Continuous collision severity prediction

Details Continuous collision severity prediction Continuous collision severity prediction

2002-11-26

2004-11-09

Zanelli, Michael J. (Department: 3661)

Data processing: vehicles, navigation, and relative location

Vehicle control, guidance, operation, or indication

Vehicle subsystem or accessory control

C280S735000

Reexamination Certificate

active

06816766

ABSTRACT:

BACKGROUND OF THE INVENTION
The present disclosure relates generally to predicting collision severity and in particular, to predicting collision severity in a continuous manner, using acceleration signals, for controlling smart occupant protection systems, such as an airbag activation device.
Safety devices for the protection of the operator and passengers of automotive vehicles have been in use for many years. Many safety features function in a collision situation without external activation. Seat reinforcement, seat headrests, and passenger compartment padding are examples of such safety items. Other safety devices such as supplemental inflatable restraints, popularly known as air bags, require external activation when a collision event is apparently occurring.
Air bags comprise an inflatable bag, an electrically actuated igniter and a gas generator. Each bag is folded and stored with its igniter and gas generator in the steering wheel pad, instrument panel, door panel or body pillar. Air bags also require a collision detection system that determines when the bags should be deployed and signals the ignition of one or more charges (or stages) of the gas generator. Current air bag, and other passive occupant protection systems, rely on acceleration sensors (detecting abrupt vehicle deceleration) and a microprocessor based controller. An acceleration sensor is a device that continually senses accelerative forces and converts them to electrical signals. The controller continually receives acceleration signals from each sensor and processes them to determine whether a collision situation is occurring that requires air bag deployment.
The content of such a collision detection system for safety device actuation usually depends upon the method or algorithm used by the controller for assessing collision severity. Most systems rely on an acceleration sensor placed in the passenger compartment, close to the center of gravity of the vehicle. This sensor is often put under the passenger seat as part of a sensing and diagnostic module (SDM) of the vehicle collision sensing system. In addition, some systems place one or more accelerometers at the center or sides of the radiator cross-tie-bar to detect vehicle front-end deceleration indicative of a collision. These front-end accelerometers have been called electronic front sensors (EFS). The collision detection controller receives signals from the acceleration sensor(s) and evaluates them in a pre-programmed manner to determine whether air bag deployment is necessary. The program may also determine the ignition of one or more charges (or stages) of the gas generator of the air bag.
Collision detection algorithms have involved increasing degrees of complexity. Acceleration values from a single sensor (e.g., the SDM sensor) have simply been compared with a pre-determined threshold acceleration value as a test for device deployment. Values from more than one sensor location have been used in the collision sensing practices. Acceleration values have been integrated over time to yield crush velocities, and further integrated to yield crush displacement values. Further, the derivative of acceleration values have been determined as “jerk” values. Such velocity and displacement values, and jerk values, have also been compared with respective pre-determined threshold values as a more selective basis for achieving timely air bag deployment. They also have been used in combination with seat occupancy information and seat belt usage.
There are variants in vehicle front-end collision modes and, of course, there can be considerable variation in the severity of a collision depending upon the construction and mass of the striking vehicle and the struck vehicle or object, and their relative velocity and moving direction at the onset of a collision. With respect to front-end collision modes, a vehicle may collide head-on with another vehicle or a stationary object such as a rigid wall, either fully or partially overlapped (called a full frontal collision mode and an offset frontal collision mode, respectively), or with a narrower stationary object such as a pole or a tree (called a frontal pole collision mode). Front-end collision of a vehicle with other vehicle or stationary object in angle is called an angular collision mode.
Actual vehicular collision testing reveals different patterns of front-end and passenger compartment crush velocities and displacements associated with different collision modes. In fact, considerable collision testing of a vehicle has been required to provide the substantial database of threshold values of jerk, acceleration, velocity and/or displacement over a collision period for use by a collision-sensing controller. Such data must be compiled from suitably instrumented test vehicles over the relevant duration of each test collision. Depending upon the nature and severity of a collision, an airbag deployment decision may be made by the controller process at any time during a period of from about 15 milliseconds to 100 milliseconds or so from the onset of the collision. A controller for a dual stage airbag system can be instructed to inflate the airbag to one of two levels. In contrast, a continuous flow airbag can be instructed to inflate the airbag to any level based on input to a variable output inflator.
BRIEF DESCRIPTION OF THE INVENTION
In an exemplary embodiment, the method comprises receiving vehicle acceleration data. A collision event is detected in response to the vehicle acceleration data. A collision mode is determined in response to detecting the collision event. Input to the collision mode determination includes the acceleration data. A collision severity value responsive to the acceleration data and to the collision mode is calculated. The collision severity value corresponds to a percentage inflation level of an airbag.
In another aspect, a method for continuous collision severity prediction comprises receiving vehicle acceleration data. A collision event is detected in response to receiving the vehicle acceleration data. An airbag deployment request is transmitted to an airbag controller in response to the detecting. A collision mode is determined in response to detecting the collision event. Input to the collision mode determination includes the acceleration data. A collision severity value responsive to the acceleration data and the collision mode is calculated. The collision severity value corresponds to a percentage inflation level of an airbag. The collision severity value is transmitted to the airbag controller.
In a further aspect, a system for continuous collision severity prediction comprises an airbag controller, a SDM accelerometer, an EFS accelerometer and a collision detection controller. The collision detection controller is in communication with the airbag controller, the SDM accelerometer and the EFS accelerometer. The collision detection controller includes instructions to implement a method comprising receiving vehicle acceleration data. A collision event is detected in response to receiving the vehicle acceleration data. A collision mode is determined in response to detecting the collision event. Input to the collision mode determination includes the acceleration data. A collision severity value responsive to the acceleration data and the collision mode is calculated. The collision severity value corresponds to a percentage inflation level of an airbag.


REFERENCES:
patent: 5411289 (1995-05-01), Smith et al.
patent: 5732374 (1998-03-01), Ohm
patent: 5967548 (1999-10-01), Kozyreff
patent: 6186539 (2001-02-01), Foo et al.
patent: 6424898 (2002-07-01), Anishetty et al.
patent: 6424899 (2002-07-01), Imai et al.
patent: 6459975 (2002-10-01), Kuhn et al.
patent: 6499763 (2002-12-01), Mishima
patent: 6532408 (2003-03-01), Breed
patent: 6540255 (2003-04-01), Blank et al.
patent: 6549836 (2003-04-01), Yeh et al.
patent: 6647331 (2003-11-01), Imai et al.
patent: 2004/0056459 (2004-03-01), Kassman et al.
patent: 4214222 (1993-04-01), None
patent: 0900702 (1999-03-01), None

Affiliated with

Also associated with

Rating

Continuous collision severity prediction does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Continuous collision severity prediction, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Continuous collision severity prediction will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3361821