Road Surface Condition Identification Based on Statistical Pattern Recognition

• US 20100250086A1
• Filed: 03/24/2009
• Published: 09/30/2010
• Est. Priority Date: 03/24/2009
• Status: Active Grant

• Current Assignee: GM Global Technology Operations LLC (General Motors Company)
• Original Assignee: GM Global Technology Operations Incorporated (Motors Liquidation Co. GUC Trust)
• Inventors: Deng, Weiwen, Zhao, Qingrong
• Granted Patent: US 8,306,712 B2
• Time in Patent Office: Days
• US Class Current: 701/82
• CPC Class Codes: B60W 40/068 Road friction coefficient

Abstract

A method is provided for determining an adhesive condition of a surface of a vehicle road using a statistical pattern recognition technique. A plurality of probability distribution functions is provided representing respective adhesive effects between the vehicle tire of a driven vehicle and the vehicle road. An index is calculated which represents a vehicle understeer characteristic. Probability analysis is applied for each of the road surface adhesive conditions as a function of the index. Each probability analysis is recursively updated. A likelihood factor is determined for each road surface adhesive condition as a function of each respective recursively updated probability analysis. Each respective road surface adhesive condition has a respective likelihood factor that identifies the likelihood of the road surface having the respective adhesive condition as a function of the index. The identified adhesive condition of the road surface is input into a vehicle control process.

Claims

1. A method of determining an adhesive condition of a surface of a vehicle road using a statistical pattern recognition technique, the method comprising the steps of:

(a) providing a plurality of probability distribution functions for a plurality of road surfaces which represent respective adhesive effects between the vehicle tire of a driven vehicle and the vehicle road;

(b) obtaining a steering wheel angle, a vehicle speed, and a yaw rate of the driven vehicle;

(c) calculating an index which represents a vehicle understeer characteristic for a vehicle traveling on a current road that is a function of the steering wheel angle input, the vehicle speed, and the yaw rate;

(d) applying a probability analysis for each of the road surface adhesive conditions as a function of the index;

(e) recursively updating the probability analysis for each respective road surface adhesive condition;

(f) determining a likelihood factor for each road surface adhesive condition as a function of each respective recursively updated probability analysis, each respective road surface adhesive condition having a respective likelihood factor that identifies the likelihood of the road surface having the respective adhesive condition as a function of the index; and

(g) inputting the identified adhesive condition of the road surface into a vehicle control process.

2. The method recited in claim 1 wherein steps (b)-(g) applied for each adhesive condition are performed in parallel.

3. The method of claim 1 wherein a plurality of probability density functions is constructed and stored in a vehicle database.

4. The method of claim 1 wherein each probability analysis for each respective adhesive condition is recursively updated a predetermined number of times.

5. The method of claim 4 wherein the probability analysis is based on Bayes'"'"' rule.

6. The method of claim 5 wherein the probability analysis based on Bayes'"'"' rule is estimated based on the following formula:

7. The method of claim 1 wherein the likelihood factor is determined by the following formula:

8. The method of claim 7 wherein the likelihood factor is output as a probability percentage of the road surface having the adhesive condition.

9. The method of claim 7 wherein the identification of the adhesive condition includes determining the higher probability between each of the respective likelihood factors.

10. The method of claim 1 wherein the index is determined as a function of the yaw rate that is derived as a function of a front and rear axle cornering stiffness.

11. The method recited in claim 1 wherein a vehicle front wheel steering angle is a vehicle operating characteristic derived by a driver'"'"'s steering input.

12. A method of determining an adhesive condition of a surface of a vehicle road using a statistical pattern recognition technique, the method comprising the steps of:

(a) providing a plurality of probability distribution functions for a plurality of road surfaces which represent respective adhesive effects between the vehicle tire of a driven vehicle and the vehicle road;

(b) setting a count to an initial setting;

(c) obtaining vehicle operating characteristic data;

(d) determining an index based on the vehicle operating characteristic data;

(e) applying a probability analysis for each adhesive condition road surface as a function of the determined index;

(f) recursively updating the probability analysis for each respective road surface adhesive condition as a function of the count;

(g) determining whether the count is less than a predetermined number;

(h) repeating steps (c)-(g) if the count is less than the predetermined count;

(i) determining a likelihood factor for each road surface adhesive condition as a function of each respective recursively updated probability analysis, each respective road surface adhesive condition having a respective likelihood factor that identifies the likelihood of the road surface having the respective adhesive condition as a function of the index;

(j) selecting the respective likelihood factor that identifies the adhesive condition of the road surface traveled by the vehicle; and

(k) inputting the identified adhesive condition of the road surface and the respective likelihood factor into a vehicle control process.

13. The method of claim 12 wherein steps (a)-(j) are repeated after inputting the identified adhesive condition into the vehicle control process.

14. The method of claim 12 wherein the likelihood factor is output as a probability percentage of the road surface having the adhesive condition.