Method and device for determining surroundings

• US 9,747,801 B2
• Filed: 04/15/2013
• Issued: 08/29/2017
• Est. Priority Date: 04/30/2012
• Status: Active Grant

• Current Assignee: Robert Bosch GmbH
• Original Assignee: Robert Bosch GmbH
• Inventors: Kanter, Pascal, App, Thomas
• Primary Examiner(s): PAIGE, TYLER D
• Application Number:
  US14/397,820
• Publication Number:
  US 20150116101A1
• Time in Patent Office: 1,597 Days
• Field of Search: 701 36, 382103
• CPC Class Codes: B60W 40/06 Road conditions
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks
• 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 1/165 for passive traffic, e.g. including static obstacles, trees
• CPC Class Codes: G08G 1/167 Driving aids for lane monitoring, lane changing, e.g. blind spot detection
• CPC Class Codes: H04N 7/183 for receiving images from a single remote source

Abstract

A method for classifying surroundings of a motor vehicle includes: providing a hypothesis regarding the class to which the surroundings belong; sampling pieces of information from the surroundings of the motor vehicle; determining a criterion which supports or weakens the hypothesis based on the sampled pieces of information; and determining a probability of the hypothesis being correct, using the criterion with the aid of Bayesian filtering.

Claims

1. A method for classifying surroundings of a motor vehicle into one of multiple predefined classes, comprising:

providing, by an estimating unit, a classification regarding one of multiple predefined classes to which the surroundings belong, the multiple predefined classes including at least two of a freeway class, a traffic-calm zone class, a city traffic zone class, or a parking situation class;

receiving, via an interface unit, sampled pieces of information from the surroundings of the motor vehicle;

determining, by a processing unit, based on the sampled pieces of information, at least one criterion which supports or weakens the classification;

storing a plurality of determined criteria; and

determining, by the processing unit, a probability of the classification being correct with the aid of Bayes filtering of the at least one criterion, the probability being determined based on the stored criteria.

2. The method as recited in claim 1, further comprising:

accepting the classification if the probability exceeds an upper threshold value, and rejecting the classification if the probability is below a lower threshold value which is lower than the upper threshold value.

3. The method as recited in claim 2, wherein a distance between the upper threshold value and the lower threshold value is predetermined.

4. The method as recited in claim 2, further comprising:

when the probability falls between the upper threshold value and the lower threshold value, one of;

indicating via a display unit that additional data is required to assess whether the classification should be accepted or rejected, and

continuing the determining, by the processing unit, the probability of the classification in the manner of a moving average including at least one past piece of information, the at least one past piece of information including data from within an interval of time relating to the sampled pieces of information.

5. The method as recited in claim 1, wherein each criterion relates to a point in time, and the criteria are cyclically stored based on respective reference points in time, so that the probability is determined with respect to a predetermined, past time period.

6. The method as recited in claim 5, wherein the number of the stored criteria is determined based on a speed of the motor vehicle.

7. The method as recited in claim 5, wherein a first weighting factor is assigned to each criterion.

8. The method as recited in claim 5, wherein the criteria are cyclically stored in a circular buffer in the motor vehicle.

9. The method as recited in claim 1, wherein the determination of each criterion includes a weighting with an uncertainty factor which increases as the latest determined probability increasingly moves away from extreme values.

10. The method as recited in claim 1, wherein the processing unit determines, based on the sampled pieces of information, at least two criterion which support or weaken the classification.

11. The method as recited in claim 1, wherein the processing unit determines a number of different classifications including the provided classification at a same time.

12. The method as recited in claim 1, wherein the multiple predefined classes including at least three of the freeway class, the traffic-calm zone class, the city traffic zone class, or the parking situation class.

13. The method as recited in claim 1, wherein the multiple predefined classes including all of the freeway class, the traffic-calm zone class, the city traffic zone class, and the parking situation class.

14. A non-transitory computer-readable data storage medium storing a computer program having program codes which, when executed on a computer, performs a method for classifying surroundings of a motor vehicle into one of multiple predefined classes, the method comprising:

providing, by an estimating unit, a classification regarding one of multiple predefined classes to which the surroundings belong, the multiple predefined classes including at least two of a freeway class, a traffic-calm zone class, a city traffic zone class, or a parking situation class;

receiving, via an interface unit, sampled pieces of information from the surroundings of the motor vehicle;

determining, by a processing unit, based on the sampled pieces of information, at least one criterion which supports or weakens the classification;

storing a plurality of determined criteria; and

determining, by the processing unit, a probability of the classification being correct with the aid of Bayes filtering of the at least one criterion, the probability being determined based on the stored criteria.

15. A device for classifying surroundings of a motor vehicle into one of multiple predefined classes, comprising:

an estimating device for providing a classification regarding one of multiple predefined classes to which the surroundings belong, the multiple predefined classes including at least two of a freeway class, a traffic-calm zone class, a city traffic zone class, or a parking situation class;

an interface for receiving sampled pieces of information from the surroundings of the motor vehicle;

a processing device for (i) determining, based on the sampled pieces of information, at least one criterion which supports or weakens the classification, and (ii) determining a probability of the classification being correct with the aid of Bayes filtering of the at least one criterion; and

a memory for storing the at least one criterion.

16. The device as recited in claim 15, wherein the memory is a circular buffer for cyclically storing the at least one criterion.

17. A method for indicating surroundings of a motor vehicle to a driver assistance system of the motor vehicle, comprising:

providing, by an estimating unit, a classification regarding one of multiple predefined classes to which the surroundings belong, the multiple predefined classes including at least two of a freeway class, a traffic-calm zone class, a city traffic zone class, or a parking situation class;

receiving, from a sampling device, sampled pieces of information from the surroundings of the motor vehicle;

determining, by a processing unit, based on the sampled pieces of information, at least one criterion which supports or weakens the classification;

determining, by the processing unit, a probability of the classification being correct with the aid of Bayes filtering of the at least one criterion;

accepting the classification if the probability exceeds a threshold; and

providing a signal indicating the surroundings of the motor vehicle to the driver assistance system of the motor vehicle.

18. The method as recited in claim 17, wherein in response to the classification being that the surroundings belong to the one of the multiple predefined classes corresponding to the freeway class, the at least one criterion includes at least one of:

i) a high speed, parallel traffic with a same direction of travel, ii) small average steering angles, or iii) multiple marked traffic lanes.

19. The method as recited in claim 18, wherein the at least one criterion includes the high speed, parallel traffic with the same direction of travel.

20. The method as recited in claim 18, wherein the at least one criterion includes small average steering angles.

21. The method as recited in claim 18, wherein the at least one criterion includes the multiple marked traffic lanes.

22. The method as recited in claim 17, wherein the sampling device includes at least one of a camera, a distance sensor and a position sensor.