COMPOSITE CONFIDENCE ESTIMATION FOR PREDICTIVE DRIVER ASSISTANT SYSTEMS

• US 20150112571A1
• Filed: 10/22/2014
• Published: 04/23/2015
• Est. Priority Date: 10/22/2013
• Status: Active Grant

• Current Assignee: Honda Research Institute Europe GmbH (Honda Motor Company)
• Original Assignee: Honda Research Institute Europe GmbH (Honda Motor Company)
• Inventors: SCHMDDERICH, Jens
• Granted Patent: US 9,308,919 B2
• Time in Patent Office: Days
• US Class Current: 701/93
• CPC Class Codes: B60W 2050/0022 Gains, weighting coefficients or weighting functions
• CPC Class Codes: B60W 2554/00 Input parameters relating to objects
• CPC Class Codes: B60W 2554/20 Static objects
• CPC Class Codes: B60W 2554/80 Spatial relation or speed relative to objects
• CPC Class Codes: B60W 2556/20 Data confidence level
• CPC Class Codes: B60W 30/08 Active safety systems predicting or avoiding probable or impending collision or attempting to minimise its consequences
• CPC Class Codes: B60W 30/14 Adaptive cruise control
• CPC Class Codes: B60W 30/16 Control of distance between vehicles, e.g. keeping a distance to preceding vehicle
• CPC Class Codes: B60W 30/18163 Lane change; Overtaking manoeuvres
• CPC Class Codes: B60W 40/04 Traffic conditions
• CPC Class Codes: B60W 50/0097 Predicting future conditions
• CPC Class Codes: B60W 60/0027 using trajectory prediction for other traffic participants
• CPC Class Codes: G06F 18/254 of classification results, e.g. of results related to same input data
• CPC Class Codes: G06N 5/04 Inference or reasoning models
• 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

Abstract

The invention relates to a driving assistance system including a prediction subsystem in a vehicle. According to a method aspect of the invention, the method comprises the steps of accepting a set of basic environment representations; allocating a set of basic confidence estimates; associating weights to the basic confidence estimates; calculating a weighted composite confidence estimate for a composite environment representation; and providing the weighted composite confidence estimate as input for an evaluation of a prediction based on the composite environment representation.

Claims

1. A method for a prediction subsystem in a driving assistance system of a vehicle, the method comprising the following steps:

accepting a set of basic environment representations, wherein each basic environment representation represents at least one first entity detected by one or more sensors in an environment of the vehicle;

allocating a set of basic confidence estimates, wherein each basic confidence estimate of the set is associated to one of the set of basic environment representations, and each basic confidence estimate represents a combination of one or more detection confidences related to the associated basic environment representation;

associating at least one weight to one of the set of basic confidence estimates, wherein the weight is related to a composite environment representation based on the set of basic environment representations and the weight indicates an effect of a detection error in the basic environment representation, to which the basic confidence estimate is associated to, on a prediction for a second detected entity;

calculating a weighted composite confidence estimate for the composite environment representation based on a combination of the set of basic confidence estimates with the associated at least one weight; and

providing the weighted composite confidence estimate as input for an evaluation of the prediction based on the composite environment representation.

2. The method according to claim 1, wherein the detection confidences are based on at least one of signal strengths associated to sensor data, error propagation indications associated to sensor data, and an application of plausibility rules.

3. The method according to claim 1, wherein the set of basic confidence estimates is allocated by selecting from a plurality of basic confidence estimates those basic confidence estimates associated to the basic environment representations contributing to the composite environment representation.

4. The method according to claim 1, wherein the weight is assigned to minimize a probability of a wrong prediction for the second detected entity.

5. The method according to claim 1, wherein multiple weights are associated in a one-to-one relation to multiple basic confidence estimates and the multiple weights are assigned relative values reflecting the relative importance of the associated basic environment representations for the prediction.

6. The method according to claim 1, wherein a weight reflects the influence of a detection error on the probability for a critical prediction.

7. The method according to claim 6, wherein the critical prediction comprises at least one of a false positive prediction result or a false negative prediction result.

8. The method according to claim 1, wherein calculating the weighted composite confidence estimate comprises at least one of a summation, weighted summation, product, weighted product, and selecting a minimum or maximum.

9. The method according to claim 1, wherein it is decided on whether or not the composite environment representation is used for a prediction by comparing a numerical value of the weighted composite confidence estimate with a numerical value of at least one of a predefined threshold value and numerical values of one or more other confidence estimates.

10. The method according to claim 1, wherein one or more of the basic environment representations and the composite environment representation comprise direct and/or indirect indicators and one or more of the basic confidence estimates and the composite confidence estimate comprise confidence indicators wherein direct and/or indirect indicators and the confidence indicators are based on a similar data structure.

11. The method according to claim 1, wherein the driving assistance system is adapted to perform a cruise control functionality.

12. A computer program product comprising program code portions for performing the method according to claim 1 when the computer program product is executed on a computing device.

13. A driving assistance system for a vehicle, the driving assistance system including a prediction subsystem and comprising:

a component adapted to accept a set of basic environment representations, wherein each basic environment representation represents at least one first entity detected by one or more sensors in an environment of the vehicle;

a component adapted to allocate a set of basic confidence estimates, wherein each basic confidence estimate of the set is associated to one of the set of basic environment representations, and each basic confidence estimate represents a combination of one or more detection confidences related to the associated basic environment representation;

a component adapted to associate at least one weight to one of the basic confidence estimates, wherein the weight is related to a composite environment representation based on the set of basic environment representations and the weight indicates an effect of a detection error in the basic environment representation, to which the basic confidence estimate is associated to, on a prediction for a second detected entity;

a component adapted to calculate a weighted composite confidence estimate for the composite environment representation based on a combination of the set of basic confidence estimates with the associated at least one weight; and

a component adapted to provide the weighted composite confidence estimate as input for an evaluation of the prediction based on the composite environment representation.

14. The system according to claim 13, wherein the prediction subsystem is adapted to perform a context based prediction and a physical prediction.

15. A vehicle comprising a system according to claim 13.