AIRBAG DEPLOYMENT DIRECTIONAL CONTROLLER
This invention relates to a vehicle occupant restraint system that controls the direction of deployment of an airbag m accordance with a vehicle occupant's seated height and/or projected trajectory. A vehicle occupant restraint system' s purpose is to reduce and possibly eliminate injury to a vehicle occupant in a crash. The addition of airbags to an automobile's safety restraint system has significantly enhanced the overall protection extended to a vehicle occupant. However, under certain circumstances the vehicle occupant may be injured by the airbag. Airbags are inflated very rapidly and could cause bodily injury to the vehicle occupant if the vehicle occupant interacts with the airbag while it is being deployed. Out of position vehicle occupants are most at risk of injury from inflating airbags. Vehicle occupants can be "out of position" by not wearing their safety belts during a crash, sitting too close to the steering wheel/instrument panel, delayed triggering of the airbag, having short seated height, or slumping forward.
There has been a huge impetus in the industry to develop a "smart" or "intelligent" safety system which customizes airbag deployment based upon the nature of the crash along with measured information about the vehicle occupant such as height, position, and motion, rather than solely on the magnitude of negative acceleration or deceleration of the vehicle. A smart airbag system would operate by processing signals from various sensors and then transmitting a specific
command to the airbag module. Depending on the circumstances of -;he crash, the controller can signal the airbag module not to deploy an airbag, to deploy the airbag partially (e.g. fill only one chamber of the airbag) , or to completely inflate the airbag.
Airbags typically require about 30 milliseconds to achieve full inflation and should be fully deployed before a vehicle occupant interacts with the airbag. The airbag may injure a vehicle occupant if the vehicle occupant interacts with the airbag before the airbag is fully deployed. Shorter vehicle occupants are more susceptible to being injured by moving into the path of a deploying airbag as opposed to taller vehicle occupants because of the deployment direction of the airbag. The airbag module is designed so the airbag is deployed at a direction towards the chest/sternum area of a 50th percentile adult male (1.75 meters, 77 kilograms), where forces can be absorbed most readily due to natural stiffness and support in the human body. Shorter vehicle occupants are more susceptible to injury because the airbag is deployed in the direction of their head and chin as opposed to their chest/sternum area possibly causing hyperextension of the cervical region of the spine. Airbag systems are designed so the airbag will be deployed toward the chest/sternum area of a 50th percentile adult male. This standard actually does not represent the 50th percentile across genders, rather it represents the 70th percentile. That is, 70% of the adult population will actually be smaller than the standard. Thus, a deployed airbag will have an initial point of contact that is higher than the
chest/sternum area. Smart airbag systems may deactivate the airbag system in low speed crashes with short occupants. The drawback in deactivating the airbag system is that the short occupant may sustain injuries by colliding with the steering wheel column or dashboard. If the airbag deployment direction could be modified for shorter occupants the chance of injury to the occupant will be reduced. In US 5 330 226 the idea of changing the deployment direction is contemplated, "If the airbag is capable of being aimed, it is desirable that the direction of deployment be controlled in response to the occupant's position." However, US 5 330 226 does not teach or suggest a means of changing the deployment angle of the airbag, rather the patent mentions the possible benefit of changing the deployment angle. This patent teaches an apparatus for controlling actuation of an airbag, comprising infrared sensors, ultrasound sensors, and crash sensors. US 5 360 231 teaches a rotatable occupant restraint wherein during deployment of the bag, the enclosure rotates about a horizontal extending axis into an activated position for direct deployment of the bag between the support structure and a vehicle occupant in the passenger seat. The purpose of the invention is to deploy an airbag in the direction of the vehicle occupant in situations where the dashboard is angled and does not concern changing the deployment angle to reduce injury to the shorter stature occupants.
The present invention as set forth in claim 1 provides a vehicle restraint system, which reduces
injury to vehicle occupants by adjusting the direction an airbag deployed in a crash.
Brief Description of Drawings
FIG. 1 shows a relationship between stature and weight among adult male and females. (National Aeronautics and Space Administration, Anthropometric Source Book. NASA Defense Publication No. 1024, 1978)
FIG. 2 is a schematic view of two airbags deployed in different directions.
FIG. 3 is a schematic representation of an airbag module on an adjustable platform during normal operation.
FIG. 4 is a schematic representation of an airbag module on a platform where the height of the adjustment has been altered to change the deployment angle of the airbag for shorter occupants.
FIG. 5 is a schematic representation of an airbag module capable of being adjusted by rotating the airbag module.
FIG. 6 is a schematic representation of an airbag module, which has been rotated to yield a modified deployment direction for an airbag.
FIG. 7 is a graph of chest compression for the baseline (dash lines) and tilt module (solid line) tests. Chest compression (cm) is plotted as a function of time.
FIG. 8 is a graph of the head triaxial resultant acceleration for the baseline (dash lines) and tilt module (solid lines) tests. The resultant plot combines the acceleration of the three different directional components and is used for the determination of the HIC.
FIG. 9 is a ,raph of the y-component flexion- extension bending moments for the baseline (dash lines) and tilt module (solid lines) tests.
Detailed Description of the Invention
The invention relates to a vehicle restraint system for actuating an airbag for protecting a vehicle occupant. The restraint system comprises an airbag module 8, which is in communication with a means for adjusting the deployment direction for the airbag. The restraint system also comprises a controller that analyzes a signal communicated thereto by a sensor which estimates the seated height of the vehicle occupant, and the controller is also in communication with the means for adjusting the deployment direction for the airbag and signals the adjustment means 5 to adjust the deployment direction for the airbag in accordance with a vehicle occupant's seated height.
One aspect of the invention is to adjust the deployment direction of the airbag in accordance with the seated height of the vehicle occupant. The seated height of the vehicle occupant can be indirectly approximated with reasonable confidence by determining the mass of a seated occupant . Another aspect of the invention is to adjust the deployment direction of the airbag in accordance with the vehicle occupant's weight. A seat occupant sensing system utilizing at least two load cells for determining the mass of the occupant is described in US 5 810 392. There are a number of other methods for measuring occupant weight, including pressure foils, inflated bladders, and strain gauges. The microprocessor can process the signal from the load cells utilizing the stature and weight distribution correlation to produce an approximate height of the occupant. FIG. 1 is a graph
of the relationship between stature and weight among adult males and females. The elliptical figure closer to the origin of the graph represents the relationship between mass and stature for a female while the elliptical figure further away from the origin shows the relationship for a male. From stature data, a reasonable approximation for the seated height of a vehicle occupant can be calculated.
A means of directly determining a vehicle occupant's seated height is computed by processing readings from headliner-to-head infrared (IR) sensors and the headrest infrared sensors rather than load cells juxtaposed between a rigid member and a seat pan member. The headliner and the headrest infrared sensors work by oscillating an infrared light emitting diode (LED) with a 38 kHz, 50% duty cycle square wave and comparing the received signal with a predetermined voltage level . Most automobile manufacturers have headliner to head distances of 7.6 TO 12.7 cm for the 50th percentile male's seated height. The IR sensors are tailored to have maximum voltage differences in this range, thus allowing fine-tuning of the module. The sensors can be altered to have a collimated angle of dispersion between 25 degrees and 45 degrees. This angle allows for high sensitivity within certain area of movement .
An exemplary IR sensor system uses a 555 timer to oscillate an IR LED at 38 KHz. The 38 KHz signal makes contact with a seated occupant and is reflected back toward the integrated receiver/filter/signal conditioner. The GP1U52X Infrared Receiver/Demodulator is a hybrid integrated
circuit/infrared detector circuit and uses a pin photo diode that has a peak sensitivity in the near infrared range. A built in filter blocks visible light to reduce or eliminate false operation caused by other light sources. The band pass filter then rejects all signals outside the pass band (40 kHz ± 4 kHz) . The remaining signal is fed to the demodulator, integrator, and wave-shaper circuit. The sensor then transmits an analog signal to an airbag controller for analysis of the reflected IR signal. A plurality of separate IR sensors could be used to determine the seated height of the occupant as described, for example, in US 5 330 226.
While the embodiment disclosed herein was reduced to practice employing infrared sensors to detect seated height of a vehicle occupant, it is understood that any suitable vehicle occupant sensing system that has the capacity to detect these variables may be used in the practice of the present invention. Other suitable types of vehicle occupant sensing systems that may be used in the practice of the present invention include, but are not limited to: optical, ultrasonic, infrared, and/or microwave emission and absorption devices as disclosed for example in airbag deployment system shown in the US 5 118 134. Another aspect of the invention is the determination of the vehicle occupant's projected trajectory. In order for the vehicle restraint system to compute the projected trajectory of the vehicle occupant, more than one vehicle occupant/crash characteristics might be measured. A simple but effective restraint system could comprise a seat
height determininc sensor in conjunction with a seat belt sensor. The input of the two parameters (height and use of restraint system) may be used to predict impact trajectory of the vehicle occupant, for the purpose of controlling the directional deployment of the airbag system (FIG. 2) .
A more complicated restraint system can be designed that collects data from several sensors. Besides sensors for height and seat belt use, the restraint system could include sensors that measure seat back angle, seat position, occupant position, vehicle temperature, and crash characteristics. A smart airbag system comprising the sensors mentioned herein is described in US 5 626 359. The use of an array of capacitive coupling proximity sensors could be integrated into a restraint system to determine the position and motion of the occupant as taught in US 5 603 734.
Another aspect of the invention is the adjustment of the orientation of the airbag module. FIG. 2 is a schematic view of the passenger compartment of an exemplary vehicle. Effective area 1 is a pictorial representation of the location of an inflated airbag deployed towards a 50th percentile male. For shorter occupants the airbag deployment direction should be adjusted as shown by effective area 2.
Sled tests were performed in which the direction of deployment for a 50th percentile male utilized was 11° above the horizontal (FIG. 3) . To represent the shorter occupant, a 5th percentile female (1.5 meters and 47.6 kilograms) was selected. An experiment was conducted to ascertain the preferred direction at
which the airbag needs to be deployed to ensure safety for a small occupant. Measurements were taken using an electric level from the top of the airbag flap to the breast line of the 5th percentile female, which yielded a preferred airbag deployment direction of 12° below the horizontal. The difference between deployment directions is 23° (FIG. 4) . It is understood that the direction of airbag deployment for a particular vehicle will be dependent upon the design and dimensions of the vehicle interior and other variables that are discussed herein.
For the conducted experiment, the binary signal from the comparator at the end of the IR sensor's signal chain was sent through a current limiting resistor to a transistor switch. The emitter of the switch was connected to an electrically controlled air valve, which was open at 12 V and closed at 0V. The air valve opens a port to an air piston 5, which lifts the platform 6 from the normal position for 50th percentile male to a raised position for the 5th percentile female.
The air tubes 7 from the piston were attached to the air valve and a source of compressed air. The platform 6 can be adjusted by pumping air into the piston 1. An alternative method for adjusting the height of the platform is by creating a vacuum system where adjustment occurs by opening the system to atmospheric gas . Most motor vehicles have ample supplies of vacuum lines to facilitate this embodiment setup, without a sacrifice of differential pressure. There are only two moving parts in this exemplary embodiment: the platform 6 and piston 5. The piston
and the pinion have a life expectancy that exceeds most car manufacturers' warranties for safety systems, thus providing a reliable and long-lasting mechanism for changing the direction of airbag deployment. Besides a pneumatic method of moving a piston, the piston could be activated hydraulically, electrically, or pyrotechnically . Another method of adjusting the orientation of the airbag could be achieved by rotating the airbag module 8 by an adjustable means 10 (FIGS. 5 and 6) .
Experiments were conducted using a HYGE Sled with a representative popular vehicle buck. The sled test crash pulse that was used was a 25-mph frontal barrier crash for this vehicle. The airbag module was the 1998 model year production, with its "depowered" inflator. The dummy used was the Hybrid III 5th percentile female.
The crash dummy was placed in the standard full forward seat position for a normally seated 5th percentile female. The test was conducted with the dummy not restrained by a seat belt . Data from the experiments were acquired using the standard accelerometers and force sensors located inside the dummy .
EXPERIMENT 1: COMPARISION OF INITIAL CONTACT AND CHEST ACCELERATION
Chest acceleration and chest compression measurements from a Hybrid III 5th percentile dummy during both baseline and tilted module sled tests were determined. The baseline test utilized a standard airbag deployment direction, which is 11° above
horizontal, while the tilted module sled test utilized an airbag with a deployment direction angle of 12° below the horizontal (a change of 23°) . All conditions for both tests were identical with the exception of the deployment angle.
The point of contact for the 5th percentile dummy should be below the sternum at about the xiphoid process area, as opposed to the baseline dummy's neck/chin area. Because the chest area is far more rigid and has more surface area than the neck area, and can therefore absorb forces more readily than the neck, the probability of severe damage is much reduced for the lower point of contact. Table 1 provides the data collected for the acceleration in three different axes (x, y, and z) on a Cartesian graph along with the resultant acceleration and maximum compressibility.
Table 1
A positive acceleration is indicative of a movement of the vehicle occupant in the direction opposite the front of the car. The largest % difference for movement along a particular axis occurred in the x direction. The maximum acceleration for both tests occurred around 70 ms into the crash, which corresponds to the time for maximum deployment. The
initial point of contact between the airbag and the dummy was different in the two tests. The tilted airbag configuration diverted the force toward the chest area while the baseline module diverted force elsewhere. Thus, the accelerometer at the chest recorded a higher positive acceleration for the tilted module than the baseline.
The difference in cushioning is illustrated in the maximum compression column (Table 1) , where the tilted module compresses the chest 2.07 cm (well below injury criteria) at 68 ms into the crash. The baseline module has a maximum compression of 5.2 mm at 43 ms into the crash (FIG. 7) . Initially, the results might lead the observer to consider the baseline test less damaging to the occupant; however, the larger chest compression in the tilt module test illustrates desired interaction with the airbag. Compressing the chest diminishes the risk of head or neck injury because initial contact with the airbag occurs with the chest, reducing the impact to the head or the neck. Table 1 shows that the maximum compressibility for the crash test dummy in the tilted module test is almost four times as large as the dummy in the baseline test. This occurs during the critical crash time of approximately 70 ms , where injury is most likely.
EXPERIMENT 2: COMPARISIONS OF HEAD AND NECK
ACCELERATIONS
Head accelerations in the three different directions differ substantially between the baseline test and the tilted module tests.
Table 2
Table 2 shows that the magnitude of head acceleration in all three directions is smaller for the tilted module test. FIG. 8 shows the Head Triaxial resultant for both tests, which combines data from accelerations in the three different directional components. Utilizing the area under the curves in FIG. 8, the head injury criteria (HIC) can be determined. The HIC is a valuable parameter in accessing vital damage potential to the head region of a vehicle occupant involved in contact with an airbag during deployment. Since 1968, the HIC parameter has been specified in the Federal Motor Safety Standard (FMVSS) 208 as one of the "performance criteria" for assessing the dummy occupant response during frontal impacts. The equation for computing HIC is as follows :
a(t) equals resultant acceleration at the head center of gravity; tl and t2 are two time points during the impact which maximize HIC; t equals time in seconds. The HIC measurement is a standard value calculated in the art .
The mathematical equation for the computation of the HIC is an integral formula, where the larger the area under the curve the larger the HIC. The calculated value for HIC for the baseline test (Table 2) is 905.8 which corresponds to about a 15 % probability of a serious head/neck injury with severe cerebrospinal damage, while a 637 HIC in the tilted test corresponds to about a 5 % probability of severe head/neck injury (%ages were determined using cumulative distribution curves by Mertz and Weber) .
EXPERIMENT 3 : COMPARISIONS OF Y-COMPONENT BENDING
MOMENTS
36% of all automobile related injuries involve the head and neck. A detailed investigation of neck moments is useful for predicting neck injuries.
Table 3
The negative values in Table 3 for the directional components correspond to a load placed on the cells in a direction away from the front of the vehicle, while the positive values correspond to the movement of the neck towards the front of the vehicle.
The likelihood of the vehicle occupant suffering neck injury can be determined by examining the neck moment along the y direction. Extension of the neck or backward movement of the head is more damaging to the neck than flexion of the neck or movement of the head forward. Extension is more injurious to the neck due to the severe angles at which the head can move backward without hindrance. Flexion movements are buffered by chin-contact and although damaging, prove to be less damaging than extension. A reduction in magnitude of the neck extension y moment by 66% is observed between the titled module test and the
baseline test. Thus, the titled module drastically reduces the chance of neck injury to a shorter vehicle occupant .
FIG. 9 demonstrates the substantial difference between the baseline and tilted module y component neck bending moments. The baseline curve (represented by dashed lines) shows a large positive bending moment at around 74 ms followed by a sharp drop in bending moments in the y direction. This portion of the curve represents the period of time in which the greatest injury is likely when the crash dummy is propelled into the airbag then rebounds backward. The tilt module (represented by a solid line) demonstrates a rather smooth transition between forward motion and airbag induced recoil because the negative drop in bending moment is not as severe around 74ms.
Table 4 shows a comparative summary of the sled test results for the baseline and tilted modules.
Table 4
The HIC value improved by thirty % for the 5th percentile female by adjusting the orientation of the airbag module. There was a 66% improvement for the neck extension y-moment, determined by examining the
individual frames from the crash video. For the baseline module configuration, the head moved backwards about 90° which would certainly cause damage to the cervical vertebrae, while a change of only around 10° was observed for the titled module configuration. The neck extension angle improved in the tilted module configuration by 89%. In most vehicle occupants, movement of the head past 45° would result in damage to the cervical vertebrae. For the neck extension angle measurements, zero degrees was assigned to a leveled head and the measured angle was computed by approximating the maximum angle formed from the head rotating backwards.