초록 ▼

A method of guiding an aircraft in the final approach phase including imaging the ground using an imaging system onboard the aircraft to obtain an image, analyzing the image to detect an airport area, detecting an airport area and selecting a runway for landing, tracking the selected runway during t

A method of guiding an aircraft in the final approach phase including imaging the ground using an imaging system onboard the aircraft to obtain an image, analyzing the image to detect an airport area, detecting an airport area and selecting a runway for landing, tracking the selected runway during the approach of the aircraft by visually tracking the selected runway in the image supplied by the imaging system, and tracking the runway when the aircraft is very close to the ground by analyzing the image supplied by the imaging system to determine the position of the aircraft relative to the runway.

대표청구항 ▼

1. A method of guiding an aircraft in the final approach phase of a landing, said method of guiding comprising the steps of:a) imaging the ground using only an onboard imaging system to obtain an image; b) analyzing said image to define an airport area; c) detecting said airport area and repeating s

1. A method of guiding an aircraft in the final approach phase of a landing, said method of guiding comprising the steps of:a) imaging the ground using only an onboard imaging system to obtain an image; b) analyzing said image to define an airport area; c) detecting said airport area and repeating said detecting step until a runway is selected for landing; d) tracking said selected runway during the approach of said aircraft by visually tracking said selected runway in said image supplied by said onboard imaging system; and e) tracking said runway when said aircraft is very close to the ground by analyzing said image supplied by said onboard imaging system to determine the position of said aircraft relative to said runway. 2. The method as claimed in claim 1 wherein a reinitialization consisting of a return to step b) is effected if said runway is lost during tracking step d).3. The method as claimed in claim 1 wherein said analyzing step b) is conducted by searching said image for substantially parallel straight line segments to identify said runway by pairing such straight line segments.4. The method as claimed in claim 3 wherein said substantially parallel straight line segments are searched for by regularly dividing the range of all possible directions into classes and associating each segment with a class whereby two segments are considered to be parallel if they belong to the same class.5. The method as claimed in claim 4 wherein two parallel segments are retained if the following conditions are satisfied:said two segments belong to the same class of directions; each of said two segments has a minimum length; said two segments are face-to-face with a minimum overlap; and the distance between said two segments is less than a predetermined value. 6. The method as claimed in claim 5 wherein a vanishing point consisting of the theoretical point of intersection of two substantially parallel straight line segments is determined for each pair of parallel straight line segments and two pairs of straight line segments are grouped together if the distance between the two corresponding vanishing points is less than a predetermined threshold.7. The method as claimed in claim 1 wherein said runway tracking step d) further comprises the steps of:d1) initialization: the selected runway is extracted from the said image and modeled and a window in which said runway has been found is defined; d2) prediction: the position of said runway in the current image is predicted as a function of the position occupied in the preceding image and a corresponding window is defined; d3) extraction of characteristics: a suitable representation of contours and/or points of interest of said window is supplied; d4) pairing: said runway is searched for in said window as a function of said prediction; and d5) if necessary, said representation of said runway and/or the speed of apparent movement of said runway is updated; and said steps d2) to d5) are executed iteratively. 8. The method as claimed in claim 1 wherein said runway tracking set forth in step e) is conducted by analyzing the color and texture of the pixels of the current image in order to classify said pixels as belonging to said runway or not.9. The method as claimed in claim 1 wherein said step of tracking said runway in step e) further comprises the step of searching said image for the edge of said runway and/or markings on said runway.10. The method as claimed in claim 1 wherein the change from step d) to step e) occurs if a predetermined condition is satisfied.11. The method as claimed in claim 10 wherein said runway has two parallel longitudinal edges connected at one end by a transverse threshold and said change from step d) to step e) occurs when said threshold is no longer visible on said image.12. A system from guiding an aircraft in the final approach phase, including a computer and a display in the cockpit of said aircraft and a video camera connected to said computer, which has means for acquiring and processing video images and for implementing a method as claimed in claim 1 and is connected to said display in said cockpit.13. The system claimed in claim 12 wherein said video camera is a charge coupled device video camera.14. The system claimed in claim 12 wherein said video camera is connected to said computer via an optical fiber link.15. The system claimed in claim 12 including two video cameras.16. The system claimed in claim 15 wherein each video camera is connected to said computer via an optical fiber link.17. The system claimed in claim 12 wherein said computer is connected to an automatic pilot.18. An aircraft including a guidance system as claimed in claim 12.