SYSTEM AND METHOD FOR CLASSIFYING A TARGET VEHICLE

• US 20090222172A1
• Filed: 03/03/2008
• Published: 09/03/2009
• Est. Priority Date: 03/03/2008
• Status: Active Grant

• Current Assignee: Ford Global Technologies, LLC (Ford Motor Company)
• Original Assignee: Ford Global Technologies, LLC (Ford Motor Company)
• Inventors: Joh, Peter Gyumyeong, Yopp, Wilford Trent
• Granted Patent: US 8,140,225 B2
• Time in Patent Office: Days
• US Class Current: 701/45
• CPC Class Codes: B60R 2021/01272 Belt tensioners
• CPC Class Codes: B60R 21/0134 responsive to imminent contact with an obstacle , e.g. using radar systems
• CPC Class Codes: G01S 13/931 of land vehicles
• CPC Class Codes: G01S 2013/9321 Velocity regulation, e.g. cruise control
• CPC Class Codes: G01S 2013/9325 for inter-vehicle distance regulation, e.g. navigating in platoons
• CPC Class Codes: G01S 2013/93271 in the front of the vehicles
• CPC Class Codes: G01S 7/412 based on a comparison between measured values and known or stored values

Abstract

A system and method for classifying a target vehicle using a sensor system, including a plurality of target vehicle sensors. The sensor system acquiring target vehicle data points that define at least an upper and lower portion of the target vehicle. The sensor system reconstructing a target vehicle shape using the target vehicle data points to provide a first and second target vehicle classification value. The sensor system determining an overall target vehicle classification based in part upon the first and second target vehicle classification values.

Claims

1. A method for classifying a target vehicle using a sensor system located within a host vehicle, the method comprising:

acquiring target vehicle data points using at least a first sensor and a second sensor, the first sensor acquiring the target vehicle data points associated with at least an upper portion of the target vehicle, and the second sensor acquiring the target vehicle data points associated with at least a lower portion of the target vehicle;

calculating a delta distance between the location of the upper portion and the lower portion based upon the acquired target vehicle data points, wherein the delta distance is compared against a first target vehicle threshold value in order to determine a first target vehicle classification value; and

classifying the target vehicle using the first target vehicle classification value.

2. The method of claim 1 further comprising calculating a relative traveling distance of the host vehicle in relation to the target vehicle using the first and second sensor, and combining the calculated relative traveling distance with the data from the upper portion and the lower portion in order to reconstruct a target vehicle shape.

3. The method of claim 2 further comprising:

calculating a standard deviation between each of a plurality of target vehicle templates and the reconstructed target vehicle shape;

determining a second target vehicle classification value corresponding to a target vehicle data template that provides the smallest standard deviation value in relation to the reconstructed target vehicle shape; and

classifying the target vehicle using the first and second target vehicle classification values.

4. The method of claim 3 further comprising:

comparing each calculated standard deviation value to each other; and

assigning the second target vehicle classification value as undefined if the difference between a pair of standard deviation values is less than a second target vehicle threshold value.

5. The method of claim 3 further comprising:

superimposing at least a first target vehicle data template and a second target vehicle template in order to align the reconstructed target vehicle shape at a common intersection point so that each standard deviation may be calculated.

6. The method of claim 3 further comprising:

applying a fuzzy logic interface to determine a first weight value for the first target vehicle classification value, wherein the first weight value biases the classification of the target vehicle;

applying the fuzzy logic interface to determine a second weight value for the second target vehicle classification value, wherein the second weight value biases the classification of the target vehicle.

7. The method of claim 6 further comprising classifying the target vehicle according to the first and second target vehicle classification values based at least in part upon the first weight value and the second weight value.

8. The method of claim 1, wherein the at least first and second sensors are distance sensors.

9. The method of claim 1 further comprising acquiring the target vehicle data points using a plurality of closing velocity (CV) sensors.

10. The method of claim 1 further comprising enabling at least one pre-crash safety system based upon the classification of the target vehicle.

11. The method of claim 1 further comprising adjusting the sensor system acquisition distance in response to the transmittance coefficient of the front windshield of the host vehicle and the acquired relative speed as the host vehicle approaches the target vehicle.

12. The method of claim 1 further comprising classifying the target vehicle as a vehicle type from the group consisting of sedan, SUV, truck, coupe, crossover, semi, motorcycle and station wagon according to the first target vehicle classification value.

13. A method for classifying a target vehicle using a CV sensor system located within a host vehicle, the method comprising:

acquiring target vehicle data points using at least a first CV sensor and a second CV sensor, the first CV sensor acquiring the target vehicle data points associated with at least an upper portion of the target vehicle, and the second CV sensor acquiring the target vehicle data points associated with at least a lower portion of the target vehicle;

calculating a relative traveling distance of the host vehicle in relation to the target vehicle using the first and second CV sensor, and combining the calculated relative traveling distance with the data from the upper portion and the lower portion in order to reconstruct a target vehicle shape;

calculating a standard deviation between each of a plurality of target vehicle templates and the reconstructed target vehicle shape, wherein a second target vehicle classification value is determined using a target vehicle template that generates the smallest standard deviation in relation to the reconstructed target vehicle shape; and

classifying the target vehicle using the second target vehicle classification value.

14. The method of claim 13 further comprising classifying the target vehicle as a vehicle type from the group consisting of sedan, SUV, truck, coupe, crossover, semi, motorcycle and station wagon according to the second target vehicle classification value.

15. The method of claim 13 further comprising:

calculating a delta distance between the location of the upper portion and the lower portion, wherein the delta distance is compared against a first target vehicle threshold value in order to determine a first target vehicle classification value; and

classifying the target vehicle using the first and second target vehicle classification values.

16. The method of claim 13 further comprising enabling at least one pre-crash safety system based upon the classification of the target vehicle.

17. The method of claim 16 further comprising enabling the at least one pre-crash safety system to a default state when the target vehicle classification value is undefined.

18. A sensor system for classifying the rear-end of a target vehicle, the sensor system comprising:

a plurality of target vehicle sensors, acquiring a series of target vehicle data points, the target vehicle data points defining at least an upper and lower portion of the target vehicle; and

a controller receiving the acquired target vehicle data points from the plurality of target vehicle sensors, wherein the controller determines a horizontal distance between the acquired target vehicle data points corresponding to the upper portion and the lower portion, compares the horizontal distance to a first target vehicle threshold value, and assigns a first target vehicle classification value based upon the first target vehicle threshold value and the horizontal distance in order to classify the target vehicle.

19. The sensor system of claim 18, wherein the acquired target vehicle data points provide a reconstructed target vehicle shape that is compared to a plurality of target vehicle templates by determining a standard deviation value corresponding to each of the plurality of target vehicle templates in relation to the reconstructed target vehicle shape, the controller assigning a second target vehicle classification value in order to classify the target vehicle based upon the comparison of each standard deviation value.

20. The sensor system of claim 18, wherein the plurality of target vehicle sensors include closing velocity (CV) sensors.