Vehicle safety systems implementing algorithms for early detection of high-speed full-lap crash events

Abstract

A method for controlling actuatable safety devices for protecting vehicle occupants includes sensing left up-front vehicle acceleration via a left up-front sensor, sensing center up-front vehicle acceleration via a center up-front sensor, and sensing right up-front vehicle acceleration via a right up-front sensor. The method also includes sensing center vehicle acceleration via a center airbag control unit. The method further includes determining the occurrence of a vehicle crash event in response to determining that each of the left, center, and right up-front vehicle accelerations exceeds a predetermined magnitude and is phase-shifted from the center vehicle acceleration.

Description

[Background technology]

[0001] Actuatable vehicle occupant protection systems, or "vehicle safety systems," include actuatable restraints, such as actuatable seat belt retractors and airbags. These vehicle safety systems include a number of event sensors, such as accelerometers, and an electronic control unit, referred to herein as an airbag control unit ("ACU"), which monitors signals provided by the sensors and determines whether to activate the actuatable restraints.

[0002] It is desirable for a vehicle safety system to distinguish between various crash or impact events ("crash events") in which a vehicle may be involved. Recognizing that real-world crash scenarios are infinite, crash tests have been developed to mimic the most common real-world crash scenarios in terms of crash type and crash severity. If a vehicle safety system can distinguish between various crash tests and activate actuatable restraints accordingly, the vehicle safety system will function in a real-world crash event.

[0003] Crash tests can include various impact types, such as frontal impacts, side impacts, offset impacts, and oblique or angled impacts, each performed at a predetermined speed. Vehicle crash events can include various impacting structures, such as a pole, a rigid barrier, or a deformable barrier. The deformable barrier can be stationary or moving. Each of these impacting structures is specifically designed to represent structures encountered in real-world crash scenarios. For example, a pole crash test can implement a pole designed to represent a typical telephone or traffic light pole. The deformable barrier can be designed to represent another vehicle involved in the crash.

[0004] The vehicle safety system can be configured or adapted to distinguish between crash events in which activation of an actuatable restraint is desired and crash events in which activation of an actuatable restraint is not desired. Crash discrimination therefore requires determining the type of crash, e.g., frontal, side, offset, oblique, angle, etc. Crash discrimination can also involve determining the type of impacting structure, e.g., a pole or deformable barrier. Crash discrimination can also involve determining the severity of the crash. Crash discrimination can further involve detecting misuse conditions in which a vehicle impact can be detected but which do not warrant activation of an actuatable restraint. Examples of misuse conditions that may occur include off-road driving, rough roads, hitting a curb, animal impact, hammer blows (simulating hitting a rock or object), potholes, and railroad crossings. Crash tests can be performed to simulate misuse conditions and verify that the vehicle safety system responds in a desired manner.

[0005] Thus, the vehicle safety system must determine whether the event is a must-fire (MF), no-fire (NF), or misuse event. Must-fire events are crash events for which airbag deployment is required. It is for these events that further discrimination is desirable because it allows for tailoring of airbag deployment to specific classified events. Non-fire events are those whose magnitude does not warrant airbag deployment. Misuse events are those where the vehicle is being used in an unconventional manner, such as off-road driving, and airbag deployment is not desired, even if the magnitude of the detected event may dictate otherwise.

[0006] The National Highway Traffic Safety Administration (NHTSA) is the U.S. government agency that oversees vehicle safety and evaluates the safety of new vehicles through the NHTSA's New Car Assessment Program (US-NCAP). Through US-NCAP, NHTSA establishes crash tests to establish the crashworthiness of new vehicles and rates them with a star rating, with a five-star rating being the highest. The standards for these tests are published as Federal Motor Vehicle Safety Standards (FMVSS). NHTSA publishes FMVSS to implement safety laws passed by Congress. FMVSS standards detail the exact test procedures used to determine the US-NCAP rating, which is determined by metrics measured on crash test dummies placed inside the vehicle during testing.

[0007] The United States is not the only country with its own new car rating program. Other countries, such as China, Japan, and Australia, as well as other groups of countries in Europe and Latin America, have their own NCAPs. The new car ratings issued by these agencies are similar, although some use crash tests in slightly different ways.

[0008] For years, safety standards have been revised and updated to "push the envelope" when it comes to automotive safety. As a result, maintaining the standards forces automobile manufacturers to constantly improve the safety of their products. As standards become more stringent, safety systems have adapted and become more complex and sophisticated. Throughout the evolution of vehicle safety systems, crash classification has been found to be one of the key aspects that helps determine the effectiveness of the system. If a safety system can accurately and robustly identify crash scenarios as defined by the safety standards, it can take tailored countermeasures to provide the best outcome for occupants involved in the accident for which the standards are designed.

[0009] Although vehicle safety systems have been developed with the ability to distinguish between various crash events, there is always a need to further classify and distinguish between crash events so that the vehicle safety system can take appropriate responsive action. Crash events that may require discrimination include frontal impact crash events for different types of vehicles.

[0010] One specific frontal crash test is a high-speed full-lap crash test, in which a vehicle strikes a rigid barrier at 56 kph (approximately 35 mph). Because this test involves a full-lap impact, the vehicle strikes the rigid barrier head-on, with the entire width of the vehicle striking the barrier. This contrasts with other crash tests, such as offset crash tests, in which less than the full width of the vehicle's front end strikes the barrier, such as the left / right side or corner of the vehicle's front end.

[0011] High-speed full-lap crash tests can simulate, for example, a head-on collision with another vehicle or a rear-end collision with another vehicle. High-speed full-lap crash events develop quickly. Therefore, it is desirable for the vehicle safety system to identify this crash scenario as early in the event as possible to maximize the time allowed for the vehicle safety system to deploy necessary safety devices. Summary of the Invention

[0012] The present invention relates to a vehicle safety system that includes right, center, and left upfront satellite sensors that provide acceleration signals to an airbag control unit ("ACU"). The ACU executes a crash classification algorithm that utilizes the acceleration signals from the upfront sensors to determine the occurrence of a high-speed, full-overlap frontal crash. The algorithm implements a crash classification metric that utilizes a phase shift between the acceleration signals of the upfront sensors compared to the acceleration signals of the ACU to determine a high-speed, full-overlap frontal crash. [Means for solving the problem]

[0013] According to one aspect, a vehicle safety system for helping to protect a vehicle occupant includes (practice inserting restated claims here once filing is complete). [Brief explanation of the drawings]

[0014] [Figure 1] 1 is a schematic diagram of a vehicle including a vehicle safety system, according to an example configuration. [Figure 2] FIG. 2 is a schematic diagram illustrating crash signal adjustments performed in a vehicle safety system. [Figure 3] FIG. 2 is a diagram showing a collision discrimination algorithm executed in a vehicle safety system. [Figure 4] 4A and 4B are graphs depicting crash test data related to the development of crash classification algorithms implemented in vehicle safety systems. [Figure 5] FIG. 2 is a schematic diagram illustrating a crash discrimination metric implemented in a crash discrimination algorithm executed in a vehicle safety system. DETAILED DESCRIPTION OF THE INVENTION

[0015] References herein may be made to the left and right sides of a vehicle. These references should be understood to refer to the forward direction of vehicle travel. Thus, a reference to the "left" side of a vehicle is meant to correspond to the driver's side ("DS") of the vehicle. A reference to the "right" side of a vehicle is meant to correspond to the passenger's side ("PS") of the vehicle.

[0016] Also, some explanation will be given herein regarding vehicle axes, specifically the X, Y, and Z axes of the vehicle. VEH The Y axis is the axis extending in the longitudinal direction of the vehicle shown in FIG. 1. The Y axis is the axis extending in the lateral direction of the vehicle, perpendicular to the X axis, and is shown in FIG. VEH The Z axis is the vertical axis of the vehicle, perpendicular to both the X and Y axes, and is shown as Z in FIG.VEH The X, Y, and Z axes intersect at or near the vehicle's center of gravity ("COG").

[0017] Vehicle Safety Systems 1, a vehicle 10 includes a vehicle safety system 100. The safety system 100 may include a plurality of actuatable vehicle safety devices, generally designated 110. The actuatable safety devices 110 may include, for example, airbags (e.g., front airbags, side impact airbags, curtain airbags, etc.) and seat belts, such as seat belt pretensioners. The protective devices 110 are also operably connected to a central airbag control unit ("ACU") 120, which controls their actuation.

[0018] The ACU 120 is mounted centrally in the vehicle 10, behind the instrument panel 12. As shown, the ACU 120 is typically mounted in a central location between the front seats of the vehicle 10. The system 100 further includes a plurality of satellite sensors 130, 132, 134. The satellite sensors 130, 132, 134 are mounted remotely from the ACU 120 and operably connected to the ACU. The satellite sensors 130, 132, 134 are accelerometers used to sense vehicle acceleration and provide signals indicative of the sensed acceleration to the ACU 120.

[0019] The satellite sensors 130, 132, 134 provide digitized acceleration signals indicative of their respective sensed accelerations to the ACU 120. The ACU 120 includes a controller that is programmed to utilize the digitized acceleration signals to determine a vehicle condition, such as a crash condition, and to control the operation of the safety device 110 in response to the determined vehicle condition.

[0020] The ACU 120 includes an ACU sensor 122 integrated into the ACU 120. The ACU sensor 122 includes a dual-axis accelerometer. The dual-axis accelerometer measures the X axis (XVEH ) and Y-axis (Y VEH ) and determines values ​​indicative of these sensed vehicle accelerations. CCU_X is a value indicative of vehicle acceleration measured in the vehicle X-axis direction at the location of ACU 120. CCU_Y is a value indicative of vehicle acceleration measured in the vehicle Y-axis direction at the location of ACU 120. ACU 120 can use these signals to determine a crash and distinguish between different crash types. CCU_X is implemented in the algorithms described herein, but CCU_Y is not.

[0021] The satellite sensors are upfront sensors mounted at the front end of the vehicle 10, for example, at or near the front bumper 14. The satellite sensors include a right upfront sensor 130 mounted at or near the front right corner of the vehicle 10, a center upfront sensor 132 mounted at or near the front center of the vehicle, and a left upfront sensor 134 mounted at or near the front left corner of the vehicle. The upfront sensors 130, 132, 134 are accelerometers that measure vehicle acceleration in the X-axis direction. The upfront sensors 130, 132, 134 are operable to generate right, center, and left upfront X-axis acceleration signals RT_UFS, MD_UFS, and LT_UFS, respectively, that are provided to the ACU 120. The ACU 120 can use these signals to detect crashes and distinguish between different crash types according to the algorithms disclosed herein.

[0022] Signal Conditioning 2 shows a block diagram illustrating the signal conditioning that the sensor data undergoes before being executed in the discrimination algorithms disclosed herein. Each of the signals RT_UFS, MD_UFS, LT_UFS, and MD_UFS is conditioned through several different functions to generate an associated conditioned signal. It is these conditioned signals that are executed in the discrimination algorithms.

[0023] 2, the acceleration signal RT_UFS from the right UFS 130 undergoes low pass filtering (LPF block 150) that can be used to remove high frequency noise from the RT_UFS acceleration signal. An acceleration moving average calculation (AMA block 160) with an adjustable window size can be used to smooth the filtered acceleration signal. This adjustment produces the filtered moving average signal RT_UFS_AMA shown in block 170.

[0024] Similarly, the acceleration signal MD_UFS from the central UFS 132 undergoes low-pass filtering (LPF block 152) that can be used to remove high-frequency noise from the MD_UFS acceleration signal. An acceleration moving average calculation (AMA block 162) with an adjustable window size can be used to smooth the filtered acceleration signal. This adjustment produces the filtered moving average signal MD_UFS_AMA shown in block 172.

[0025] Similarly, the acceleration signal LT_UFS from the left UFS 134 undergoes low-pass filtering (LPF block 154) that can be used to remove high-frequency noise from the LT_UFS acceleration signal. An acceleration moving average calculation (AMA block 164) with an adjustable window size can be used to smooth the filtered acceleration signal. This adjustment produces the filtered moving average signal LT_UFS_AMA shown in block 174.

[0026] The acceleration signal CCU_X from the ACU sensor 122 also undergoes low-pass filtering (LPF block 156) which can be used to remove high frequency noise from the CCU_X acceleration signal. The low-pass filtered CCU_X acceleration signal is provided to a damped-spring-mass (MSD) model 166 which uses damped-spring-mass modeling to calculate the axial X acceleration resulting from the shock that produced the CCU_X acceleration. VEH , which generates a modeled value for the relative displacement along the vehicle axis X_REL_DISP. This can be done according to known modeling methods based on the particular vehicle architecture and based on an occupant with particular characteristics. Examples of this signal conditioning and modeling are described in detail in U.S. Pat. No. 5,935,182 to Foo et al. and U.S. Pat. No. 6,036,225 to Foo et al., the disclosures of which are incorporated herein by reference in their entireties. This conditioning generates a modeled relative displacement signal X_REL_DISP, shown in block 176.

[0027] An early detection algorithm for high-speed full-lap collisions. 3 illustrates a discrimination algorithm 200 executed by the vehicle safety system 100 to provide early detection of a high-speed full-overlap crash. The discrimination algorithm 200 utilizes the conditioned signals determined according to FIG. 2 to achieve early detection of a high-speed full-overlap crash event. As a result, the vehicle safety system 100 can respond by deploying one or more of the safety devices 110. The discrimination algorithm 200 implements three discrimination metrics to provide early crash detection.

[0028] Right UFS discrimination metric 180 evaluates the magnitude of RT_UFS_AMA signal 170 and X_REL_DISP signal 176. If the metric exceeds a predetermined right UFS threshold, a collision is determined and right UFS discrimination metric 180 outputs a logic 1. Otherwise, right UFS discrimination metric 180 outputs a logic 0.

[0029] Central UFS discrimination metric 182 evaluates the magnitude of MD_UFS_AMA signal 172 and X_REL_DISP signal 176. If the metric exceeds a predetermined central UFS threshold, a collision is determined and central UFS discrimination metric 182 outputs a logic 1. Otherwise, central UFS discrimination metric 182 outputs a logic 0.

[0030] Left UFS discrimination metric 184 evaluates the magnitude of LT_UFS_AMA signal 174 and X_REL_DISP signal 176. If the metric exceeds a predetermined left UFS threshold, a collision is determined and left UFS discrimination metric 184 outputs a logic 1. Otherwise, left UFS discrimination metric 184 outputs a logic 0.

[0031] The outputs of the discrimination metrics 180, 182, 184 are provided to a logical AND block 186. As shown, if all three discrimination metrics 180, 182, 184 match, i.e., if all three discrimination metrics indicate a collision has occurred (a logical 1), the AND block 186 outputs a logical 1, indicating that a high-speed full-lap collision has been detected. The discrimination algorithm 200 generates a high-speed full-lap collision detection signal, shown in block 190 of FIG. 3. In response to the high-speed full-lap collision detection signal 190, the ACU 120 activates one or more of the safety devices 110 (see FIG. 1).

[0032] To detect a high-speed full-wrap crash event, the discrimination algorithm 200 requires a match from all three UFS discrimination metrics 180, 182, 184. Because of this match, a high-speed full-wrap crash is detected with high confidence, and no further validation of the event is required. The discrimination algorithm 200 requires a match across the entire front end of the vehicle 10. Therefore, discrimination of a high-speed full-wrap crash event is performed to the exclusion of other front impact crash events, even if they may be similar.

[0033] For example, a high-speed offset crash event, such as a high-speed left offset crash event, may produce a phase shift that causes the left UFS discrimination metric 184 to discriminate the crash. Depending on the degree of offset and the severity of the offset crash, the center UFS discrimination metric 182 may also discriminate the crash. However, because the right UFS 130 is not impacted and therefore no phase shift is detected, the right UFS discrimination metric 180 will not discriminate the crash. This lack of agreement by only one UFS discrimination metric is sufficient to prevent the discrimination algorithm 200 from detecting the high-speed full-lap crash event.

[0034] This is not to say that the safety system 100 will not detect and respond to the example left offset crash event described in the previous paragraph. The safety system 100 can and certainly will include other discrimination algorithms tasked with detecting wrap crashes and will undoubtedly respond in an appropriate manner to offset crash events. The important point here is that the discrimination algorithm 200 provides a rapid response to high-speed full-wrap crash events and follows other discrimination algorithms to discriminate other crash types.

[0035] Characteristics of high-speed full-wrap frontal collisions 4A-B are graphs representing recorded crash test data. The graphs show how the various sensors described herein respond to different types of crash scenarios. FIG. 4A shows the sensor response to a high-speed full-lap frontal crash test. FIG. 4B shows the sensor response to an abuse condition crash test. For the example of FIG. 4B, the abuse condition is a front curb event in which the vehicle drives over a curb in a direction perpendicular to the curb.

[0036] As shown in FIG. 4A, because the crash event is a full-frontal, high-speed event, the upfront sensors record the crash acceleration before the ACU sensors record the crash acceleration. This is because the UFS sensors are mounted upfront on the vehicle, and the vehicle deforms (e.g., crushes or crushes) in response to the crash. This delays the acceleration of the ACU, which is mounted rearward, e.g., in the center, of the vehicle. Therefore, it can be seen that there is a delay between the UFS acceleration and the ACU acceleration. As a result, there is a phase shift between the acceleration signals output by the upfront and ACU sensors. In the crash test example of FIG. 4B, the phase shift is approximately 7 milliseconds for the center UFS and approximately 3 milliseconds for the left and right UFS.

[0037] In comparison, as shown in FIG. 4B, for an abuse event, there is no significant phase shift in the acceleration measured by the upfront sensor and the ACU sensor. This is because the magnitude of the abuse event is smaller than that of a high-speed, full-wrap frontal crash event. Because of this, the vehicle does not deform significantly in response to the event, and all of the acceleration sensors, both the upfront sensor and the ACU sensor, respond to vehicle acceleration more simultaneously, i.e., with little or no delay. Therefore, there is no significant phase shift in the acceleration signals generated by the upfront sensor and the ACU sensor.

[0038] These phenomena are supported by the example crash discrimination metric in Figure 5, which shows the upfront sensor response to the events of Figures 4A and 4B. For a high-speed, full-frontal overlap event, the metric initially exhibits a high slope, crosses the required activation threshold, and then levels off. This is the result of the phase shift between the upfront sensor and the ACU sensor, as shown in Figure 4A. During the initial phase of the crash, during the initial vehicle deformation and before the ACU sensor experiences the full vehicle acceleration, the magnitude of the UFS_AMA signal measured by the upfront sensor is high compared to the X_REL_DISP signal measured by the ACU sensor. Therefore, the magnitude of the metric is primarily due to the upfront sensor acceleration.

[0039] The required activation threshold is configured, or shaped, to facilitate detection of a phase shift between upfront acceleration and ACU acceleration. As shown in FIG. 5, the initial portion of the required activation threshold, i.e., the left side as shown, is lower in magnitude than the remainder of the threshold, i.e., the right side as shown. In the exemplary configuration of the metric shown in FIG. 5, this lower magnitude portion of the threshold is generally V-shaped. However, the shape can be different while maintaining the lower magnitude characteristic. The lower magnitude portion can be, for example, square or rectangular, curved (e.g., U-shaped), stepped, etc. Due to the lower magnitude of this phase shift detection portion of the threshold, phase shifts indicated by a steeper slope of the metric can be detected more quickly and easily, thereby improving the metric's response.

[0040] After the initial vehicle deformation, the ACU sensors accelerate and their outputs increase accordingly, while the outputs of the upfront sensors decrease as their acceleration approaches completion. As a result, we see the metric level off, as expected, because the magnitude of the metric is primarily due to the ACU acceleration.

[0041] From the above, it can be seen that when the upfront sensor acceleration and the ACU sensor acceleration are not phase-shifted and occur simultaneously or substantially simultaneously, the slope of the metric is less steep. Additionally, when the magnitude of acceleration is not high enough, the mandatory activation threshold is not crossed, regardless of any phase shift. Thus, it can be seen that the mandatory activation threshold implemented in the metric is configured, or shaped, such that an initial, steep acceleration, indicative of a high magnitude acceleration measured by the upfront sensor that is phase-shifted from the ACU sensor, is captured as a mandatory activation event.

[0042] Looking at the metric for the abuse event, we can see that because there is no delay / phase shift between the upfront sensor and the ACU sensor, the magnitude of the metric increases along a less steep slope, i.e., the rise and run of the metric are more equal. This is because both the upfront sensor and the ACU sensor are experiencing more simultaneous acceleration. Additionally, in the case of an abuse event, the magnitude of the acceleration is not sufficient to cross the required activation threshold. Therefore, the metric does not meet the required activation threshold.

[0043] Phase-shift sensing enables early collision detection Typically, crash discrimination involves two steps: crash detection and crash verification / classification. Crash detection can be triggered by various sensors or combinations of sensors. For example, measured ACU sensor acceleration exceeding a threshold magnitude can be used to detect that a crash has occurred. When this occurs, the safety system generates a request to activate one or more vehicle safety devices. Simultaneously, a crash classification algorithm utilizes the sensed crash data to classify the type and / or severity of the sensed crash. As a result of this verification, the safety system deploys protective devices.

[0044] Advantageously, the vehicle safety system 100 disclosed herein implements a sensor structure and algorithms that avoid the need to classify / verify a sensed crash condition. Eliminating this need results in the safety system 100's ability to deploy the protection device 110 early in a high-speed, full-overlap crash event. Because the safety system 100 implements three upfront sensors 130, 132, 134, the algorithm 200 can distinguish the occurrence of a particular event, namely, a high-speed, full-overlap crash event, in response to the immediate and substantially simultaneous response of the three discrimination metrics 180, 182, 184 associated with these sensors. Agreement between the three upfront sensors at this early stage of this particular type of crash event is sufficient to deploy the safety device 110 without further classification or verification.

[0045] The sensor architecture and implementation of the discrimination algorithm improves the time-to-fire ("TTF") performance of the vehicle safety system 100 in the event of a high-speed, full-wrap crash event. Testing has shown that a TTF in the range of 3-5 milliseconds can be reliably achieved in response to a high-speed, full-wrap crash event, while simultaneously discriminating between "no fire" conditions in response to a misuse event.

[0046] From the above description of the invention, those skilled in the art will recognize improvements, changes, and modifications. For example, the diagrams illustrating various metrics implemented in the disclosed vehicle safety system show thresholds having particular "shapes." The illustrated thresholds are merely examples. Those skilled in the art will understand that the characteristics of these metrics, i.e., the shapes of the lines identifying the thresholds, may vary depending on a variety of factors, such as the configuration of the particular vehicle in which the safety system is implemented (i.e., the particular platform or model of the vehicle) and the design criteria for the safety system implemented in the particular vehicle. Such improvements, changes, and / or modifications within the skill of those skilled in the art are intended to be covered by the appended claims.

Claims

1. 1. A method of controlling an actuatable safety device for protecting an occupant of a vehicle, comprising: sensing left up-front vehicle acceleration via a left up-front sensor; sensing a center up-front vehicle acceleration via a center up-front sensor; sensing right up-front vehicle acceleration via a right up-front sensor; sensing central vehicle acceleration via a central airbag control unit; determining the occurrence of a vehicle crash event in response to determining that each of the left up-front vehicle acceleration, the center up-front vehicle acceleration, and the right up-front vehicle acceleration exceeds a predetermined magnitude and is phase shifted from the center vehicle acceleration; The determination that each of the left up-front vehicle acceleration, the center up-front vehicle acceleration, and the right up-front vehicle acceleration exceeds a predetermined magnitude and is phase shifted from the center vehicle acceleration includes determining, for each up-front acceleration: performing an upfront discrimination metric that evaluates averaged upfront acceleration (UFS_AMA) and longitudinal vehicle displacement (X_REL_DISP); performing phase shift detection in the upfront discrimination metric using a threshold configured to detect a phase shift between the upfront acceleration and the central vehicle acceleration; determining a required activation state in response to the metric crossing the threshold; The method of claim 1, wherein the threshold of the upfront discrimination metric is reduced in a predetermined range of vehicle displacement in the vehicle longitudinal direction relative to other ranges of vehicle displacement in the vehicle longitudinal direction, and the range of vehicle displacement in the vehicle longitudinal direction for which the threshold is reduced improves detection of the phase shift between the upfront acceleration and the central vehicle acceleration.

2. 10. The method of claim 1, further comprising the step of actuating the actuatable safety device in response to determining the occurrence of a vehicle crash.

3. The method of claim 1 , wherein the reduced magnitude portion of the threshold is V-shaped.

4. The method of claim 1 , wherein the vehicle crash event comprises a high-speed full-lap crash event.

5. A vehicle safety system, comprising: an actuatable safety device; Left up-front sensor, a central upfront sensor; Right upfront sensor, and a central airbag control unit comprising a controller configured to perform the method of claim 1 to activate the actuatable safety device in response to detecting the occurrence of a vehicle crash.

6. 6. The vehicle safety system of claim 5, wherein the left up-front sensor is mounted at a front left position of the vehicle, the center up-front sensor is mounted at a front center position of the vehicle, and the right up-front sensor is mounted at a front right position of the vehicle.

7. 2. The vehicle safety system of claim 1, further comprising detecting the occurrence of a vehicle crash in response to detecting a simultaneous phase shift of the left up-front vehicle acceleration, the center up-front vehicle acceleration, and the right up-front vehicle acceleration from the center vehicle acceleration.

8. A vehicle safety system, comprising: an actuatable safety device; Left up-front acceleration sensor; a central up-front acceleration sensor; a right up-front acceleration sensor; an airbag control unit (ACU) operably connected to the safety device, the left up-front acceleration sensor, the center up-front acceleration sensor, and the right up-front acceleration sensor; the airbag control unit includes an ACU acceleration sensor and a controller configured to detect the occurrence of a vehicle crash in response to determining that the acceleration sensed by each of the left up-front acceleration sensor, the center up-front acceleration sensor, and the right up-front acceleration sensor exceeds a predetermined magnitude and is phase shifted from the acceleration sensed by the ACU acceleration sensor; A determination that each of the left up-front vehicle acceleration, the center up-front vehicle acceleration, and the right up-front vehicle acceleration exceeds a predetermined magnitude and is phase shifted from the center vehicle acceleration may be made by the airbag control unit determining, for each up-front acceleration: Implementing an upfront discrimination metric that evaluates averaged upfront acceleration (UFS_AMA) and longitudinal vehicle displacement (X_REL_DISP); performing phase shift detection in the upfront discrimination metric using a threshold configured to detect a phase shift between the upfront acceleration and the central vehicle acceleration; determining a required activation state in response to the metric crossing the threshold; A vehicle safety system, wherein the threshold of the upfront discrimination metric is reduced in a predetermined range of vehicle displacement in the vehicle longitudinal direction more than in other ranges of vehicle displacement in the vehicle longitudinal direction, and the range of vehicle displacement in the vehicle longitudinal direction for which the threshold is reduced improves detection of the phase shift between the upfront acceleration and the central vehicle acceleration.

Images