Abstract: | In the past years computer vision techniques are getting wide spread in numerous applications. Therefore, increased research and development activities were performed in order to create algorithms for optical tracking of various landing surfaces. Several applications already were implemented in robotic and automotive application including obstacle detection and positioning. Only some research activities were conducted in airborne application. However none of these attempts have reached the status of a certified airborne navigation system. Apart from that unmanned airborne applications using computer vision deal with distances much smaller than those in aircraft landing scenarios. Unmanned aerial vehicles track obstacles in distances not greater than several hundred meters. In aviation application certified SBAS LPV 200 approaches allow ground proximities down to 200 ft decision height by autopilot. Pilots have to take control of the aircraft by manual flight from decision height down to the landing. An innovative approach is to use an airborne computer vision application to overcome the gap between current decision height minima and touch down point by vision aided automatic landing systems realized by increased navigation performance even though facing visual distances more than several thousand meters. The Institute of Flight Guidance, of the Technische Universität Braunschweig runs the joined research project C2Land. Its main goal is the development of a visionaided automatic landing system. The landing system uses a navigation system without further ground based infrastructure for navigation and an optical sensor system to recognize environment and target structures. In order to execute an automatic landing, present attempts of optic sensor coupling have to be extended. Future target is an automatic landing system especially for concrete runways without usage of the traditional Instrument Landing System (ILS). Therefore, performances have to be reached that meet aviation standards in terms of integrity, accuracy, availability and continuity. This paper introduces the hardware architecture of overall custom-build system approach. Furthermore, the paper outlines the principles how to extract and to apply filter on the optical raw data in order to give a lateral and vertical guidance relative to the desired runway touchdown point. This filter algorithm is an innovative approach dealing with misleading information of the optical system and how to distinguish and evaluate the proper information with the usage of integrated INSGNSS state vector. The algorithms have been evaluated not just by a self-developed simulation environment but also by first real conducted flight trials in on-time processing. |
Published in: |
Proceedings of the ION 2015 Pacific PNT Meeting April 20 - 23, 2015 Marriott Waikiki Beach Resort & Spa Honolulu, Hawaii |
Pages: | 694 - 701 |
Cite this article: | Angermann, M., Wolkow, S., Schwithal, A., Tonhäuser, C., Hecker, P., "High Precision Approaches Enabled by an Optical-Based Navigation System," Proceedings of the ION 2015 Pacific PNT Meeting, Honolulu, Hawaii, April 2015, pp. 694-701. |
Full Paper: |
ION Members/Non-Members: 1 Download Credit
Sign In |