METHOD FOR MIGRATING RADAR SENSOR LIMITATIONS WITH VIDEO CAMERA INPUT FOR ACTIVE BRAKING FOR PEDESTRIANS

• US 20170205506A1
• Filed: 07/21/2015
• Published: 07/20/2017
• Est. Priority Date: 07/25/2014
• Status: Active Grant

• Current Assignee: Robert Bosch GmbH, Robert Bosch LLC (Robert Bosch GmbH)
• Original Assignee: Robert Bosch GmbH
• Inventors: Voorheis, Thomas, Schwindt, Oliver F.
• Granted Patent: US 10,444,346 B2
• Time in Patent Office: Days
• CPC Class Codes: G01S 13/66 Radar-tracking systems; Analogous systems
• CPC Class Codes: G01S 13/867 Combination of radar systems with cameras
• CPC Class Codes: G01S 13/87 Combinations of radar systems, e.g. primary radar and secondary radar
• CPC Class Codes: G01S 13/931 of land vehicles
• CPC Class Codes: G01S 2013/93185 Controlling the brakes
• CPC Class Codes: G01S 2013/9323 Alternative operation using light waves
• CPC Class Codes: G01S 2013/93271 in the front of the vehicles
• CPC Class Codes: G01S 2013/93272 in the back of the vehicles

Abstract

A pedestrian collision avoidance system for a vehicle including a radar sensor, a video camera, and an electronic control unit (ECU). The ECU detects an object in the video information and classifies the object as a pedestrian based on a comparison of the video information with a database. The ECU determines a distance to the pedestrian based on the radar information and determines a characteristic of the pedestrian based on the video information, the distance, and the database. When the pedestrian ceases to be detected by the radar sensor, the ECU determines an updated distance to the pedestrian based on the video information and the characteristic of the pedestrian. The ECU determines whether a potential for collision exists between the vehicle and the pedestrian based in part on the distance to the pedestrian, and when the potential for collision is present, the ECU activates an automatic vehicle response.

Claims

1. A method of operating a pedestrian collision avoidance system on a vehicle, the method comprising:

sensing a surrounding area of the vehicle with a radar sensor and a video camera;

sending radar information from the radar sensor and video information from the video camera to an electronic control unit;

detecting an object in the video information with the electronic control unit;

classifying the object as a pedestrian based on a comparison of the video information with a database;

determining, with the electronic control unit, a distance between, based on the radar information, an object classified as a pedestrian and the vehicle;

determining, with the electronic control unit, a characteristic of the object classified as a pedestrian based on the video information, the distance, and the database;

storing the characteristic of the object classified as a pedestrian in memory;

when the object classified as a pedestrian ceases to be detected by the radar sensor, determining, with the electronic control unit, an updated distance to the object classified as a pedestrian based on the video information and the characteristic of the object classified as a pedestrian;

determining, with the electronic control unit, whether a potential for collision between the vehicle and the object classified as a pedestrian is present based in part on the distance to the object classified as a pedestrian; and

when the potential for collision is present, activating an automatic vehicle response.

2. The method of claim 1, wherein the detecting of an object includes tracking an object using a bounding box and wherein determining a characteristic of the object classified as a pedestrian includes determining a height of the object classified as a pedestrian based on the bounding box.

3. The method of claim 1, wherein classifying the object as a pedestrian includes comparing a height of the object classified as a pedestrian with a height of one or more other detected objects in the video information.

4. The method of claim 1, wherein classifying the object as a pedestrian includes comparing a height of the object to a height of a plurality of stored image models representing different classes of pedestrians in different poses.

5. The method of claim 1, further comprising:

when the object classified as a pedestrian is in a field of view of both the radar sensor and the video camera, updating the distance determination, with the electronic control unit, based on the radar information and the video information, and

when the object classified as a pedestrian is in the field of view of only the video camera, updating the distance determination, with the electronic control unit, based at least in part on the stored characteristic of the object classified as a pedestrian.

6. The method of claim 5, wherein the stored characteristic of the object classified as a pedestrian is a height of the object classified as a pedestrian.

7. The method of claim 5, further comprising determining whether a potential for collision between the vehicle and the object classified as a pedestrian is present, with the electronic control unit, based on the updated distance determination.

8. The method of claim 1, wherein an automatic vehicle response includes

determining a braking time, with the electronic control unit, the braking time being a time to apply braking to avoid a collision between the object classified as a pedestrian and the vehicle;

and activating a braking system based on the braking time.

9. The method of claim 8, wherein the braking time is determined based on known braking characteristics of the vehicle and a safety margin.

10. The method of claim 8, further comprising calculating a time to collision, with the electronic control unit, between the vehicle and the object classified as a pedestrian and activating the braking system when the time to collision approaches the braking time.

11. The method of claim 1, wherein determining a distance to the object classified as a pedestrian includes comparing detected objects determined from the video information with detected objects determined from the radar information.

12. The method of claim 1, wherein determining whether a potential for collision between the vehicle and the object classified as a pedestrian is present is also based on speed and direction of the vehicle.

13. The method of claim 1, wherein the radar sensor is a long-range radar sensor with a field of view and the video camera has a wider field of view than the long-range radar sensor.

14. A pedestrian collision avoidance system for a vehicle, the system comprising:

a radar sensor positioned on the vehicle;

a video camera positioned on the vehicle;

an electronic control unit, electrically connected to the radar sensor and the video camera, the electronic control unit configured to

sense a surrounding area of the vehicle with the radar sensor and the video camera;

receive radar information from the radar sensor and video information from the video camera;

detect an object in the video information;

classify the object as a pedestrian based on a comparison of the video information with a database;

determine a distance between, based on the radar information, an object classified as a pedestrian and the vehicle;

determine a characteristic of the object classified as a pedestrian based on the video information, the distance, and the database;

store the characteristic of the object classified as a pedestrian in memory;

when the object classified as a pedestrian ceases to be detected by the radar sensor, determining an updated distance to the object classified as a pedestrian based on the video information and the characteristic of the object classified as a pedestrian;

determine whether a potential for collision between the vehicle and the object classified as a pedestrian is present based in part on the distance to the object classified as a pedestrian; and

when the potential for collision is present, activate an automatic vehicle response.

15. The pedestrian collision avoidance system of claim 14, wherein the electronic control unit is further configured to track the object using a bounding box and determine a height of the object classified as a pedestrian based on the bounding box.

16. The pedestrian collision avoidance system of claim 14, wherein the electronic control unit is further configured to compare a height of the object classified as a pedestrian with a height of one or more other detected objects in the video information.

17. The pedestrian collision avoidance system of claim 14, wherein the electronic control unit is further configured to compare a height of the object classified as a pedestrian to a height of a plurality of stored image models representing different classes of pedestrians in different poses.

18. The pedestrian collision avoidance system of claim 14, wherein the electronic control unit is further configured to

when the object classified as a pedestrian is in a field of view of both of the radar sensor and the video camera, update the distance determination based on the radar information and the video information, and

when the object classified as a pedestrian is in the field of view of only the video camera, update the distance determination based at least in part on the recorded characteristic of the object classified as a pedestrian.

19. The pedestrian collision avoidance system of claim 18, wherein the stored characteristic of the object classified as a pedestrian is a height of the object classified as a pedestrian.

20. The pedestrian collision avoidance system of claim 18, wherein the electronic control unit is further configured to determine whether a potential for collision between the vehicle and the object classified as a pedestrian is present based on the updated distance determination.

21. The pedestrian collision avoidance system of claim 14, wherein an automatic vehicle response includes

determining a braking time, the braking time being a time to apply braking to avoid a collision between the object classified as a pedestrian and the vehicle;

and activating a braking system based on the braking time.

22. The pedestrian collision avoidance system of claim 21, wherein the braking time is determined based on known braking characteristics of the vehicle and a safety margin.

23. The pedestrian collision avoidance system of claim 21, wherein the radar sensor is a long-range radar sensor and the video camera has a wider field of view than the long-range radar sensor.