, we have to cope with a large number of features which cannot be handled by such an approach in real-time.

Download PDF sample

Rated 4.30 of 5 – based on 3 votes