Robust position estimation using images from an uncalibrated camera

Doehler, Hans-Ullrich; Korn, Bernd

doi:10.1109/dasc.2003.1245917

Cited by 12 publications

(6 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Secondly, we note that sophisticated technologies for conventional (non-neural-based) fusion, ground-object-acquisition and hazard detection are also being developed by the present team 11 11, collaborating colleagues [18,20], and others. A hybridization of AE and conventional algorithms, on a local-area image basis, will be investigated on OUT program, with the goal of achieving the most efficient processing architecture.…”

Section: D4-7mentioning

confidence: 98%

Neural net based processor for robust, high-integrity multisensor and synthetic vision fusion

Kerr¹,

Luk

Hammerstrom

et al. 2003

22nd Digital Avionics Systems Conference Proceedings (Cat No 03CH37449) DASC-03

View full text Add to dashboard Cite

Section: D4-7mentioning

confidence: 98%

Neural net based processor for robust, high-integrity multisensor and synthetic vision fusion

Kerr¹,

Luk

Hammerstrom

et al. 2003

22nd Digital Avionics Systems Conference Proceedings (Cat No 03CH37449) DASC-03

View full text Add to dashboard Cite

“…The fusion process for extracting the aircraft's position relative to the runway threshold as well as the automatic detection of runway structures in both radar images and infrared images have been described in detail in previous publications [3][5][6] [7]. In the following section, a method for MMW radar based navigation is described using additional features (in particular shadows) which allow for a position determination before the runway is in the scanning range of the radar sensor.…”

Section: The Advise-pro Systemmentioning

confidence: 99%

“…Significant structures, e.g. like the runway itself, are extracted from the sensor data (automatically by means of "machine vision") and checked whether they match with the navigation data in combination with database information [7][5] [3]. Furthermore, the sensor images are analyzed to detect obstacles on the runway [4].…”

Section: Introductionmentioning

confidence: 99%

"A System is More Than the Sum of Its Parts" - Conclusion of DLR'S Enhanced Vision Project "ADVISE-PRO"

Korn¹,

Döhler²

2006

2006 Ieee/Aiaa 25TH Digital Avionics Systems Conference

View full text Add to dashboard Cite

Within its research project ADVISE-PRO (Advanced visual system for situation awareness enhancement -prototype, 2003 -2006) that will be presented in this contribution, DLR has combined elements of Enhanced Vision and Synthetic Vision to one integrated system to allow all low visibility operations independently from the infrastructure on ground. The core element of this system is the adequate fusion of all information that is available on-board. This fusion process is organized in a hierarchical manner. The most important subsystems are a) the sensor based navigation which determines the aircraft's position relative to the runway by automatically analyzing sensor data (MMW, IR, radar altimeter) without using neither (D)GPS nor precise knowledge about the airport geometry, b) an integrity monitoring of navigation data and terrain data which verifies on-board navigation data ((D)GPS + INS) with sensor data (MMW-Radar, IR-Sensor, Radar altimeter) and airport / terrain databases, c) an obstacle detection system and finally d) a consistent description of situation and respective HMI for the pilot.

show abstract

“…With our colleagues at DLR (German Aerospace Labs), 9,10 we are working on applications of EVS imagery which go well beyond pilot displays, into the realm of machine vision. In conjunction with the paradigm of "verified SVS" in an integrated EVS-SVS system, demonstrated functionality includes • real-time correlation of a terrain/cultural database with an EVS image • runway acquisition with separate-thread generation of an accurate, 3D navigation signal (including a virtual ILS) • correction of residual (attitude) registration errors between EVS and displayed SVS • automated hazard detection .…”

Section: Advanced Implementationsmentioning

confidence: 99%

Runway infrared range concept for EVS

Kerr¹

2008

Enhanced and Synthetic Vision 2008

View full text Add to dashboard Cite

A major problem in obtaining FAA approval for infrared EVS-based operations under poor-visibility conditions is the lack of correlation between runway visible range and IR attenuation or range. The "IR advantage" in fog, although often substantial, varies greatly as a function of detailed droplet-size distribution. Real-time knowledge of the IR extinction at a given destination is key to reliable operations with lower decision heights. We propose the concept of a Runway Infrared Range (RIRR), to be measured by a ground-based IR transmissometer. Although RVR determination now utilizes single-point scatterometry, the very (Mie) scattering mechanism that often affords a significant IR range advantage necessitates a return to two-point transmissometry. As an adaptation of RVR technology, RIRR will include separate determinations of background-scene and runway/approach lights ranges, respectively. The latter algorithm, known as Allard's Law, will encompass background level, light-settings, visible extinction, and camera performance (usually at short-wave IR). The assumptions and validity of this RIRR parallel those for the traditional RVR. Also, through extended monitoring at a hub, the RIRR may be inexpensively surveyed throughout a fog season, thus predicting the economic benefits of IR-based EVS for that site.

show abstract

Robust position estimation using images from an uncalibrated camera

Cited by 12 publications

References 4 publications

Neural net based processor for robust, high-integrity multisensor and synthetic vision fusion

Neural net based processor for robust, high-integrity multisensor and synthetic vision fusion

"A System is More Than the Sum of Its Parts" - Conclusion of DLR'S Enhanced Vision Project "ADVISE-PRO"

Runway infrared range concept for EVS

Contact Info

Product

Resources

About