Methods of occupancy state determination and computer programs

• US 5,890,085 A
• Filed: 10/11/1996
• Issued: 03/30/1999
• Est. Priority Date: 04/12/1994
• Status: Expired due to Term

• Current Assignee: Robert Bosch Corporation (Robert Bosch GmbH)
• Original Assignee: Robert Bosch Corporation (Robert Bosch GmbH)
• Inventors: Corrado, Anthony P., Decker, Stephen W., Benbow, Paul K.
• Primary Examiner(s): Cuchlinski, Jr., William A.
• Assistant Examiner(s): Pipala, Edward J.
• Application Number:
  US08/731,355
• Time in Patent Office: 900 Days
• Field of Search: 701/45, 701/47, 280/734, 280/735, 307/10.1, 364/550, 364/551.01, 180/271, 180/272
• US Class Current: 701/47
• CPC Class Codes: B60N 2/002 Seats provided with an occupancy detection means mounted therein or thereon detection means responsive to presence or absence of children B60N2/266
• CPC Class Codes: B60R 21/01532 using electric or capacitive field sensors
• CPC Class Codes: B60R 21/01534 using electromagneticwaves, e.g. infrared
• CPC Class Codes: B60R 21/01536 using ultrasonic waves
• CPC Class Codes: B60R 21/0154 in combination with seat heating
• CPC Class Codes: G01S 15/04 Systems determining presence of a target
• CPC Class Codes: G01S 15/523 for presence detection burglar, theft or intruder alarms G08B13/00, e.g. G08B13/16
• CPC Class Codes: G01S 15/86 Combinations of sonar systems with lidar systems; Combinations of sonar systems with systems not using wave reflection
• CPC Class Codes: G01S 15/88 Sonar systems specially adapted for specific applications seismic or acoustic prospecting or detecting G01V1/00
• CPC Class Codes: G01S 7/526 Receivers
• CPC Class Codes: G01S 7/539 using analysis of echo signal for target characterisation; Target signature; Target cross-section
• CPC Class Codes: G06F 18/256 of results relating to different input data, e.g. multimodal recognition
• CPC Class Codes: G06V 10/147 Details of sensors, e.g. sensor lenses fingerprint or palmprint sensors G06V40/13; vascular sensors G06V40/145; eye sensors G06V40/19
• CPC Class Codes: G06V 10/803 of input or preprocessed data
• CPC Class Codes: G06V 20/593 Recognising seat occupancy
• CPC Class Codes: G06V 40/10 Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands

Abstract

A system for sensing the presence, position and type classification of an occupant in a passenger seat of a vehicle, as well as for sensing the presence of a rear-facing child seat therein, for use in controlling a related air bag activator control system to enable, disable or control inflation rate or amount of inflation of an air bag. The sensor system employs sensor fusion, a process of combining information provided by two or more sensors (24, 26), each of which "sees" the world in a unique sense. In a preferred embodiment, infrared sensor inputs (78) and ultrasonic sensor inputs (79) are combined in a microprocessor by a sensor fusion algorithm (80) to produce an occupancy state output signal (85) to the air bag controller.

Claims

1. Method of defining occupancy states for at least one interior space comprising the steps of:

(a) obtaining at least one signal from at least one first type of sensor, which signal is characteristic of said space, and which space has a predetermined configuration of objects therein and is defined as a space-empty state;

(b) processing said at least one first sensor characteristic signal to generate a plurality of first sensor feature signals;

(c) obtaining at least one signal from at least one second type of sensor, which signal is characteristic of said space-empty state;

(d) processing said at least one second sensor characteristic signal to generate a plurality of second sensor feature signals;

(e) fusing feature signals selected from said first feature signals, said second feature signals and combinations thereof, to produce at least one fused feature signal having at least one value characteristic of said space-empty space;

(f) repeating steps a-e on each one of a plurality of known objects placed in said spaces to produce a corresponding plurality of fused feature signal values, each of which fused feature signal values is characteristic of said objects associated therewith, to define a series of space-occupied state values;

(g) storing said characteristic values of said space-empty state and said space-occupied state fused feature signal values as a set of weighted occupancy state values, which when subsequently compared to fused feature signal values of at least one sensed unknown object are used to generate probability confidence levels selected from at least one of the nature and location of said unknown object in said space.

2. Method of constructing a weighted characteristic matrix for use in at least one of defining and determining occupancy states for at least one space comprising the steps of:

(a) obtaining at least one signal, from at least one of a first type of sensor, characteristic of said space having therein a predetermined configuration of objects, said characteristic signal defining a first sensor space-empty signal profile;

(b) processing said at least one first sensor signal to generate a plurality of first sensor space-empty feature signals;

(c) obtaining at least one signal from at least one of a second type of sensor characteristic of said space-empty state defining a second sensor space-empty signal profile;

(d) processing said at least one second sensor signal to generate a plurality of second sensor space-empty feat signals;

(e) fusing a plurality of said space-empty feature signals selected from said first and said second space-empty feature signals to produce at least one fused feature signal having at least one characteristic value defining a space-empty state;

(f) repeating steps (a)-(e) on each member of a plurality of known objects in said space to produce a corresponding plurality of fused feature signal characteristic values each of which corresponds to at least one of said known objects associated therewith defining a series of space-occupied states;

(g) ordering said characteristic known state values in a confidence weighted matrix of occupancy state values;

(h) storing said matrix for comparison to fused feature signal values of at least one sensed unknown object to generate probability confidence levels selected from at least one of the nature and location of said sensed unknown object in said space.

3. A method as in claim 2 which includes the step of:

(a) fusing a first raw signal from at least one of said first and second types of sensors with at least one feature signal selected from raw and processed feature signals from at least one of the sensors.

4. A method for determining the occupancy state of a vehicle interior, comprising the steps of:

(a) providing a plurality of sensors, each of said sensors providing quantifiable data concerning an aspect of the contents of said interior selected from at least one of the nature and location of said contents, and;

(b) comparing said data to a multidimensional confidence criteria matrix, each element of said confidence criteria matrix representing a probability that data from a particular sensor corresponds to a known aspect of the contents of said vehicle interior for determination of said occupancy state.

5. A method as in claim 4, wherein:

(a) at least two different types of sensors are provided; and

(b) the confidence criteria matrix is a two dimensional matrix.

6. A method as in claim 4, wherein a set of aspects of the contents represented in the multidimensional confidence criteria matrix includes:

an unoccupied car seat;

an occupied car seat;

a rearward facing child seat, a blocked sensor and a forward facing child seat.

7. A method as in claim 6, wherein the set of aspects of contents represented in the multidimensional confidence criteria matrix includes variants of occupant position and occupant type.

8. A method as in claim 5, wherein the different types of sensors include an ultrasound sensor and an infrared sensor.

9. A method of constructing a multidimensional confidence criteria matrix to represent a variety of occupancy states involving different types and positions of occupants of an automobile seat, comprising:

(a) obtaining a plurality of data sets from a plurality of sensors under a plurality of different occupant type and position combinations;

(b) combining elements of said data sets to generate a weighted confidence level correlating a probability that each such element represents a particular occupant type and position combination; and

(c) assigning each of the confidence levels determined in step (b) to a position in a multidimensional confidence criteria matrix corresponding to a particular sensor and a particular occupant type and position combination.

10. A method as in claim 9 wherein said step of obtaining said data sets includes the steps of:

(a) obtaining feature signals from said sensors; and

(b) fusing said feature signals.

11. A method as in claim 10 wherein, said plurality of sensors includes at least two different type of sensors.

12. A method as in claim 11 wherein;

(a) said sensor types are selected from infrared sensors, ultrasound sensors, and capacitance sensors.

13. A method as in claim 10 wherein said data sets are obtained empirically from known occupancy states.