Integrated collision avoidance enhanced GN&C system for air vehicle

• US 20080243383A1
• Filed: 12/04/2007
• Published: 10/02/2008
• Est. Priority Date: 12/12/2006
• Status: Active Grant

• Current Assignee: American GNC Corporation
• Original Assignee: American GNC Corporation
• Inventors: Lin, Ching-Fang
• Granted Patent: US 8,509,965 B2
• Time in Patent Office: Days
• US Class Current: 701/213
• CPC Class Codes: G01C 21/1652 with ranging devices, e.g. LIDAR or RADAR
• CPC Class Codes: G01C 21/1656 with passive imaging devices, e.g. cameras
• CPC Class Codes: G01S 13/933 of aircraft or spacecraft
• CPC Class Codes: G01S 13/935 for terrain-avoidance
• CPC Class Codes: G01S 19/15 Aircraft landing systems
• CPC Class Codes: G08G 5/0078 for monitoring traffic from the aircraft radar or analogous systems specially adapted for traffic control G01S13/91
• CPC Class Codes: G08G 5/0086 for monitoring terrain radar or analogous systems specially adapted for terrain avoidance G01S13/935
• CPC Class Codes: G08G 5/045 Navigation or guidance aids, e.g. determination of anti-collision manoeuvers radar or analogous systems specially adapted for anti-collision between aircraft G01S13/933

Abstract

Collision with ground/water/terrain and midair obstacles is one of the common causes of severe aircraft accidents. The various data from the coremicro AHRS/INS/GPS Integration Unit, terrain data base, and object detection sensors are processed to produce collision warning audio/visual messages and collision detection and avoidance of terrain and obstacles through generation of guidance commands in a closed-loop system. The vision sensors provide more information for the Integrated System, such as, terrain recognition and ranging of terrain and obstacles, which plays an important role to the improvement of the Integrated Collision Avoidance System.

Claims

1. An integrated collision avoidance method, comprising the steps of:

(a) receiving global positioning system signals to an integrated positioning/ground proximity warning system processor, wherein said global positioning system signals is arranged for selectively deriving position, velocity and time information, and is arranged for selectively doing pseudorange and delta range measurement;

(b) receiving vehicle angular rate and specific force information for computing an inertial navigation solution, including position, velocity, and altitude of said vehicle, by solving inertial navigation equations and outputting said inertial navigation solution to said integrated positioning/ground proximity warning processor;

(c) measuring air pressure, and computing barometric measurements which is output to said integrated positioning/ground proximity warning processor;

(d) measuring time delay between transmission and reception a radio signal form a terrain surface, and computing radio altitude measurement which is output to said integrated positioning/ground proximity warning processor;

(e) recognizing terrain and obstacles near said host vehicle for ground proximity warning and obstacle collision avoidance decision;

(f) accessing a terrain database for obtaining current vehicle position and surrounding terrain height data which is output to said integrated positioning/ground proximity warning processor;

(g) receiving said position, velocity and time information or said pseudorange and delta range measurements of said global positioning system, said inertial navigation solution, said radio altitude measurement, said terrain and obstacles recognition information and said current vehicle position and surrounding terrain height data, and computing optimal positioning solution data and optimal ground proximity warning solution data;

2. The method, as recited in claim 1, wherein, in the step (e), said recognizing is based on statistical model which comprises two classification methods containing supervised classification and unsupervised classification.

3. The method, as recited in claim 2, wherein said supervised classification comprises three modules:

system training and machine learning which is trained for recognition and segmentation of natural objects and cultural objects through feature extraction, feature data analysis and statistical modeling;

information system and modeling which consists of data storage of sampled object images and their features, and model optimization tools for feature selection to structure an image classifier; and

objects extraction which comprises an aerial-image objects classifier, natural objects and cultural objects recognition and labeling.

4. The method, as recited in claim 2, wherein said unsupervised classification is using a Fuzzy C-Means (FCM) based method to extract cultural and natural objects from an aerial image.

5. The method, as recited in claim 4, wherein said Fuzzy C-Means (FCM) based method comprises the steps of:

(e1) preprocessing an aerial image for generalization and contrast adjustment;

(e2) extracting multiple features to form a high-dimension feature space;

(e3) inputting said data set of each pixel in said feature space into an FCM classifier to be labeled; and

(e4) applying morphological operations to remove small islands or holes on the segmentation masks generated by said FCM classifier.

6. The method, as recited in claim 1, wherein said step (g) provides information fusion based terrain collision avoidance which comprises the steps of:

(g1) determining said terrain environment by information sources;

(g2) predicting said flight path by said platform altitude, position and velocity;

(g3) predicting said terrain collision based on said flight path prediction and terrain environment; and

(g4) providing visual and voice alert if a terrain collision is possible.

7. The method, as recited in claim 1, wherein said step (g) provides information fusion based midair collision avoidance which comprises the steps of:

(g1′

) predicting said flight path by said platform altitude, position and velocity;

(g2′

) recognizing and tracking midair obstacles;

(g3′

) providing visual and voice alert if a terrain collision is possible; and

(g4′

) designing optimal escaping method.

8. The method as recited in claim 7 wherein, in the step (g2′

), imaging sensor and vision based obstacle recognition method is used; and

radar and radar based detecting and tracking method is used.

9. The method, as recited in claim 3, wherein said step (g) provides information fusion based terrain collision avoidance which comprises the steps of:

(g1) determining said terrain environment by information sources;

(g2) predicting said flight path by said platform altitude, position and velocity;

(g3) predicting said terrain collision based on said flight path prediction and terrain environment; and

(g4) providing visual and voice alert if a terrain collision is possible.

10. The method, as recited in claim 3, wherein said step (g) provides information fusion based midair collision avoidance which comprises the steps of:

(g1′

) predicting said flight path by said platform altitude, position and velocity;

(g2′

) recognizing and tracking midair obstacles;

(g3′

) providing visual and voice alert if a terrain collision is possible; and

(g4′

) designing optimal escaping method.

11. The method, as recited in claim 10, wherein, in the step (g2′

), imaging sensor and vision based obstacle recognition method is used; and

radar and radar based detecting and tracking method is used.

12. The method, as recited in claim 5, wherein said step (g) provides information fusion based terrain collision avoidance which comprises the steps of:

(g1) determining said terrain environment by information sources;

(g2) predicting said flight path by said platform altitude, position and velocity;

(g3) predicting said terrain collision based on said flight path prediction and terrain environment; and

(g4) providing visual and voice alert if a terrain collision is possible.

13. The method, as recited in claim 5, wherein said step (g) provides information fusion based midair collision avoidance which comprises the steps of:

(g1′

) predicting said flight path by said platform altitude, position and velocity;

(g2′

) recognizing and tracking midair obstacles;

(g3′

) providing visual and voice alert if a terrain collision is possible; and

(g4′

) designing optimal escaping method.

14. The method as recited in claim 13 wherein, in the step (g2′

), imaging sensor and vision based obstacle recognition method is used; and

radar and radar based detecting and tracking method is used.

15. An integrated collision avoidance system, comprising:

an imaging sensor for terrain and obstacle recognizing and target tracking;

a global positioning system receiver receiving global positioning system signals to an integrated positioning/ground proximity warning system processor, wherein said global positioning system signals is arranged for selectively deriving position, velocity and time information, and is arranged for selectively doing pseudorange and delta range measurement

an inertial navigation system for solving navigation equations with angular rate and specific force information from an inertial measurement unit and obtaining an inertial navigation solution;

a baro altimeter for providing baro altitude measurements;

a radio altimeter for providing radio altitude measurement from the terrain surface;

a terrain database for providing global terrain data and obtains the terrain height of the host vehicle position and surrounding terrain height data;

an object detection system providing position and dynamic data of near objects;

a vehicle control and management system for providing performance and configuration data of said host vehicle; and

an integrated positioning/ground proximity warning system processor for receiving data from said imaging sensor, said global positioning system receiver, said inertial navigation system, said terrain database, said baro altimeter, said radio altimeter, and said object detection system, wherein said positioning/ground proximity warning system processor also receives vehicle performance and configuration data from a vehicle control and management system and providing optimal position, velocity, altitude navigation information, and an optimal proximity warning information.

16. The integrated collision avoidance system, as recited in claim 15, wherein said imaging sensor recognizes said terrain and obstacles based on statistical model which comprises two classification methods:

supervised classification and unsupervised classification.

17. The integrated collision avoidance system, as recited in claim 16, wherein said imaging sensor comprises a stereo camera to provide range and 3D measurements of said target, which are used for positioning.

18. The integrated collision avoidance system, as recited in claim 17, wherein said stereo camera is fused with other sensors to provide target tracking.

19. The integrated collision avoidance system, as recited in claim 15, wherein also comprises a radar sensor for obstacle detecting and tracking.

20. The integrated collision avoidance system, as recited in claim 18, wherein also comprises a radar sensor for obstacle detecting and tracking.

21. The integrated collision avoidance system, as recited in claim 15, wherein say integrated positioning/ground proximity warning system processor fuses the information form said imaging sensor, said global positioning system receiver, said inertial navigation system, said terrain database, said baro altimeter, said radio altimeter, and said object detection system to provide terrain collision avoidance and midair obstacle collision avoidance.

22. The integrated collision avoidance system, as recited in claim 20, wherein say integrated positioning/ground proximity warning system processor fuses the information form said imaging sensor, said global positioning system receiver, said inertial navigation system, said terrain database, said baro altimeter, said radio altimeter, and said object detection system to provide terrain collision avoidance and midair obstacle collision avoidance.