Electrical transmission or interconnection systems – Vehicle mounted systems – Automobile
Reexamination Certificate
2000-08-25
2002-11-05
Fleming, Fritz (Department: 2836)
Electrical transmission or interconnection systems
Vehicle mounted systems
Automobile
C307S009100
Reexamination Certificate
active
06476516
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates to a method and apparatus for classifying the weight of a seat occupant. Specifically, an estimated seat occupant weight is compared to a series of moving thresholds to compile a weight class history that provides a more accurate, stable, and robust weight classification for airbag deployment decisions.
2. Related Art
Most vehicles include airbags for the driver and passenger. It is important to control the deployment force of the airbags based on the size of the driver or the passenger. One way to control the deployment force is to monitor the weight of the seat occupant. If a smaller person such as a child or infant in a car seat is in the passenger seat, the weight on the seat will be less than if an adult occupies the seat.
Current systems for measuring the weight of a seat occupant are complex and expensive. Sensors are placed at a plurality of locations in the seat bottom and the combined output from the sensors is used to determine the weight of the seat occupant. Each sensor experiences a substantially vertical force, due to the weight of the seat occupant, but is also subject to longitudinal and lateral forces caused by acceleration, deceleration, turning, or adverse road conditions. The lateral and longitudinal forces picked up by the sensor incorporate an error component into the weight measurement. These sensors often cannot correct error due to changes in occupant seating position or adverse road conditions.
Some systems attempt to classify seat occupants into predetermined customer-specified classes usually based only on occupant weight. The classification information is then used to modify the deployment of the airbag. These systems do not provide accurate and consistent classification over a wide range of adverse road conditions and/or occupant seating conditions.
Thus, it is desirable to have an improved seat occupant classification system that provides accurate classification by eliminating error caused by adverse road conditions and changes in occupant position.
SUMMARY OF THE INVENTION
A method and apparatus is provided that classifies a seat occupant into one of several different weight classes based on an estimated value of the seat occupant weight. Each of the weight classes has upper and lower thresholds that define the class. Over time, several comparisons are made between the estimated weight and the thresholds of the weight classes and each comparison results in a weight class sample. The seat occupant is assigned a specific weight class designation once a predetermined number of consistent and consecutive weight class samples is achieved. The specific weight class designation remains locked until a certain number of inconsistent weight class samples are observed.
In a disclosed embodiment of this invention, the method for classifying a seat occupant into a weight class includes the following steps. The seat occupant weight is measured resulting in an estimated weight. The estimated weight is compared to a series of weight classes with thresholds to determine a class sample. The previous steps are repeated until a predetermined number of class samples having the same value is achieved and the class sample becomes locked as the occupant weight class.
Additional steps include generating an occupant weight class signal corresponding to the locked occupant weight class, transmitting the occupant weight class signal to a control unit, and modifying deployment of an airbag based on the occupant weight class signal. The weight class is unlocked when a predetermined number inconsistent class samples is observed. When the class is unlocked, the process repeats.
Once the occupant has been classified into a weight class, that class becomes the known class for the next comparison. Preferably, each weight class is assigned an upper threshold and a lower threshold. At each iteration, the estimated weight is compared to the upper and lower thresholds for the last known weight class. The new class sample is designated the same as the last known weight class if the estimated weight is between the upper and lower thresholds for the last known weight class. The sample is set equal to a next higher weight class if the estimated weight is greater than the upper threshold for the last known weight class or the class sample is set equal to a next lower weight class if the estimated weight is less than the lower threshold for the last known weight class.
In one disclosed embodiment, the value of the upper threshold of the class sample is increased by a first predetermined amount and the value of the lower threshold of the class sample is decreased by a second predetermined amount after the class sample is locked. The upper and lower thresholds are returned to their initial values when the class sample becomes unlocked.
The occupant weight classification system includes a sensor assembly for measuring the weight of a seat occupant to generate an estimated weight signal. A series of weight class data is also included where each weight class has an upper threshold and a lower threshold. The system includes a control unit for receiving the estimated weight signal, comparing the signal to the upper and lower thresholds to assign the signal an appropriate weight class designation, and for locking the signal into an occupant specific weight class when a predetermined number of consistent and consecutive weight class designations is achieved. The control unit generates and transmits a control signal to an airbag controller to modify deployment of the airbag based on the occupant specific weight class.
The subject invention uses varying weight class thresholds and class sample histories to produce a more stable, accurate and robust classification process that reduces errors caused by changes in occupant seating position and adverse road conditions. The more accurate classification system is used to generate control signals, which are used to modify airbag deployment.
REFERENCES:
patent: 5413378 (1995-05-01), Steffens
patent: 5474327 (1995-12-01), Schousek
patent: 5539290 (1996-07-01), Lu et al.
patent: 5573269 (1996-11-01), Gentry
patent: 5626359 (1997-05-01), Steffens
patent: 5732375 (1998-03-01), Cashler
patent: 5821633 (1998-10-01), Burke
patent: 5906393 (1999-05-01), Mazur
patent: 5987370 (1999-11-01), Murphy et al.
patent: 6070115 (2000-05-01), Oestreicher et al.
patent: 0 656 283 (1997-08-01), None
patent: 97 10115 (1997-03-01), None
patent: 98 14345 (1998-04-01), None
patent: WO99/24285 (1999-05-01), None
patent: 99 38731 (1999-08-01), None
Written Opinion dated Aug. 17, 2001 regarding International Application No. PCT/US00/23681.
DeBeradinis Robert L
Fleming Fritz
Siemens VDO Automotive Inc.
LandOfFree
Method and apparatus for classifying seat occupant weight does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for classifying seat occupant weight, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for classifying seat occupant weight will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2974631