JP2001294115A - Vehicular shock severeness discrimination device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a shock discrimination device for discriminating the front shock severeness of a vehicle. SOLUTION: This device is composed of an acceleration detector and a shock severeness discrimination circuit electrically connected to it. The discrimination circuit uses an acceleration signal of the acceleration detector to generate a shock severeness discrimination signal on the output port of the shock severeness discrimination circuit. The shock severeness discrimination circuit determines whether shock is severe or not based on displacement formed by the acceleration signal in a predetermined period when the acceleration value of the acceleration signal exceeds a predetermined start value during the shock. When the shock is not severe nor determined, the shock severeness discrimination circuit continues to discriminate the shock severeness based on the displacement formed by the acceleration signal, the cumulative number of times of the jerk of the acceleration signal exceeding a threshold value, and a period of the jerk first exceeding the threshold value, until the determination signal is generated and outputted.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a vehicle impact severity identification device, and more particularly, to a vehicle impact severity identification device for identifying a front impact severity of a vehicle.

[0002]

BACKGROUND OF THE INVENTION Airbags have become a standard accessory in quite a number of vehicles. Whether or not the airbag is triggered now depends on the severity of the vehicle's forward impact. There are several prior art algorithms for determining when an airbag should be triggered, for example, analysis of energy and energy variations and acceleration variations.
There are also ways to trigger the airbag when the integral of the acceleration detected by the acceleration detector exceeds a certain threshold. However, each of these methods has both advantages and disadvantages, none of which can adequately identify when the airbag should be triggered. Therefore, more accurately identify the severity of the impact of the vehicle,
Providing the identified information to the airbag trigger electronic control unit is a very important task.

[0003]

SUMMARY OF THE INVENTION It is, therefore, an object of the present invention to provide a vehicle impact identification device for identifying the severity of a frontal impact of a vehicle.

[0004]

SUMMARY OF THE INVENTION An object of the present invention is to provide an acceleration detector provided on a vehicle for detecting acceleration of the vehicle and generating an acceleration signal; and an acceleration provided on the vehicle for identifying the severity of impact of the vehicle. A shock severity identification circuit electrically connected to the detector, the identification circuit uses an acceleration signal generated by the acceleration detector and generates an impact severity identification signal at an output port of the impact severity identification circuit. Then, during the impact, when the acceleration value of the acceleration signal exceeds a predetermined starting value, the impact severity identification circuit determines whether the impact is severe based on the displacement formed from the acceleration signal during a predetermined period. If the impact is not a severe impact or cannot be determined, the impact severity identification circuit generates an identification signal and outputs the displacement formed by the acceleration signal and the acceleration signal until output. A vehicle impact severity identification device that identifies the front impact severity of the vehicle, which continues to identify the impact severity based on the cumulative number of times the jerk has crossed the threshold and the duration the jerk has initially exceeded the threshold. Achieved.

[0005]

FIG. 1 is a functional block diagram of a vehicle impact severity identification apparatus 10 according to the present invention. The present invention relates to a vehicle impact severity identification device 10 for identifying a forward impact severity of a vehicle 12. The vehicle impact severity identification device 10 detects the deceleration of the vehicle 12 and electrically connects the acceleration detector 14 that generates an acceleration signal G ′ and the acceleration detector 14 that converts the acceleration signal G ′ into a digital acceleration signal G. The connected A / D (analog / digital) converter 15 and the digitized acceleration signal G
It comprises an impact severity identification circuit 16 electrically connected to an A / D converter 15 for identifying the impact severity of the vehicle 12 and generating an identification signal S at an output port 17 thereof. In addition, the vehicle 12 includes an airbag device 18, an airbag device 18, and an electronic control unit (ECU) 20 that is electrically connected to the vehicle impact severity identification device 10.
The electronic control unit 20 is a vehicle impact severity identification device 1
0 is used to control the triggering of the airbag device 18 by means of the identification signal S generated by 0 and the corresponding signal S 'relating to the dynamic information of the occupant.

Referring to FIG. 2, this is a control sequence of the shock severity identification circuit 16 shown in FIG. The algorithm used in the impact severity identification circuit 16 is a “two-stage (two stages) fuzzy control algorithm”. When the acceleration value of the digital acceleration signal G in the first step _(t 0 _{~t 1)} exceeds a predetermined starting value during an impact, impact severity identifying circuit 16 of the digital acceleration signal G in a predetermined time period t ₁ Displacement obtained from the double integral (di
Based on sp1), it is determined whether or not it is “severe impact”. Impact if not identified as a "harsh impact" the second step of calculating (t 1 _~) is started. The derivative of the digital acceleration signal G with respect to time is “jerking (jer
k) "." Cumulative number "(njerk)
Is determined as the number of times the jerk exceeds the threshold after the digital acceleration signal G exceeds the starting value. "Time interval"
(Tw) is determined as the interval between the time when the digital acceleration signal G exceeds the starting value and the time when the jerk first exceeds the threshold value. In the second stage (t _1- ), the shock severity identification circuit 16 generates the digital acceleration signal G after the acceleration value of the digital acceleration signal G exceeds a predetermined starting value until the identification signal S is generated and output. G, the total displacement (disp2) formed by G and the "cumulative number" of the number of times the jerk of the acceleration signal has exceeded the threshold value of the digital acceleration signal G;
After the digital acceleration signal G exceeds the starting value, the impact severity of the vehicle 12 is continuously identified based on the "time interval" at which the jerk first exceeded the threshold. The second stage impact severity has three categories of impact: "severe impact", "moderate impact", and "light impact".

Referring to FIG. 3, this is a functional block diagram of the shock severity identification circuit 16 shown in FIG. The shock severity identification circuit 16 electrically connects the memory 22 that stores a plurality of fuzzy rules and the A / D converter 15 that converts the digital acceleration signal G into an identification signal S that indicates the severity of the impact according to the plurality of fuzzy rules. And a fuzzy control unit 24 to be connected. The digital acceleration signal G generated by the A / D converter 15 is fetched, and the first displacement fuzzy input variable “disp1”, the second displacement fuzzy input variable “disp2”, and the time interval fuzzy input variable “t”
w ", the accumulated number of fuzzy input variables" njerk "
Occurs. The first displacement fuzzy input variable "disp
1 "represents a displacement which is formed from the acceleration signal during a predetermined period of time t ₁ after the acceleration value exceeds the starting value of the digital acceleration signals G, formed of time for double integration of the displacement digital acceleration signals G The second displacement fuzzy input variable "disp2" represents the total displacement because the acceleration signal has exceeded the starting value.
Is expressed as the interval between the time when the acceleration signal exceeds the starting value and the time when the jerk first exceeds the threshold. The cumulative number of fuzzy input variables "njerk" is expressed as the number of times that the jerk exceeds a threshold after the jerk of the acceleration signal exceeds the starting value.

The fuzzy rule has a first displacement fuzzy input variable “disp1” and a second displacement fuzzy input variable “dsp1”.
isp2 ", the if-t between the time interval fuzzy input variable" tw "and the cumulative number of fuzzy input variables" njerk "to the shock severity fuzzy output variable" svty ".
Used to determine the hen relationship. The fuzzy control unit 24 receives the first displacement fuzzy input variable “disp
1 ", the second displacement fuzzy input variable" disp2 ", the time interval fuzzy input variable" tw ", the membership grade of the added number of fuzzy input variables" njerk ", the impact severity fuzzy output variable by the interface of the fuzzy rule Convert to "svty" membership grade.

As shown in FIG. 3, the shock severity identification circuit 16 further includes a first displacement input membership function module 26, a second displacement input membership function module 28, and a time interval displacement input membership function module 30. , Cumulative number of input membership function modules 32, severity output membership function module 3
4 inclusive. The first displacement input membership function module 26 is stored in the memory 22 and converts the first displacement fuzzy input variable “disp1” into the first displacement fuzzy value “f” by the first displacement input membership function μ (disp1).
_disp1 ". The second displacement input membership function module 28 is
And the second displacement input membership function μ (d
isp1), the second displacement fuzzy input variable “dis
p2 "is used to convert to a first displacement fuzzy value" f _disp2 ". The time interval input membership function module 30 is stored in the memory 22 and time interval fuzzy by the time interval input membership function μ (tw). input variable "tw" the time interval fuzzy value _{"f t w"}
Used to convert to The accumulated number of input membership function modules 32 are stored in the memory 22 and the accumulated number of input membership functions μ (nje
rk), the number of fuzzy input variables “nje
rk ”is added to the number of fuzzy values“ f _njerk ”
Used to convert to The severity output membership function module 34 is stored in the memory 22 and formed from the severity output membership function μ (svty) after being defuzzified by using the center of area (COA) defuzzification method. It is used to convert the fuzzy value " _fstvty " to a severity output value, i.e., an impact severity fuzzy output variable "svty". The severity fuzzy value “f _svty ” is a first displacement fuzzy value “f _disp1 ” and a second displacement fuzzy value “f”.
_disp2 ", the time interval fuzzy _{value" f tw} "corresponding to a minimum grade. Thus the fuzzy control unit 24 severity fuzzy _value" "cumulative number of fuzzy _{value" f Njerk} the identification signal S a _{f s vty} " Convert, which represents the severity of the impact of the vehicle 12, ie, the value of the impact severity fuzzy output variable "svty", which is then output to the output port 17.

When a vehicle 12 collides, the acceleration detector 14 of the vehicle impact severity identification system 10 first produces an acceleration signal G 'due to the deceleration of the vehicle along the direction of motion. Next, the A / D converter 15 converts the acceleration signal G ′ into a digital acceleration signal G. Finally, the impact severity identification circuit 16 identifies the impact severity of the vehicle 12 with the digital acceleration signal G based on the "two-stage fuzzy control algorithm" design, and generates an identification signal S, which is an airbag device. 18 is one of the control signals of the electronic control unit 20 that triggers 18.

Referring to FIG. 4, this is a design flowchart of the shock severity identification circuit 16 shown in FIG.
The impact severity identification circuit 16 of the vehicle impact severity identification device 10 is designed based on a “two-stage fuzzy control algorithm”. The design process of the shock severity identification circuit 16 includes four steps, as shown in FIG. 4, to establish a test model, design a fuzzy rule, design a membership function, and perform simulation and test.

The purpose of establishing a test model is that the physical acceleration characteristics detected by the acceleration detector 14 during a collision are simulated by a computer program. The design of the shock severity identification circuit 16 of this embodiment is compatible with two shock pulse data sets AKF1, AKF2 from the branch of Autoliv, France. These are cited in Reference Example A (A Predictive Based A).
lgorithm for Actuation of
an Airbag, T .; Gioutsos, Out
omotive System Lab .. , Inc.
SAE920479) and two datasets at 30 mph and irregular roads and Reference B (The Useo).
f Signal Processing Techn
Iaues in an Occupant Dete
ction System, E.C. J. Gillis,
T. Gioutsos, Automotive Sy
stem Lab .. , Inc. SAE 940906) at 17 mph and 8 mph. These data include four physical properties: a first displacement fuzzy input variable "disp1", a second displacement fuzzy input variable "disp2", a time interval fuzzy input variable "tw", and an accumulated number of fuzzy input variables. “Nje
rk ". By analyzing these physical properties, their relationship to impact severity is found and provided as a basis for designing the fuzzy rules of the impact severity identification circuit 16. You.

Referring to FIG. 5, this is a table of the fuzzy rules of the shock severity identification circuit 16 shown in FIG.
The plurality of fuzzy rules of the vehicle impact severity identification device 10 are kernels of the impact severity identification circuit 16. Each fuzzy rule has the form "if ... then ...". "If ..." is the impact severity identification circuit 1.
6 indicates the input state, and the part "then..." Indicates the reaction state of the shock severity identification circuit 16. Each fuzzy rule derives a fuzzy value through a fuzzy inference process,
Fuzzy values are relaxed (defuzzification)
It is converted to a digital value by a process. In order to make the fuzzy rules more accurately handle the various types of impact, the impact severity identification circuit 16 applies a "two-stage fuzzy control algorithm" as described below.

The first stage fuzzy rule design uses a first displacement fuzzy input variable "disp1" to determine if the impact is a severe impact. If it is determined by the first displacement fuzzy input variable “disp1” that it is not severe, the impact severity identification circuit 16
Stay in the "waiting" state. When the acceleration value of the digital acceleration signal G exceeds a predetermined start value during a shock, the fuzzy control unit 24 of the shock severity identification circuit 16 starts calculating the displacement within a predetermined period, and the first displacement fuzzy input is performed. Generate the variable "disp1". If the first displacement fuzzy input variable “disp1” exceeds a predetermined value, the impact severity identification circuit 16 identifies the impact as a severe impact; if the first displacement fuzzy input variable “disp1” exceeds the predetermined value, If it is low, the impact severity identification circuit 16 decides not later but at a later time. That is, the impact severity identification circuit 16 determines whether the impact is severe in the second stage.

The design of the second stage fuzzy rule is to determine whether the impact is a "severe impact", a "medium impact", or a "light impact" by using a first displacement fuzzy input variable "disp1". ", A second displacement fuzzy input variable" disp2 ", a time interval fuzzy input variable" tw ",
The accumulated number of fuzzy input variables "njerk" is used. If the impact severity identification circuit 16 identifies that the impact is not a severe impact during a predetermined period, the fuzzy control unit 24 of the impact severity identification circuit 16 determines that the digital acceleration signal G has a predetermined start value, that is, the second acceleration value. After the displacement fuzzy input variable "disp2" has been exceeded, the total displacement obtained from the digital acceleration signal G is subsequently calculated, which is the predetermined starting value, i.e. the accumulated number of fuzzy input variables "njer".
k ", the number of times the jerk of the digital acceleration signal G has exceeded the threshold is subsequently calculated, and the digital acceleration signal G is set to a predetermined starting value, ie, the time interval fuzzy input variable" t ".
The time interval over which the jerk exceeds the threshold for the first time beyond w "is subsequently calculated. These four factors are used as a basis for identifying the severity of the impact in a second step.

Based on the fuzzy rule design concept of the first and second stages, ten fuzzy rules are constituted by the impact severity identification circuit 16 of the vehicle impact severity identification device 10 as shown in FIG. Is done.

Referring to FIG. 6, this is a corresponding table of linguistic items and fuzzy variables of the shock severity identification circuit 16 shown in FIG. The fuzzy variable of the shock severity identification circuit 16 is a first displacement fuzzy input variable “disp”.
1 ", a second displacement fuzzy input variable" disp2 ", an accumulated number of fuzzy input variables" njerk ", a time interval fuzzy input variable" tw ", and an impact severity fuzzy output variable" svty ". A displacement fuzzy input variable “disp2”, an accumulated number of fuzzy input variables “njerk”, a time interval fuzzy input variable “tw”,
The shock severity fuzzy output variable "svty" is designed to consist of three linguistic terms: low, medium and high as shown in FIG. Each membership function of a linguistic item is a triangular function.

Referring to FIGS. 7 to 11, FIG. 7 is a graph of the first displacement input membership function μ “disp1” of the shock severity identification circuit 16 shown in FIG. FIG.
4 is a graph of a second displacement input membership function μ “disp2” of the shock severity identification circuit 16 shown in FIG. FIG. 9 is a graph of the time interval input membership function μ “tw” of the shock severity identification circuit 16 shown in FIG. FIG. 10 shows the cumulative number of input membership functions μ (njer) of the shock severity identification circuit 16 shown in FIG.
It is a graph of k). FIG. 11 shows the severity output membership function μ (s) of the impact severity identification circuit 16 shown in FIG.
vty). The input membership function is the fuzzy inference "i
f. . . The embodiment adapts the physical properties of disp1, disp2, tw, njerk as fuzzy input variables. According to the analysis and test results of the above test model, these fuzzy input variables are used. The input membership function by
It is designed as shown in The first displacement input membership function μ (disp1) is shown in FIG. The second displacement input membership function μ (disp2) is shown in FIG. Time interval input membership function μ (tw)
Is shown in FIG. The cumulative number of input membership functions μ (njerk) is shown in FIG. Furthermore, the output membership function is a fuzzy inference "the
n. . . Since the impact severity identification circuit 16 is used to identify the severity of the impact, the severity output membership function μ (svty) representing the output parameter is as shown in FIG. Is shown.

The goal of simulation and testing is to determine if the shock severity is timely identified by the shock severity identification circuit 16 under various shock types. Thereafter, the use of the indicated test models, fuzzy rules, membership functions, and "two-step fuzzy control algorithm" will increase the utility and accuracy of the impact severity identification circuit 16 of the vehicle impact severity identification device 10 according to the present invention. It was tested using the above impact data to verify.

Referring to FIGS. 12 to 17, FIG. 12 is a table showing test results of the shock severity identification circuit 16 based on the shock pulse data set AKF1. FIG. 13 shows an impact severity identification circuit 16 based on the impact pulse data set AKF2.
5 is a table of test results. 14 to 17 are 30 mp
7 is a table of test results of the impact severity identification circuit 16 based on the impact data set at h, irregular road, 17 mph, and 8 mph. The analysis of the test results based on two shock pulse data sets AKF1, AKF2 from a branch of Autoliv, France and the shock data sets at 30 mph, irregular roads, 17 mph, 8 mph from References A and B are described below.

(1) A test based on 52 sets of data from AKF1 shown in FIG. 12 showed that the test results were desirable.

(2) A test based on 15 sets of data from AKF2 shown in FIG. 13 showed a successful test result.

(3) 30 mph from Reference Examples A and B,
Tests based on impact data sets at irregular terrain, 17 mph, and 8 mph resulted in successful tests as shown in FIGS. During the impact test at 30 mph as shown in FIG.
15ms 3.8ms faster than the result of Reference A in ms
Identified by During impact tests on rough terrain, light impacts were successfully identified as shown in FIG. 17 mph
During the impact test at, a moderate impact was identified at 28 ms as shown in FIG. During the impact test at 8 mph, light impacts were successfully identified as shown in FIG. These results indicate that moderate shock is distinguishable from the two similar shock waves shown in Reference B.

From the above simulation and test results, the shock severity identification circuit 16 was able to accurately identify various types of shock waves using the "two-stage fuzzy control algorithm". In Reference A, the concept of jerk is 30 mph
Introduced to distinguish between the severity of impacts on the road and on uneven roads, the rate of change in speed did not. However, Reference A does not provide an algorithmic method for identifying the severity of the impact. Reference B used a signal processing noise reduction method and an occupant position sensor for the detection of occupant displacement in order to distinguish between severe shock waves and similar ones. Reference B does not, however, disclose the trigger time. The vehicle shock severity identification apparatus 10 according to the present invention applies a "two-stage fuzzy control algorithm" to process the "predict" and "distinguish" functions, thereby distinguishing between different types of shock waves. Trigger the airbag in a timely manner. In the impact test at 30 mph from Reference A, the present invention identified a severe impact 3.8 ms earlier than Reference A. On uneven terrain, the present invention was able to distinguish between other types of impact and impact from irregular terrain. Vehicle impact severity identification device 10
Can also distinguish 17 mph shock waves from 8 mph shock waves, which are similar to each other and trigger the airbag within 30 ms if the impact is determined to be "moderate". These test results indicate the passage of time, and a “two-stage fuzzy control algorithm”
Distinguish the ability. It is used for most types of forward impact, and its impact discriminating ability is desirable.

Compared with the prior art method, the vehicle impact severity identification device 10 according to the present invention has an impact severity identification circuit 1.
The design of No. 6 employs a “two-stage fuzzy control algorithm”. In this first stage, the impact severity identification circuit 16 identifies whether the impact is a severe impact during a predetermined period based on the displacement formed from the digital acceleration signal G. If the impact is not identified as a “severe impact”, the impact severity identification circuit 16 then proceeds to determine whether the impact is a fuzzy input variable “disp1”, “disp2”,
“Tw” and “njerk” identify whether the condition is severe, medium, or mild. Furthermore, the simulation and test results of two shock pulse data sets AKF1, AKF2 from a branch of Autoliv, France, and four typical shock data sets from cited examples A, B show that the vehicle shock severity according to the invention is The degree discriminating apparatus 10 operates in a timely manner and has a good front impact severity discriminating ability.

[Brief description of the drawings]

FIG. 1 is a functional block diagram of a vehicle impact severity identification device according to the present invention.

FIG. 2 is a diagram showing a control sequence of the shock severity identification circuit shown in FIG.

FIG. 3 is a functional block diagram of the shock severity identification circuit shown in FIG. 1;

FIG. 4 is a design flowchart of the shock severity identification circuit shown in FIG. 3;

FIG. 5 is a table of a fuzzy rule by the impact severity identification circuit shown in FIG. 3;

FIG. 6 is a table of linguistic items and fuzzy variables by the impact severity identification circuit shown in FIG. 3;

FIG. 7 is a graph of a first displacement input membership function μ (disp1) of the shock severity identification circuit shown in FIG. 3;

FIG. 8 is a graph of a second displacement input membership function μ (disp2) of the shock severity identification circuit shown in FIG. 3;

FIG. 9 is a graph of a time interval input membership function μ (tw) of the shock severity identification circuit shown in FIG. 3;

FIG. 10 is a graph of a cumulative number input membership function μ (njerk) of the shock severity identification circuit shown in FIG. 3;

FIG. 11 is a graph of a severity output membership function μ (svty) of the impact severity identification circuit shown in FIG. 3;

FIG. 12 is a table of a test rule based on the shock severity identification circuit of the present invention based on the shock pulse data set AKF1.

FIG. 13 is a table of a test rule based on the shock severity identification circuit of the present invention based on the shock pulse data set AKF2.

FIG. 14: 30 mph, rough road, 17 mph, 8 m
9 is a test result of an impact severity identification circuit based on an impact data set at ph.

FIG. 15: 30 mph, rough road, 17 mph, 8 m
9 is a test result of an impact severity identification circuit based on an impact data set at ph.

FIG. 16: 30 mph, rough road, 17 mph, 8 m
9 is a test result of an impact severity identification circuit based on an impact data set at ph.

FIG. 17: 30 mph, rough road, 17 mph, 8 m
9 is a test result of an impact severity identification circuit based on an impact data set at ph.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 10 Impact severity identification device 12 Vehicle 14 Acceleration detector 15 A / D converter 16 Impact severity identification circuit 17 Output port 18 Airbag device 20 Electronic control unit 22 Memory 24 Fuzzy control unit 26 First displacement input membership function Module 28 Second displacement input membership function module 30 Time interval input membership function module 32 Cumulative number of input membership function modules 34 Severity output membership function module

 ──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Luo Kuan-Tsoo Taiwan, Taipei City, She-Rin, Chun-Sheraud, Section 1, 82-2 No. (72) Inventor Lin Y-Ming Taiwan, Tao -Yuan Xian, Chun-Li City, Jung-An 2 Street, 53rd F term (reference) 3D054 DD28 EE14 EE25 EE44 FF16

Claims (7)

[Claims] An acceleration detector provided on a vehicle for detecting acceleration of the vehicle and generating an acceleration signal; and electrically connected to an acceleration detector provided on the vehicle and for identifying a shock severity of the vehicle. A shock severity identification circuit, which uses an acceleration signal generated by an acceleration detector to generate an impact severity identification signal at an output port of the impact severity identification circuit, When the acceleration value of the signal exceeds a predetermined starting value, the impact severity identification circuit identifies whether the impact is severe based on the displacement formed from the acceleration signal during a predetermined period, and determines whether the impact is severe. If it is not a shock or cannot be determined, the shock severity identification circuit generates an identification signal and outputs the identification signal until the number of times the displacement formed by the acceleration signal and the number of times the jerk of the acceleration signal exceeds the threshold are reached. The number that has been pressurized, jerk initially based on the period exceeds the threshold value continues to identify the impact severity, vehicle impact severity identifying device for identifying the front impact severity of vehicles. 2. The accumulative number is the number of times the jerk, which is a derivative of the acceleration signal with respect to time, exceeds a threshold, during which time the acceleration signal first exceeds a starting value and the jerk first determines a threshold. The impact severity identification device according to claim 1, wherein the period is a period between the time and the time when the time exceeds the time. 3. When the impact severity identification circuit determines that the impact is not a severe impact at the end of the predetermined period, the impact severity identification circuit generates an identification signal until the identification signal is generated and output.
The total displacement formed from the acceleration signal since the acceleration signal exceeded the predetermined starting value and the number of times the jerk of the acceleration signal exceeded the threshold value from when the acceleration signal exceeded the predetermined starting value was added. The impact severity identification device according to claim 1, wherein the impact severity is continuously identified based on the number. 4. An impact severity identification circuit comprising: a time period from when the acceleration signal exceeds a start value to a time when the jerk of the acceleration signal exceeds a threshold until an identification signal is generated and output; 2. The impact severity identification device according to claim 1, wherein the impact severity is identified by a displacement formed from the acceleration signal. 5. When the impact severity identification circuit identifies that the impact is a severe impact due to the displacement of the acceleration signal during a predetermined period, the impact severity identification circuit identifies the impact at the output port. The impact severity identification device according to claim 1, which outputs a signal. 6. A memory for storing a plurality of fuzzy rules; and a fuzzy control unit electrically connected to the acceleration detector through an analog / digital converter for converting an acceleration signal into an identification signal according to the plurality of fuzzy rules. The impact severity identification device according to claim 1, comprising: 7. The acceleration signal generated by the acceleration detector causes the fuzzy control unit to generate a first displacement fuzzy input variable representing a displacement formed from the acceleration signal during a predetermined period, and to cause the acceleration signal to exceed a starting value. A second displacement fuzzy input variable representing the total displacement since the start, a time interval fuzzy input variable representing a period from the time when the acceleration signal exceeds the starting value to the time when the jerk of the acceleration signal exceeds the threshold value,
After the acceleration signal exceeds the starting value, an accumulative number of fuzzy input variables representing the number of times the jerk of the acceleration signal exceeds the threshold is generated, and the fuzzy control unit generates a first displacement fuzzy input variable and a second And the displacement fuzzy input variable of
A time interval fuzzy input variable and an accumulated number of fuzzy input variables are converted into an impact severity fuzzy output variable by a plurality of fuzzy rules, and the fuzzy control unit converts the impact severity fuzzy output variable into an identification signal. 6
The impact severity identification device as described.