Method of distinguishing between adult and cinched car seat occupants of a vehicle seat

• US 7,233,852 B2
• Filed: 05/06/2005
• Issued: 06/19/2007
• Est. Priority Date: 05/06/2005
• Status: Expired due to Fees

• Current Assignee: Delphi Technologies, Inc. (BorgWarner Incorporated)
• Original Assignee: Delphi Technologies, Inc. (BorgWarner Incorporated)
• Inventors: Murphy, Morgan D., Prieto, Raymundo, Wagner, Steven J.
• Primary Examiner(s): Camby; Richard M.
• Application Number:
  US11/123,419
• Publication Number:
  US 20060253238A1
• Time in Patent Office: 774 Days
• Field of Search: 701/45, 701/46, 701/47, 180/271, 180/282, 280/734, 280/735
• US Class Current: 701/45
• CPC Class Codes: B60R 2021/01325 Vertical acceleration
• CPC Class Codes: B60R 21/01556 Child-seat detection systems

Abstract

A normally seated adult occupant of a vehicle seat is distinguished from a cinched car seat occupant based on the static seat force and first and second sets of identified dynamic characteristics that are respectively correlative of a normally seated adult and a cinched car seat. Confidence levels are assigned to the static seat force and each of the identified dynamic characteristics, and overall confidence levels for the two classifications are determined based on the assigned confidence levels. The overall confidence level for the adult occupant classification is compared with the overall confidence level for the cinched car seat classification, and the occupant is classified based on the highest overall confidence level.

Claims

1. A method of distinguishing between a normally seated adult and a cinched car seat in an occupant classification system for a seat of a vehicle, the method comprising the steps of:

measuring parameters during movement of said vehicle including a force applied to said seat and a z-axis acceleration of said vehicle;

determining an average of the measured seat force;

determining one or more dynamic characteristics correlative of a normally seated adult occupant of said seat based on said measured parameters;

determining one or more dynamic characteristics correlative of a cinched car seat occupant of said seat based on said measured parameters;

computing a first overall confidence level in favor of a normally seated adult occupant based on the average seat force and the dynamic characteristics correlative of a normally seated adult occupant;

computing a second overall confidence level in favor of a cinched car seat occupant based on the average seat force and the dynamic characteristics correlative of a cinched car seat occupant;

determining that a normally seated adult occupant is present if the first overall confidence level is greater than the second overall confidence level; and

determining that a cinched child seat is present if the second overall confidence level is greater than the first overall confidence level.

2. The method of claim 1, including the steps of:

assigning confidence levels to the average seat force in favor of a normally seated adult occupant and to each of the dynamic characteristics correlative of a normally seated adult occupant; and

computing said first overall confidence level based on the assigned confidence levels.

3. The method of claim 2, where there are at least two dynamic characteristics correlative of a normally seated adult occupant and the method includes the steps of:

computing an average of the confidence levels assigned to said at least two dynamic characteristics; and

computing said first overall confidence level based on the computed average and the confidence level assigned to said average seat force.

4. The method of claim 1, including the steps of:

assigning confidence levels to the average seat force in favor of a cinched car seat occupant and to each of the dynamic characteristics correlative of a cinched car seat occupant; and

computing said second overall confidence level based on the assigned confidence levels.

5. The method of claim 4, where there are at least two dynamic characteristics correlative of a cinched car seat occupant and the method includes the steps of:

computing an average of the confidence levels assigned to said at least two dynamic characteristics; and

computing said second overall confidence level based on the computed average and the confidence level assigned to said average seat force.

6. The method of claim 1, including the steps of:

determining Mahalanobis distances for the average seat force and for said dynamic characteristics correlative of a normally seated adult occupant;

using the determined Mahalanobis distances to assign confidence levels to the average seat force in favor of a normally seated adult occupant and to each of the dynamic characteristics correlative of a normally seated adult occupant; and

computing said first overall confidence level based on the assigned confidence levels.

7. The method of claim 1, including the steps of:

determining Mahalanobis distances for the average seat force and for said dynamic characteristics correlative of a cinched car seat occupant;

using the determined Mahalanobis distances to assign confidence levels to the average seat force in favor of a cinched car seat occupant and to each of the dynamic characteristics correlative of a cinched car seat occupant; and

computing said second overall confidence level based on the assigned confidence levels.

8. The method of claim 1, including the steps of:

determining probability distribution function values for the average seat force and for said dynamic characteristics correlative of a normally seated adult occupant;

using the determined probability distribution function values to assign confidence levels to the average seat force in favor of a normally seated adult occupant and to each of the dynamic characteristics correlative of a normally seated adult occupant; and

computing said first overall confidence level based on the assigned confidence levels.

9. The method of claim 1, including the steps of:

determining probability distribution function values for the average seat force and for said dynamic characteristics correlative of a cinched car seat occupant;

using the determined probability distribution function values to assign confidence levels to the average seat force in favor of a cinched car seat occupant and to each of the dynamic characteristics correlative of a cinched car seat occupant; and

computing said second overall confidence level based on the assigned confidence levels.

10. The method of claim 1, including the steps of:

determining quality loss function values for the average seat force and for said dynamic characteristics correlative of a normally seated adult occupant;

using the determined quality loss function values to assign confidence levels to the average seat force in favor of a normally seated adult occupant and to each of the dynamic characteristics correlative of a normally seated adult occupant; and

computing said first overall confidence level based on the assigned confidence levels.

11. The method of claim 1, including the steps of:

determining quality loss function values for the average seat force and for said dynamic characteristics correlative of a cinched car seat occupant;

using the determined quality loss function values to assign confidence levels to the average seat force in favor of a cinched car seat occupant and to each of the dynamic characteristics correlative of a cinched car seat occupant; and

computing said second overall confidence level based on the assigned confidence levels.