Collision warning system

• US 8,082,101 B2
• Filed: 01/19/2005
• Issued: 12/20/2011
• Est. Priority Date: 04/08/2004
• Status: Active Grant

• Current Assignee: MobilEye Vision Technologies Ltd. (Intel Corporation)
• Original Assignee: Mobileye Technologies Limited (Intel Corporation)
• Inventors: Stein, Gideon P., Dagan, Erez, Mano, Ofer, Shashua, Amnon
• Primary Examiner(s): PERUNGAVOOR, SATHYANARAYA V
• Application Number:
  US10/599,667
• Publication Number:
  US 20090143986A1
• Time in Patent Office: 2,526 Days
• Field of Search: 382/104
• US Class Current: 701/301
• CPC Class Codes: B60Q 9/008 for anti-collision purposes
• CPC Class Codes: B60T 7/22 initiated by contact of vehicle, e.g. bumper, with an external object, e.g. another vehicle , or by means of contactless obstacle detectors mounted on the vehicle
• CPC Class Codes: G06T 2207/10004 Still image; Photographic image
• CPC Class Codes: G06T 2207/30232 Surveillance
• CPC Class Codes: G06T 2207/30236 Traffic on road, railway or crossing
• CPC Class Codes: G06T 2207/30241 Trajectory
• CPC Class Codes: G06T 2207/30252 Vehicle exterior; Vicinity of vehicle
• CPC Class Codes: G06T 2207/30256 Lane; Road marking
• CPC Class Codes: G06T 7/248 involving reference images or patches
• CPC Class Codes: G06T 7/337 involving reference images or patches
• CPC Class Codes: G06T 7/74 involving reference images or patches
• CPC Class Codes: G06V 20/58 Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
• CPC Class Codes: G08G 1/16 Anti-collision systems road vehicle drive control systems for predicting or avoiding probable or impending collision otherwise than by control of a particular sub-unit B60W30/08
• CPC Class Codes: G08G 1/161 Decentralised systems, e.g. inter-vehicle communication
• CPC Class Codes: G08G 1/166 for active traffic, e.g. moving vehicles, pedestrians, bikes

Abstract

A method of estimating a time to collision (TTC) of a vehicle with an object comprising: acquiring a plurality of images of the object; and determining a TTC from the images that is responsive to a relative velocity and relative acceleration between the vehicle and the object.

Claims

1. A method of estimating a time-to-collision (TTC) of a vehicle with an object comprising the step of:

(a) acquiring a plurality of images of the object at known time intervals between the times t at which the images are acquired;

(b) determining a momentary time T_v(t), which is equal to an instantaneous distance between the vehicle and the object at time t divided by an instantaneous relative velocity, from information derived from the images and the time intervals; and

determining the time-to-collision (TTC) based on a rate of change over times t of momentary time T_v(t).

2. The method according to claim 1, further comprising the step of:

(c) determining a function of the relative velocity from the images and using the function of the relative velocity to determine TTC.

3. The method according to claim 1, wherein said (b) determining a momentary time T_v(t) is based on a change in scale of images of at least a portion of the object.

4. The method according to claim 3, wherein said determining the change in scale includes determining a ratio between a dimension of the object in a first one of the images and the same dimension of the object in a second one of the images.

5. The method according to claim 4, wherein said determining the time-to-collision (TTC) includes determining the momentary time T_v(t)=[1/(S(t)−

1)]Δ

T where S is the ratio and Δ

T is the time interval between the acquisition of two images of the images.

6. The method, according to claim 1, further comprising the step of (c) determining a function of the relative acceleration from the images and using said function of the relative acceleration to determine the TTC.

7. The method according to claim 6, wherein said (c) determining said function of the relative acceleration includes determining a time derivative of a function of the relative velocity.

8. The method according to claim 1, further comprising the step of:

(e) determining whether the vehicle and the object are on a course that leads to a collision at the TTC.

9. A method according to claim 8, wherein said determining whether the vehicle and object are on a course that leads to a collision at the TTC includes:

(i) determining respective motions of at least two features of the object relative to the vehicle from the images; and

(ii) determining, from the relative motions, whether at TTC if the real space horizontal coordinates of said at least two features straddle the real-space horizontal coordinate of at least a part of the vehicle.

10. A system which performs the method steps of claim 1, for determining the time-to-collision (TTC) of the vehicle with the object, the system comprising:

(a) at least one camera mounted in the vehicle and adapted for said acquiring of the images at the known time intervals; and

(b) a processor which is configured in real time to calculate the momentary time T_v(t) equal to the instantaneous distance between the vehicle and the object at time t divided by the instantaneous relative velocity from information derived from the images and the time intervals, and, based on the rate of change over time t of momentary time T_v(t), is configured to calculate in real time the time-to-collision (TTC).

11. The system, according to claim 10 wherein the at least one camera is a single camera.

12. The system, according to claim 10, further comprising:

(c) an alarm apparatus operatively connected to the processor, the alarm apparatus configured for alerting a driver of the vehicle to a possible collision with the object responsive to the TTC.

13. The system, according to claim 10, further comprising:

(c) an alarm apparatus, operatively connected to the processor, wherein the alarm apparatus is configured to alert, based on the TTC, at least one person outside of the vehicle of a possible collision of the vehicle with the object.

14. The system, according to claim 10, wherein the at least one camera images an environment in front of the vehicle.

15. The system, according to claim 10, wherein the at least one camera images an environment in back of the vehicle.

16. The system, according to claim 10, wherein the at least one camera images an environment to a side of the vehicle.

17. A method of estimating a time-to-collision (TTC) of a vehicle with an object comprising the step of:

(a) acquiring a plurality of images of the object at known time intervals between the times at which the images of the plurality of images are acquired; and

(b) determining the time-to-collision (TTC) from information derived from the images and the time intervals, wherein said determining the TTC is based on a relative velocity and relative acceleration between the vehicle and the object, wherein said (b) determining the time-to-collision (TTC) includes determining a change in scale of an image of at least a portion of the object and using the change in scale for determining a function of the relative velocity, wherein said determining the change in scale includes determining a ratio between a dimension of the object in a first one of the images and the same dimension of the object in a second one of the images, wherein said determining a function of the relative velocity includes determining a function T_v=[1/(S−

1)]Δ

T, where S is the ratio and Δ

T is a time lapse between two images of the images, wherein a function of the relative acceleration is determined based on a time derivative T′

_v, of function T_v.

18. The method according to claim 17, wherein the TTC is determined responsive to a function of T_v and T′

_v.

19. The method according to claim 17, wherein TTC is determined responsive to the expression:

TTC (t)=[T_v/C)][1−

(1+2C)]^1/2, where C=T′

_v+1.

20. A method of determining whether a vehicle and an object are on a collision course, the method comprising the steps of:

(a) acquiring a plurality of images of the object from a position in the vehicle at a plurality of known times;

(b) determining, from the images, respective motions in image space of at least two features of the object relative to the vehicle;

(c) determining from the images an estimate of a time to collision (TTC) of the vehicle and the object; and

(d) determining from the motions whether at the TTC, the real-space horizontal coordinates of said at least two features straddle the real-space horizontal coordinate of the position in the vehicle, whereby the vehicle and object are on a collision course.

21. The method according to claim 20, wherein said (b) determining respective motions of the at least two features includes determining lateral motion of the features relative to the vehicle.

22. The method according to claim 20, wherein said (d) determining includes extrapolating lateral locations of the at least two features at TTC from the respective motions at the known times of said acquiring.

23. A method according to claim 20, wherein said (c) determining includes determining TTC solely from the images and time intervals between the known times of said acquiring of the images.

24. A system which performs the method steps of claim 20, for determining whether a vehicle and an object are on a collision course, the system comprising:

(a) a camera mounted in the vehicle and adapted for said acquiring of the images; and

(b) a processor which determines from the images, respective motions of at least two features of the object relative to the vehicle and from the images determines an estimate of a possible time to collision (TTC) of the vehicle and the object; and

from the relative motions determines whether at the TTC, the real space horizontal coordinates of the first and second features straddle the real-space horizontal coordinate the camera.