SYSTEM AND METHOD FOR PREVENTING A COLLISION

• US 20110160950A1
• Filed: 07/15/2009
• Published: 06/30/2011
• Est. Priority Date: 07/15/2008
• Status: Active Grant

• Current Assignee: Airbus Defence and Space GmbH
• Original Assignee: Airbus Defence and Space GmbH
• Inventors: Langtaler, Peter, Naderhirn, Michael
• Granted Patent: US 8,825,266 B2
• Time in Patent Office: Days
• US Class Current: 701/28
• CPC Class Codes: G05D 1/0202 specially adapted to aircraft
• CPC Class Codes: G05D 1/106 Change initiated in response to external conditions, e.g. avoidance of elevated terrain or of no-fly zones
• 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

Controlling a vehicle to automatically prevent the vehicle from colliding with obstacles includes identifying and locating obstacles in front of the vehicle, wherein relative position of an obstacle is determined; measuring relative speed of the obstacles; assessing whether there is a risk of a collision between the vehicle and the obstacle as a function of the relative position and relative speed. If there is a risk of a collision, the method further includes: calculating an unsafe region around the obstacle as a function of known measurement errors; calculating an evasion point within or on an edge of the unsafe region; defining a protection zone around the evasion point; defining an evasion route which is in the form of a circular path and has a predefined radius of curvature; controlling the vehicle at a critical distance, so that the vehicle follows the evasion route which is tangential to the protection zone.

Claims

1. A method for controlling a vehicle to automatically avoid collisions of the vehicle with obstacles;

the method comprising;

detecting and localizing an obstacle in front of the vehicle, wherein relative position of the obstacle with respect to the vehicle is defined by relative coordinates;

measuring the relative velocity with respect to the detected obstacle;

evaluating a risk of collision of the vehicle with the obstacle dependent upon the current relative position and relative velocity;

performing of the following steps, if there is a risk of collision with the obstacle;

computing a domain of uncertainty around the obstacle dependent upon measurement errors;

computing an avoidance point at a boundary of the domain of uncertainty;

defining a protection zone having a radius around the avoidance point;

defining a circular exit path of a given curve radius; and

navigating the vehicle at a critical distance such that the vehicle follows the circular exit path, wherein the circular exit path is tangential to the protection zone.

2. The method according to claim 1, wherein upon measuring the relative velocity with respect to the detected obstacle below and below a horizontal plane, the plane extending through the vehicle, the relative position, or the relative velocity respectively, are projected in the plane, and wherein the projected position or velocity, respectively, are used as measured values for computing the uncertainty domain, the avoidance point, and the protection zone.

3. The method according to claim 2, wherein, while the vehicle is situated on the circular route, the distance toward the obstacle is being measured, and wherein, if that distance toward the obstacle is increasing, the circular exit path is left and a contrary motion commences.

4. The method according to claim 3, wherein, upon having commenced the contrary motion, a route control in the direction of one waypoint is activated, if the absolute value of the angular difference between the waypoint and a compass direction and the vehicle and the compass direction is lower than a given threshold value.

5. The method according to claim 1, wherein the detected obstacle is classified into one of three groups:

namely a first group “

stationary obstacles”

, a second group “

moving and cooperating obstacles”

, and a third group “

moving and non-cooperating obstacles”

, wherein cooperating obstacles follow pre-defined avoidance rules, and wherein non-cooperating obstacles ignore the vehicle.

6. The method according to claim 5, wherein the uncertainty domain, and thus also the avoidance point, is determined dependent upon the classification as stationary, moving and cooperating, and moving and non-cooperating obstacle.

7. The method according to claim 5, wherein, in case of stationary obstacles, four vertices of the stationary uncertainty domain are determined according to formulas:

point 1;

[(R_SENSE−

R_MIN)cos(Ψ

−

Ψ

_MIN), (R_SENSE−

R_MIN)sin(Ψ

−

Ψ

_MIN)],

point 2;

[(R_SENSE+R_MAX)cos(Ψ

−

Ψ

_MIN), (R_SENSE+R_MAX)sin(Ψ

−

Ψ

_MIN)],

point 3;

[(R_SENSE+R_MAX)cos(Ψ

+Ψ

_MAX), (R_SENSE−

R_MIN)sin(Ψ

+Ψ

_MAX)], and

point 4;

[(R_SENSE−

R_MIN)cos(Ψ

+Ψ

_MAX), (R_SENSE−

R_MIN)sin d(Ψ

+Ψ

_MAX)],

wherein the domain of uncertainty is characterized by a nominal value Ψ

, by R_SENSE of the Azimut Angle, the distance, and corresponding uncertainty values Ψ

_MIN, Ψ

_MAX, R_MIN, R_MAX.

8. The method according to claim 7, wherein the avoidance point is equal to point 4, if avoidance is to be performed in right direction, and wherein the avoidance point is equal to point 1, if avoidance is to be performed in left direction.

9. The method according to claim 8, wherein the avoidance point is determined according to the following rules:

when avoiding in right direction;

point 4,

when avoiding in left direction;

point 1.

10. The method according to claim 5, wherein for moving and cooperating obstacles the protection zone around the avoidance radius is increased.

11. The method according to claim 5, wherein for stationary obstacles the critical distance is reached, if the following inequality is satisfied:

2·

ρ

·

R_RISK+R_RISK²+2·

ρ

·

R_SENSE·

sin(Ψ

)−

R_SENSE²>

0,

wherein ρ

represents the radius of the exit path, R_RISK the radius of a circular protection zone, Ψ

the measured Azimut Angle of the obstacle, and R_SENSE the measured distance between the obstacle and the vehicle.

12. The method according to claim 5, wherein, for stationary obstacles, the critical distance is reached upon satisfying the following inequality for any Azimut Angle Ψ

of a domain of uncertainty Γ

, and for any distance R_SENSE of a domain of uncertainty Σ

;

2·

ρ

·

R_RISK+R_RISK²+2·

ρ

·

Σ

·

sin(Γ

)−

Σ

²>

0,

wherein ρ

represents the radius of the exit path and R_RISK the radius of a circular protection zone.

13. The method according to claim 5, wherein, for moving and cooperating obstacles, the critical distance is reached upon satisfying the following inequality:

2·

ρ

·

R+R²+2·

ρ

·

R_SENSE·

sin(Ψ

)−

R_SENSE²>

0,

wherein R=ρ

+R_RISK and ρ

is the radius of the exit path, R_RISK represents the radius of a circular protection zone, Ψ

the measured Azimut Angle of the obstacle, and the R_SENSE measured distance between obstacle and vehicle.

14. The method according to claim 13, wherein, for stationary obstacles, the critical distance is reached upon satisfying the following inequality for any Azimut Angle Ψ

of a domain of uncertainty Γ

, and for any distance R_SENSE of a domain of uncertainty Σ

;

2·

ρ

·

R_RISK+R_RISK²+2·

ρ

·

R_SENSE·

sin(Ψ

)−

R_SENSE²>

0,

wherein ρ

represents the radius of the exit path and R_RISK the radius of a circular protection zone.

15. The method according to claim 5, wherein, for moving and non-cooperating obstacles, a collision is anticipated.

16. The method according to claim 15, wherein a collision is anticipated in that the inequality

sin(Ψ

−

X_r)−

(R_RISK/R_SENSE)²>

0

is being evaluated, and a risk of collision is assumed, if the inequality is not satisfied, wherein Ψ

is the measured Azimut Angle of the obstacle, R_RISK is the measured distance toward the obstacle, and X_r is the angle included by the vector of the relative velocity of the obstacle and the direction of motion.

17. The method according to claim 16, wherein the domains of uncertainty of Ψ

, R_SENSE, and X_r are taken into account.

18. The method according to claim 5, wherein the protection zone is a movement corridor of the obstacle, the width of the movement corridor being twice the mandatory minimum distance (R_RISK) toward the obstacle.

19. The method according to claim 5, wherein the protection zone is a movement corridor of the obstacle, the minimum width of the movement corridor being twice the width of the mandatory distance (R_RISK) toward the obstacle, wherein the corridor per unit length increases in width for 2 sin(α

), wherein α

is the maximum angular error of the measured velocity vector of the obstacle.

20. The method according to claim 19, wherein the protection zone defined by the corridor extends in front of obstacle the length of R_RISK+v₀t, where R_RISK is the mandatory minimum distance between the obstacle and the vehicle, v₀ is the measured velocity of the obstacle, and t is the time period, which the vehicle on the exit path requires to depart from the corridor.

21. The method according to claim 5, wherein, in case of currents, stationary obstacles are considered dynamic and non-cooperative obstacles have a relative velocity with respect to the vehicle, this relative velocity being directed opposite to the velocity of the currents, but its absolute value being equal to it.

22. A system for obstacle detection and for computing an avoidance maneuver in a vehicle, the system comprising:

an autopilot unit for controlling the movement of the vehicle, the autopilot unit having an interface adapted to receive motion parameters, the motion parameters describing a target movement of the vehicle, and wherein the autopilot unit is adapted to navigate the vehicle dependent upon the motion parameters such that the actual movement of the vehicle approximately equals the target movement;

a sensor unit having at least one environmental sensor adapted to detect an obstacle, and adapted to provide sensor data dependent upon relative position of the obstacle and the relative velocity of the obstacle;

a data processing unit adapted to process the sensor data, and adapted to provide measured values for the relative position and the relative velocity; and

a unit for computing an avoidance maneuver adapted to compote motion parameters for a target movement using the measured values, and to provide these motion parameters to the interface of the autopilot unit.

23. The system according to claim 22, wherein the obstacle is arranged in a particular azimut angle and in a particular elevation angle in front of the vehicle and the obstacle has a particular distance toward the vehicle, and wherein the environmental sensor captures a certain area in front of the vehicle, the environmental sensor providing the distance, the azimut angle, the elevation angle, and the relative velocity of the obstacle as sensor data.

24. The system according to claim 22, wherein the sensor unit comprises a second environmental sensor, wherein the first environmental sensor is an optical sensor and the second environmental sensor is a radar sensor.

25. The system according to claim 24, wherein the sensor unit is adapted to conflate sensor data of at least the first and the second environmental sensor, and adapted to compute Azimut Angle, elevation angle, distance, and relative velocity of the obstacle from information comprised in the sensor data.

26. The system according to claim 25, wherein the first optical sensor is an area scan camera providing an image of the environment, and wherein the sensor unit is adapted to:

detect and classify potential obstacles on the image,

determine the actual size of the obstacles from a database on the basis of that classification, and

compute the distance between the respective obstacle and the vehicle from the actual size of each obstacle and the image.

27. The system according to claim 26, wherein the relative position of the respective obstacle is a criterion for the danger of an obstacle, wherein the one obstacle is selected to be most dangerous, whose distance toward the center of the circular exit path is least.

29. The system according to claim 23, wherein the data processing unit is adapted to project the relative position of an obstacle in a horizontal plane, in order to determine the horizontal distance and provide it as measured value for the relative position.

30. The system according to claim 23, wherein the sensor data for the relative position of each obstacle provided by the sensor unit are within a domain of uncertainty, and wherein the date processing unit is adapted to reduce the sensor data by conflating a number of obstacles which are located close to each other, and by computing a common relative position for conflated obstacles with a correspondingly increased domain of uncertainty.

31. The system according to claim 22, wherein the unit for computing an avoidance maneuver is adapted to compute motion parameters for an emergency avoidance maneuver, if the distance of an obstacle is lower than a critical distance.

32. The system according to claim 22, wherein the unit for computing an avoidance maneuver comprises a unit for classifying the obstacle, the unit for classifying the obstacle being adapted to classify the obstacle on the basis of the measured data, in order to deduce the behavior of the obstacle thereof.

33. The system according to claim 32, wherein the unit for classifying the obstacle is adapted to determine from the sensor data, if the obstacle is a stationary obstacle, a moving and cooperating obstacle, or a moving and non-cooperating obstacle.

34. The system according to claim 32, wherein the unit for computing an avoidance maneuver is adapted to provide motion parameters for the autopilot unit dependent upon the relative position and upon the relative velocity of the most dangerous obstacle, and dependent upon the result of the classification.

28. (canceled)

35. A method for detecting obstacles and for computing an avoidance maneuver in a vehicle, the method comprising:

locating an obstacle using a first environmental sensor and providing corresponding sensor data representing the relative position and relative velocity of the obstacle;

processing the sensor data and providing measured values for the relative position and for the relative velocity;

computing motion parameters for a target movement of an avoidance maneuver based upon the measured values; and

applying the motion parameters to an interface of an autopilot unit for controlling the movement of the vehicle, the autopilot unit being adapted to navigate the vehicle on the basis of the motion parameters such that the actual movement of the vehicle approximates the target movement.

36. The method according to claim 35, further comprising:

measuring a distance, an azimut angle, an elevation angle, and the relative velocity of the obstacle, wherein the distance, the azimut angle, and the elevation angle represent the relative position of the obstacle in front of the vehicle.

37. The method according to claim 35, wherein the first environmental sensor is an optical sensor providing an image of the environment, the method further comprising:

providing sensor data of a second environmental sensor being implemented as a radar sensor.

38. The method according to claim 37, further comprising:

conflating of sensor data of the optical sensor and of the radar sensor and computing the Azimut Angle, the elevation angle, the distance, and the relative velocity of the obstacle from the information comprised in the sensor data.

39. The method according to claim 37, wherein the optical sensor is at least one area scan camera providing an image of the environment of the vehicle;

the method further comprising;

detecting and classifying potential obstacles on the image on the basis of the sensor data of the area scan camera;

determining the actual size of the obstacles from a database based upon the classification;

computing the distance between the respective obstacle and the vehicle form the actual size of each obstacle and the image.

40. The method according to claim 38, further comprising:

detecting potential obstacles on the pictures based upon sensor data of area scan camera;

approximately determining the relative position of the obstacle with respect to the vehicle on the basis of the sensor data of the area scan camera;

sorting the detected obstacles according to their danger with respect to given criteria and selecting the most dangerous obstacle;

swinging the radar sensor in the direction of the most dangerous obstacle on the basis of the approximately determined relative position; and

measuring the relative position using the radar sensors with higher accuracy than the approximately determined relative position.

41. The method according to claim 40, wherein the relative position of the respective obstacle serves as criterion for the danger of an obstacle, wherein the one obstacle is selected to be most dangerous, whose distances toward the center of the circular exit path is least.

42. The method according to claim 36, further comprising:

projecting the relative position of an obstacle in a horizontal plane in order to determine the horizontal distance;

providing the horizontal distance as a measured value for the relative position.

43. The method according to claim 36, wherein the sensor data for the relative position of each obstacle are each within a domain of uncertainty;

the method further comprising;

conflating a number of obstacles which are located close to each other;

computing a common relative position with a correspondingly increased domain of uncertainty for the conflated obstacles in order to reduce sensor data.

44. The method according to claim 35, further comprising:

computing motion parameters for an emergency avoidance maneuver, if the distance of an obstacle is lower than a critical distance.

45. The method according to claim 35, further comprising:

classifying the detected obstacle on the basis of the measured data in order to deduce the behavior of the obstacle thereof.

46. The method according to claim 45, wherein classifying the obstacle comprises:

determining, if the obstacle is a stationary object, a moving and cooperating object, or a moving and non-cooperating object based upon the sensor data.

47. The method according to claim 45, further comprising:

computing motion parameters for the exit path dependent upon the relative position and the relative velocity of the most dangerous obstacle and dependent upon the result of the classification; and

providing motion parameters for the interface of the autopilot unit.