Empty seat detection via spectral analysis

• US 7,577,506 B2
• Filed: 07/25/2005
• Issued: 08/18/2009
• Est. Priority Date: 07/25/2005
• Status: Active Grant

• Current Assignee: Robert Bosch Corporation (Robert Bosch GmbH), Robert Bosch GmbH
• Original Assignee: Robert Bosch GmbH
• Inventors: Fleet, Chad Van
• Primary Examiner(s): Jeanglaud; Gertrude Arthur
• Application Number:
  US11/188,318
• Publication Number:
  US 20070021891A1
• Time in Patent Office: 1,485 Days
• Field of Search: 701/45, 701/46, 280/734, 280/735, 180/268, 180/271, 180/272, 180/273
• US Class Current: 701/45
• CPC Class Codes: B60R 21/01516 using force or pressure sensing means

Abstract

A module and method for determining whether a seat is empty. In one embodiment, the module includes a projection module that is configured to receive force data that includes a plurality of modes. The projection module is also configured to isolate data for one of the modes. A transform module is configured to communicate with the projection module, and to transform the data for one of the modes into a frequency domain, thus generating spectral mode data. In addition, a spectral analysis module is configured to communicate with the transform module, and to classify the spectral mode data.

Claims

1. A module for determining whether a seat is empty, the module comprising:

a projection module configured to receive force data that includes a plurality of modes, and to isolate data for one of the modes;

a transform module configured to communicate with the projection module, and to transform the data for one of the modes into a frequency domain, thus producing spectral mode data; and

a spectral analysis module configured to communicate with the transform module, and to classify the spectral mode data.

2. The module of claim 1 further comprising a filter module configured to communicate with the transform module and the spectral analysis module, and to filter the transformed mode data.

3. The module of claim 1, wherein the spectral analysis module includes a probabilistic model.

4. The module of claim 3, wherein the spectral analysis module is further configured to classify the spectral mode data by:

creating a plurality of probability vectors according to the probabilistic model;

determining a plurality of joint probability vectors;

determining a log-likelihood of the joint probability vectors; and

applying a classification rule to the log-likelihood to classify the spectral mode data.

5. The module of claim 1, wherein the plurality of modes include vibrational modes.

6. The module of claim 1, wherein one of the plurality of modes includes a vertical mode.

7. A system for determining a weight of an object on a seat, the system comprising:

a sensor configured to detect a force exerted by an object;

a first module configured to use spectral analysis to determine an offset of the sensor; and

a second module configured to determine the weight of the object as a function of the force exerted on the seat and the offset of the sensor.

8. The system of claim 7, wherein the second module determines the weight of the object by subtracting the offset from the force exerted on the seat.

9. The system of claim 7, wherein the force includes a plurality of modes, and the first module includes:

a projection module in communication with the a plurality of sensors, and configured to receive weight data that includes a plurality of modes and isolate data for one of the modes;

a transform module in communication with a linear projection module, and configured to transform the data for one of the modes into a frequency domain, thus producing spectral mode data; and

a spectral analysis module in communication with the transform module, and configured to classify the spectral mode data.

10. The system of claim 9, wherein the first module further includes a filter module in communication with the transform module and the spectral analysis module and configured to filter the transformed mode data.

11. A method for determining whether a seat is empty, the method comprising:

isolating one of a plurality of modes of force data;

transforming one of the modes of the force data into a frequency domain, thus producing transformed mode data; and

performing spectral analysis on the transformed mode data to classify the transformed mode data.

12. The method of claim 11, wherein the transformed mode data is classified as “

empty.”

13. The method of claim 11, wherein the transformed mode data is classified as “

not empty.”

14. The method of claim 11, wherein the transformed mode data is classified as “

unknown.”

15. The method of claim 11 further comprising filtering the transformed mode data.

16. The method of claim 15, wherein spectral analysis is performed according to a probabilistic model.

17. The method of claim 16, wherein performing spectral analysis further includes:

creating a plurality of probability vectors according to the probabilistic model;

determining a plurality of joint probability vectors;

determining a log-likelihood of the joint probability vectors; and

applying a classification rule to the log-likelihood to classify the spectral mode data.

18. A computer readable storage medium comprising computer-executable instructions for analyzing a force on a seat by:

isolating one of a plurality of modes of force data;

transforming one of the modes of the force data into a frequency domain, thus producing transformed mode data; and

performing spectral analysis on the transformed mode data to classify the transformed mode data.

19. A signal including representations of computer-executable instructions for analyzing a force on a seat by:

isolating one of a plurality of modes of force data;

transforming one of the modes of the force data into a frequency domain, thus producing transformed mode data;

performing spectral analysis on the transformed mode data to classify the transformed mode data; and

determining the weight of the object as a function of the total force and the force that is not due to the weight of the object.

20. An device for analyzing a load on a surface, the device comprising:

a projection module configured to receive force data that includes a plurality of modes and to isolate data for one of the modes;

a transform module configured to communicate with the projection module and to transform the data for one of the modes into a frequency domain, thus producing spectral mode data; and

a spectral analysis module configured to communicate with the transform module and to classify the spectral mode data.

21. The device of claim 20 further comprising a filter module configured to communicate with the transform module and the spectral analysis module, and to filter the transformed mode data.

22. The device of claim 20, wherein the spectral analysis module includes a probabilistic model.

23. The device of claim 22, wherein the spectral analysis module is further configured to classify the spectral mode data by:

creating a plurality of probability vectors according to the probabilistic model;

determining a plurality of joint probability vectors;

determining a log-likelihood of the joint probability vectors; and

applying a classification rule to the log-likelihood to classify the spectral mode data.

24. The device of claim 20, wherein the plurality of modes include vibrational modes.

25. The module of claim 20, wherein one of the plurality of modes includes a vertical mode.

26. A vehicle having a system to detect the presence and weight of an occupant, the system comprising:

a projection module configured to receive force data that includes a plurality of modes, and to isolate data for one of the modes;

a transform module in communication with the projection module, and configured to transform the data for one of the modes into a frequency domain, thus producing spectral mode data; and

a spectral analysis module in communication with the transform module, and configured to classify the spectral mode data.

27. A vehicle of claim 26 further comprising:

a seat;

a control unit configured to generate an output; and

at least one sensor configured to sense a weight of the seat and, when present, a weight of an object in the seat and to communicate with the control unit.

28. A vehicle of claim 27 further comprising an occupant restraint system configured to receive a command signal based on the output of the control unit.