DEVICE AND METHOD FOR DETECTING ABNORMALITY OF MOTOR FOR GENERATING HYDRAULIC PRESSURE, AND NON-TRANSITORY COMPUTER-READABLE STORAGE MEDIUM STORING PROGRAM FOR PERFORMING THE METHOD

Abstract
A hydraulic pressure generation motor abnormality detection device may detect abnormality of a hydraulic pressure generation motor configured to drive a pump to generate a hydraulic pressure corresponding to an input displacement of a brake pedal of a vehicle, and includes a memory in which one or more instructions are stored and a processor configured to execute the one or more instructions, wherein the processor executes the one or more instructions to input input data to an artificial neural network model configured to receive the input data and output an estimation value of an output variable related to an output of the hydraulic pressure generation motor, obtain the estimation value of the output variable, and detect whether the hydraulic pressure generation motor is abnormal by using a difference between the estimation value of the output variable and an actual measurement value of the output variable.
Description
CROSS-REFERENCE TO RELATED APPLICATIONS

This application claims priority to and the benefit of Korean Patent Application No. 10-2023-0117732, filed on Sep. 5, 2023, and Korean Patent Application No. 10-2024-0060705, filed on May 8, 2024, the disclosures of which are incorporated herein by reference in their entireties.


TECHNICAL FIELD

The present disclosure relates to a hydraulic pressure generation motor abnormality detection device and method and a non-transitory computer-readable storage medium in which a program for performing the method is stored, and more specifically, to a hydraulic pressure generation motor abnormality detection device and method for detecting the abnormality of a hydraulic pressure generation motor for driving a pump to generate a hydraulic pressure for performing braking corresponding to an input displacement of a brake pedal of a vehicle, and a non-transitory computer-readable storage medium in which a program for performing the method is stored.


RELATED ART

Brake systems for braking are essentially installed in vehicles. Recently, integrated dynamic brake (IDB) systems have been proposed to improve braking stability and efficiency.


An integrated dynamic brake system includes a pedal displacement sensor for detecting an input depth (displacement) of a brake pedal, a motor which operates according to a control signal generated on the basis of detected information of the pedal displacement sensor, and a pump including a slave cylinder for converting a rotational force of the motor to a linear motion to generate a hydraulic pressure required for braking.


When the motor fails in the integrated dynamic brake system, a braking force corresponding to the input depth of the brake pedal may not be generated. As a result, a severe problem for safety can occur in a vehicle.


In general, an output torque of a hydraulic pressure generation motor can be reduced due to demagnetization due to degradation, insulation material corrosion, etc. Such output torque reduction of a motor gradually occurs. Therefore, it is preferable that abnormality of the motor, which generates a hydraulic pressure for braking, be prognosed to improve stability of a brake system. However, the conventional approach related to the abnormality detection of the motor generating the hydraulic pressure for braking is limited to measures for post-diagnosis and follow-up for failure.


SUMMARY

The present disclosure is directed to solving the above-described problems and providing a hydraulic pressure generation motor abnormality detection device and method for detecting performance degradation of a hydraulic pressure generation motor configured to drive a pump to generate a hydraulic pressure for performing braking corresponding to an input displacement of a brake pedal of a vehicle, before the hydraulic pressure generation motor fails and a non-transitory computer-readable storage medium in which a program for performing the method is stored.


The present disclosure is also directed to providing a hydraulic pressure generation motor abnormality detection device and method for efficiently prognosing output degradation of a hydraulic pressure generation motor configured to generate a hydraulic pressure for braking using controller area network (CAN) signals in a vehicle and a non-transitory computer-readable storage medium in which a program for performing the method is stored.


The objects of the present disclosure are not limited to the above-described objects, and other objects that are not mentioned will be able to be clearly understood by those skilled in the art to which the present disclosure pertains from the following description.


In accordance with one aspect of the present disclosure, there is provided a hydraulic pressure generation motor abnormality detection device for detecting abnormality of a hydraulic pressure generation motor configured to drive a pump to generate a hydraulic pressure for performing braking corresponding to an input displacement of a brake pedal of a vehicle, is provided, the hydraulic pressure generation motor abnormality detection device including a memory in which one or more instructions are stored and a processor configured to execute the one or more instructions, wherein the processor executes the one or more instructions to input input data to an artificial neural network model configured to receive the input data and output an estimation value of an output variable related to an output of the hydraulic pressure generation motor, obtain the estimation value of the output variable, and detect whether the hydraulic pressure generation motor is abnormal by using a difference between the estimation value of the output variable and an actual measurement value of the output variable.


In the hydraulic pressure generation motor abnormality detection device according to one aspect of the present disclosure, the input data may include one or more of a speed of the vehicle, a lateral acceleration of the vehicle, the input displacement of the brake pedal, and a wheel speed of the vehicle.


In the hydraulic pressure generation motor abnormality detection device according to one aspect of the present disclosure, the output variable may include a fluid pressure generated by the pump.


In the hydraulic pressure generation motor abnormality detection device according to one aspect of the present disclosure, the input data may be obtained through a CAN of the vehicle.


In the hydraulic pressure generation motor abnormality detection device according to one aspect of the present disclosure, the artificial neural network model may be formed in a generative adversarial network (GAN) including a generator configured to receive the input data and generate the estimation value of the output variable and a discriminator configured to receive the input data and the actual measurement value of the output variable and output an actual measurement related discrimination value.


In the hydraulic pressure generation motor abnormality detection device according to one aspect of the present disclosure, the discriminator may further receive the input data and the estimation value of the output variable and output an estimation related discrimination value.


In the hydraulic pressure generation motor abnormality detection device according to one aspect of the present disclosure, the artificial neural network model may be built by alternately performing learning of the generator and the discriminator, and the input data and the actual measurement value of the output variable used for the learning may be obtained in a normal state of each of the vehicle and the hydraulic pressure generation motor.


In the hydraulic pressure generation motor abnormality detection device according to one aspect of the present disclosure, the generator may be provided as a multivariate transformer.


In the hydraulic pressure generation motor abnormality detection device according to one aspect of the present disclosure, the processor may execute the one or more instructions to input error data related to the difference between the estimation value of the output variable and the actual measurement value of the output variable to an abnormality detection model and determine whether the hydraulic pressure generation motor is abnormal.


In the hydraulic pressure generation motor abnormality detection device according to one aspect of the present disclosure, the abnormality detection model may use a one-class support vector machine (OCSVM) algorithm.


In the hydraulic pressure generation motor abnormality detection device according to one aspect of the present disclosure, there are a plurality of data sets each including the input data, the actual measurement value of the output variable, and the estimation value of the output variable, and the error data may include an average and a standard deviation of errors between the actual measurement values of the output variables and the estimation values of the output variables obtained from the plurality of data sets, a maximum absolute error between the actual measurement values of the output variables and the estimation values of the output variables obtained from the plurality of data sets, and discrimination values of the discriminator for the input data and the actual measurement values of the output variables included in the plurality of data sets.


In accordance with another aspect of the present disclosure, there is provided a hydraulic pressure generation motor abnormality detection method for detecting abnormality of a hydraulic pressure generation motor configured to drive a pump to generate a hydraulic pressure for performing braking corresponding to an input displacement of a brake pedal of a vehicle is provided, the hydraulic pressure generation motor abnormality detection method including executing, by a processor, one or more instructions to input input date to an artificial neural network model configured to receive the input data and output an estimation value of an output variable related to an output of the hydraulic pressure generation motor and obtaining the estimation value of the output variable and detecting, by the processor, whether the hydraulic pressure generation motor is abnormal by using a difference between the estimation value of the output variable and an actual measurement value of the output variable.


In the hydraulic pressure generation motor abnormality detection method according to one aspect of the present disclosure, the input data may include one or more of a speed of the vehicle, a lateral acceleration of the vehicle, the input displacement of the brake pedal, and a wheel speed of the vehicle.


In the hydraulic pressure generation motor abnormality detection method according to one aspect of the present disclosure, the output variable may include a fluid pressure generated by the pump.


In the hydraulic pressure generation motor abnormality detection method according to one aspect of the present disclosure, the artificial neural network model may be formed in a GAN including a generator configured to receive the input data and generate the estimation value of the output variable and a discriminator configured to receive the input data and the actual measurement value of the output variable and output an actual measurement related discrimination value.


In the hydraulic pressure generation motor abnormality detection method according to one aspect of the present disclosure, the artificial neural network model may be built by alternately performing learning of the generator and the discriminator, and data used for the learning may be obtained in a normal state of each of the vehicle and the hydraulic pressure generation motor.


In the hydraulic pressure generation motor abnormality detection method according to one aspect of the present disclosure, the detecting of whether the hydraulic pressure generation motor is abnormal may include inputting, by the processor, the input data and the actual measurement value of the output variable to the discriminator and obtaining the actual measurement related discrimination value generated by the discriminator and inputting, by the processor, the actual measurement related discrimination value and error data including values related to the difference between the estimation value of the output variable and the actual measurement value of the output variable to an abnormality detection model and obtaining an output of the abnormality detection model.


In the hydraulic pressure generation motor abnormality detection method according to one aspect of the present disclosure, the abnormality detection model may use an OCSVM algorithm.


In the hydraulic pressure generation motor abnormality detection method according to one aspect of the present disclosure, there are a plurality of data sets each including the input data, the actual measurement value of the output variable, and the estimation value of the output variable, and the error data may include an average and a standard deviation of errors between the actual measurement values of the output variables and the estimation values of the output variables obtained from the plurality of data sets, a maximum absolute error between the actual measurement values of the output variables and the estimation values of the output variables obtained from the plurality of data sets, and discrimination values of the discriminator for the input data and the actual measurement values of the output variables included in the plurality of data sets.


In accordance with still another aspect of the present disclosure, there is provided a non-transitory computer-readable storage medium in which a program including at least one instruction for performing the hydraulic pressure generation motor abnormality detection method is stored.





BRIEF DESCRIPTION OF THE DRAWINGS

The above and other objects, features and advantages of the present disclosure will become more apparent to those of ordinary skill in the art by describing exemplary embodiments thereof in detail with reference to the accompanying drawings, in which:



FIG. 1 is a schematic view illustrating a configuration of an integrated dynamic brake system of a vehicle;



FIG. 2 is a view illustrating a configuration of a hydraulic pressure generation motor abnormality detection device according to one embodiment of the present disclosure;



FIG. 3 is a view illustrating models used for obtaining an estimation value or detecting abnormality in the hydraulic pressure generation motor abnormality detection device according to one embodiment of the present disclosure;



FIG. 4 is a view illustrating a detailed configuration of an artificial neural network model;



FIG. 5 is a view illustrating a detailed configuration of a generator of the artificial neural network model;



FIG. 6 is a view illustrating operation of the hydraulic pressure generation motor abnormality detection device according to one embodiment of the present disclosure;



FIG. 7 is a flowchart illustrating a hydraulic pressure generation motor abnormality detection method according to one embodiment of the present disclosure; and



FIG. 8 is a detailed flowchart illustrating detecting whether a hydraulic pressure generation motor is abnormal in the hydraulic pressure generation motor abnormality detection method according to one embodiment of the present disclosure.





DETAILED DESCRIPTION OF EMBODIMENTS

Hereinafter, embodiments of the present disclosure will be described in detail so that those skilled in the art to which the present disclosure pertains can easily carry out the embodiments. The present disclosure may be implemented in many different forms and is not limited to the embodiments described herein. In order to clearly describe the present disclosure, portions not related to the description are omitted from the accompanying drawings, and the same or similar components are denoted by the same reference numerals throughout the specification.


The words and terms used in the specification and the claims are not limitedly construed as their ordinary or dictionary meanings, and should be construed as meaning and concept consistent with the technical spirit of the present disclosure in accordance with the principle that the inventors can define terms and concepts in order to best describe their invention.


In the specification, it should be understood that the terms such as “comprise” or “have” are intended to specify the presence of features, numbers, steps, operations, components, parts, or combinations thereof described in the specification and do not preclude the possibility of the presence or addition of one or more other features, numbers, steps, operations, components, parts, or combinations thereof.



FIG. 1 is a schematic view illustrating a configuration of an integrated dynamic brake system of a vehicle.


An integrated dynamic brake system 200 is disposed in a vehicle and performs braking of the vehicle. The integrated dynamic brake system 200 is configured to drive a pump 270 according to an electric control signal generated according to a detected input depth (displacement) of a brake pedal 210 to supply a hydraulic pressure required for the braking to a brake 280 disposed on each wheel of the vehicle. The integrated dynamic brake system 200 provides quick response performance and facilitates implementation of a by-wire function of the vehicle.


Referring to FIG. 1, the integrated dynamic brake system 200 includes the brake pedal 210, a pedal simulator 220, a pedal displacement sensor 230, a vehicle sensor unit 240, an electronic control unit 250, a hydraulic pressure generation motor 260, the pump 270, the brake 280, and a pressure sensor 290.


When a pressing force of a driver is input to the brake pedal 210, a position of the brake pedal 210 is changed toward one side. The pedal simulator 220 provides a repulsive force to the brake pedal 210. Accordingly, the driver may feel a proper pedaling sensation.


The pedal displacement sensor 230 detects an input depth of the brake pedal 210, that is, an input displacement of the brake pedal 210. The input displacement of the brake pedal detected by the pedal displacement sensor 230 may be transmitted to the electronic control unit 250.


The vehicle sensor unit 240 may detect one or more pieces of information related to a state of the vehicle. For example, the vehicle sensor unit 240 may include a vehicle speed sensor 241, a wheel speed sensor 242, and an acceleration sensor 243. A speed of the vehicle detected by the vehicle speed sensor 241, a wheel speed detected by the wheel speed sensor 242, and a lateral acceleration of the vehicle measured by the acceleration sensor 243 may be transmitted to the electronic control unit 250.


The electronic control unit 250 generates a control signal for controlling the hydraulic pressure generation motor 260 on the basis of one or more pieces of input data. In this case, the one or more pieces of input data may include one or more of the input displacement of the brake pedal, the speed of the vehicle, and the wheel speed or the lateral acceleration of the vehicle.


The hydraulic pressure generation motor 260 receives the control signal from the electronic control unit 250 and is driven to correspond to the control signal. A rotational driving force of the hydraulic pressure generation motor 260 is transmitted to the pump 270.


The pump 270 is provided with a driving mechanism that converts a rotational force of the hydraulic pressure generation motor 260 to a linear motion. The pump 270 generates a hydraulic pressure such that a braking hydraulic pressure corresponding to operation of the hydraulic pressure generation motor 260 is transmitted to the brake 280 disposed on the wheel of the vehicle.


The brake 280 may perform braking using the hydraulic pressure supplied according to operation of the pump 270. In this regard, a hydraulic pressure circuit for supplying the hydraulic pressure may be disposed between the pump 270 and the brake 280.


The pressure sensor 290 detects a fluid pressure generated by the pump 270. The pressure sensor 290 may be disposed in the pump 270 or the hydraulic pressure circuit. The fluid pressure detected by the pressure sensor 290 may be fed back to the electronic control unit 250.



FIG. 2 is a view illustrating a configuration of a hydraulic pressure generation motor abnormality detection device according to one embodiment of the present disclosure. FIG. 3 is a view illustrating models used for obtaining an estimation value or detecting abnormality in the hydraulic pressure generation motor abnormality detection device according to one embodiment of the present disclosure.


A hydraulic pressure generation motor abnormality detection device 100 according to one embodiment of the present disclosure detects abnormality of the hydraulic pressure generation motor 260 which drives the pump 270 to generate a braking hydraulic pressure corresponding to an input displacement of the brake pedal 210 of the vehicle. More specifically, the hydraulic pressure generation motor abnormality detection device 100 may detect output degradation of the hydraulic pressure generation motor 260 before the hydraulic pressure generation motor 260 fails.


The hydraulic pressure generation motor abnormality detection device 100 according to one embodiment of the present disclosure may prognose function degradation of the hydraulic pressure generation motor 260 through a digital twin algorithm based on an artificial intelligence before the hydraulic pressure generation motor 260 fails. The digital twin algorithm may virtually build the integrated dynamic brake system of the vehicle.


Referring to FIG. 2, the hydraulic pressure generation motor abnormality detection device 100 according to one embodiment of the present disclosure may include a memory 110 and a processor 120.


The memory 110 stores one or more instructions. The one or more instructions may be executed by the processor 120.


The memory 110 may include a hardware device configured store and execute a program instruction. For example, the memory 110 may include storage media such as a read only memory (ROM), a random-access memory (RAM), and a flash memory. In addition, the memory 110 may also include magnetic media such as a floppy disk and a magnetic tape, optical media such as a compact disk read only memory (CD-ROM) and a digital video disk (DVD), magneto-optical media such as a floptical disk, etc.


The processor 120 executes the one or more instructions. For example, the processor 120 may be a hardware unit which performs operation and control in a computer. The processor 120 may include at least one arithmetic logic unit (ALU) and a register.


The processor 120 executes the one or more instructions to input input data to an artificial neural network model 20, which receives the input data and outputs an estimation value of an output variable related to an output of the hydraulic pressure generation motor 260, and obtain the estimation value of the output variable.


The input data may include one or more of a speed of the vehicle, a lateral acceleration of the vehicle, an input displacement of the brake pedal, and a wheel speed of the vehicle. For example, the input data may include the speed of the vehicle, the lateral acceleration of the vehicle, the input displacement of the brake pedal, and the wheel speed of the vehicle. In this case, the wheel speed of the vehicle may be obtained from each of a plurality of wheels of the vehicle.


The input data may be obtained through a controller area network (CAN) 300 of the vehicle. For example, the input data may be transmitted to the memory 110 through the CAN 300.


As described above, the input data may be obtained through the CAN 300 of the vehicle. Accordingly, in the case of the present disclosure, an additional sensor related to abnormality detection of the hydraulic pressure generation motor does not need to be used.


In this regard, the memory 110 may be directly or indirectly connected to the CAN 300 of the vehicle. The memory 110 may receive the input data through the CAN 300. Accordingly, the processor 120 may obtain the input data through the memory 110.


The output variable may include a fluid pressure generated by the pump 270. In one embodiment of the present disclosure, the fluid pressure may be the output variable.


The artificial neural network model 20 receives the input data and outputs the estimation value of the output variable related to the output of the hydraulic pressure generation motor 260. More specifically, the artificial neural network model 20 may receive the input data and output an estimation value of the fluid pressure generated by the pump.


The estimation value of the output variable may be obtained to be compared with an actual measurement value of the output variable. In this case, the actual measurement value of the output variable may be obtained through the pressure sensor disposed in the pump or the hydraulic pressure circuit. That is, the actual measurement value of the output variable may be the fluid pressure detected by the pressure sensor.


In one embodiment of the present disclosure, the artificial neural network model 20 may be formed in a generative adversarial network (GAN).


In one embodiment of the present disclosure, the artificial neural network model 20 may be a neural network model based on deep learning for the integrated dynamic brake system 200 of the vehicle. In other words, the artificial neural network model 20 may serve as a virtual twin model of the integrated dynamic brake system.



FIG. 4 is a view illustrating a detailed configuration of the artificial neural network model.


Referring to FIG. 4, the artificial neural network model 20 may include a generator 21 and a discriminator 22.


The artificial neural network model 20 may be built by alternately performing learning of the generator 21 and the discriminator 22. In this case, data used for the learning may be obtained in a normal state of each of the vehicle and the integrated dynamic brake system. Accordingly, an estimation value of the output variable may follow an actual measurement value of the output variable in the normal state of each of the vehicle and the integrated dynamic brake system.


The generator 21 receives the input data and generates an estimation value of the output variable. The generator 21 may include a neural network. The generator 21 may be provided as a multivariate transformer. That is, the artificial neural network model 20 may be the GAN based on the multivariate transformer.


In one embodiment of the present disclosure, the generator 21 may receive the input data. As described above, the input data may include one or more of a speed of the vehicle, a lateral acceleration of the vehicle, an input displacement of the brake pedal, and a wheel speed of the vehicle.


The generator 21 may operate the input data and output an estimation value of the output variable. In this case, a fluid pressure generated by the pump may be the output variable. In one embodiment of the present disclosure, an estimation value of the output variable estimated by the generator 21 follows an actual measurement value of the output variable when the integrated dynamic brake system of the vehicle is in a normal state.



FIG. 5 is a view illustrating a detailed configuration of the generator of the artificial neural network model.


Referring to FIG. 5, the generator 21 may include an encoder 21a and a decoder 21b.


The encoder 21a and the decoder 21b are formed in a network for obtaining an estimation value of the output variable. In other words, the generator 21 may be formed in the network with an encoder 21a-decoder 21b structure trained in an end-to-end method to obtain an estimation value of the output variable.


The encoder 21a may be designed to extract high-level features from the input data using a transformer block and a linear operation. In this case, the transformer block may include an embedding layer, a self-attention layer, and a feed-forward layer.


First, in the transformer block, the embedding layer may encode spatiotemporal information into input sequences using a trainable embedding matrix.


Next, deep features may be extracted through the self-attention layer, and refinement may be performed using the feed-forward layer.


In this case, the self-attention layer may use a scaled dot-product attention mechanism. The self-attention layer may be an important component in a deep neural network (DNN) architecture based on a transformer which allows a model to process and understand dependence in an input sequence.


The scaled dot-product attention may apply layer normalization to the input sequence first, and processes a result thereof as three unique vectors of a quarry Q, a key K, and a value V through the linear operation.


Next, an attention score AS for each position of the input sequence may be calculated on the basis of induced vectors. A calculation equation of the attention score AS may be given as Equation 1 below.









AS
=


σ

(


Q
·
K






x
input



dim



)

·
V





[

Equation


1

]







(σ(·) means a softmax operation, ∥Xinputdim means an input feature dimension, and (·) means dot-product.)


The attention score quantifies relative importance of each position in the input sequence related to a given quarry. That is, a higher score is assigned to a more appropriate position.


An output of the transformer block may be provided to two different linear operations in order to calculate a quarry, a key vector, and an initial input value yi of the decoder 21b.


The decoder 21b may use obtained values to estimate one or more estimation values in an automatic regression method similar to sequence-to-sequence recurrent neural networks.


Like the encoder 21a, the decoder 21b may include a transformer block and a linear operation. The decoder 21b may also further includes a cross-attention layer in the transformer block.


The cross-attention layer may perform a scaled dot-product attention using the quarry and the key vector of the encoder 21a in order to training a dynamic correlation between an input signal and an estimation signal.


The discriminator 22 receives the input data and an actual measurement value of the output variable and outputs a discrimination value. The discrimination value output by the discriminator 22 for the input data and the actual measurement value of the output variable may be defined as an actual measurement related discrimination value.


The discriminator 22 is a classifier that aims to determine the conditional probability. The discriminator 22 determines the probability of an example being real or fake given that set of input features. The discrimination value from the discriminator 22 are classification labels. For example, the discrimination value from the discriminator 22 is 0 or 1.


Meanwhile, the discriminator 22 may further receive the input data and an estimation value of the output variable and output a discrimination value. The discrimination value output by the discriminator 22 for the input data and the estimation value of the output variable may be defined as an estimation related discrimination value. This process may be performed while the generator 21 and the discriminator 22 performs leaning.


As described above, the discriminator 22 outputs the discrimination value for the input data and the actual measurement value of the output variable in an operation process for detecting abnormality. Meanwhile, in an operation process for training, the discriminator 22 may output a discrimination value for the input data and an estimation value of the output variable.


The discriminator 22 may be formed in a discrimination network used for adversary training the GAN. In other words, the generator 21 and the discriminator 22 may be trained in an adversary training method to improve estimation performance and set a GAN model.


More specifically, optimization of the discriminator 22 and the generator 21 may be alternately performed in order to solve a Wasserstein min-max problem as in Equation 2.











min
G


max
D



e

x

D
,
Real



[

D

(

x

D
,
Real


)

]


-


E

x

D
,
Fake



[

D

(

x

D
,
Fake


)

]





[

Equation


2

]







(XD,Real denotes an actual data set including input data and an actual measurement value, and XD,Fake means a virtual data sample including the input data and an estimation value.)


In the artificial neural network model 20, the generator 21 is trained to deceive the discriminator 22 which classifies spatiotemporal characteristics of XD,Real and XD,Fake. Accordingly, the generator 21 may generate an estimation value following the integrated dynamic brake system in the normal state.


In this regard, a loss function for training the GAN may be defined as in Equation 3 below.













L
G

=



1
N







"\[LeftBracketingBar]"



y
G

-

y
G




"\[RightBracketingBar]"




+


k
N





[

-

D

(

x

D
,
Fake


)


]











L
D

=



1
N





[


D

(

x

D
,
Fake


)

-

D

(

x

D
,
Real


)


]



+


1
N





[


(








x

D
,
Fake




D

(

x

D
,
Fake


)




2

-
1

)

2

]











[

Equation


3

]







(N is an arrangement size, LG is a loss function of the generator, and Lp is a loss function of the discriminator.)


A first loss item of the generator 21 corresponds to an average absolute error loss LMAE of guidance training. The remaining loss item of the generator 21 constitutes a GAN loss function LGAN with a loss of the discriminator 22.


In addition, the processor 120 detects whether the hydraulic pressure generation motor is abnormal by using a difference between an estimation value of the output variable and an actual measurement value of the output variable. As described above, the output variable may be a fluid pressure generated by the pump according to operation of the hydraulic pressure generation motor. In other words, the processor 120 may compare an estimation value with an actual measurement value of the fluid pressure to detect abnormality of the hydraulic pressure generation motor. In this case, the actual measurement value may be measured by the pressure sensor 290.


The processor 120 may execute the one or more instructions to input error data related to the difference between the estimation value of the output variable and the actual measurement value of the output variable to an abnormality detection model 30 to determine abnormality of the hydraulic pressure generation motor. For example, the abnormality detection model 30 may use a one-class support vector machine (OCSVM) algorithm.


A plurality of data sets each including the input data, an actual measurement value of the output variable, and an estimation value of the output variable may be present. The error data may include an average and a standard deviation of errors between the actual measurement values of the output variables and the estimation values of the output variables obtained from the plurality of data sets, a maximum absolute error between the actual measurement values of the output variables and the estimation values of the output variables obtained from the plurality of data sets, and discrimination values of the discriminator for the input data and the actual measurement values of the output variables included in the plurality of data sets.


A plurality of data sets each including the input data, an estimation value of the output variable, and an actual measurement value of the output variable may be processed as one batch. For example, 100 to 150 data sets (specifically, 128 data sets) may be processed as one batch, and the error data may be obtained for each batch.


The error data obtained for one batch may be input to the abnormality detection model 30. In addition, a level of abnormality of the hydraulic pressure generation motor (performance degradation of the hydraulic pressure generation motor) may be detected on the basis of an output (feature) obtained from the abnormality detection model 30 which receives the error data.


An F1 score given as in Equation 4 may be considered in relation to an anomaly detection metric.










F

1

=


2
×
TP



2
×
TP

+
FP
+
FN






[

Equation


4

]







(TP denotes a true positive, FP denotes a false positive, FN denotes a false negative, and positive denotes performance degradation of the hydraulic pressure generation motor.)


Meanwhile, as described above, the discriminator 22 may calculate not only a discrimination value for input data and an actual measurement value of the output variable included in the data set but also a discrimination value for the input data and an estimation value of the output variable included in the data set. The discrimination value of the discriminator 22 related to the estimation value of the data set may be fed back to the generator 21 of the artificial neural network model 20 and used for training.


The configuration of the hydraulic pressure generation motor abnormality detection device 100 according to one embodiment of the present disclosure has been described in detail. Hereinafter, an operation of the hydraulic pressure generation motor abnormality detection device 100 will be described in detail.



FIG. 6 is a view illustrating operation of the hydraulic pressure generation motor abnormality detection device according to one embodiment of the present disclosure.


Referring to FIG. 6, the hydraulic pressure generation motor abnormality detection device 100 according to one embodiment of the present disclosure may operate as described below.


First, the memory 110 stores input data ØEcu and an actual measurement value Pm of an output variable. In this case, a fluid pressure generated by the pump of the integrated dynamic brake system may be the output variable. That is, the actual measurement value of the output variable may include an actual measurement value of the fluid pressure.


As described above, the input data ØEcu may include one or more of a speed of the vehicle, a lateral acceleration of the vehicle, an input displacement of the brake pedal, and a wheel speed of the vehicle. In addition, a plurality of pieces of actual measurement data including the input data ØEcu and the actual measurement value Pm may be processed as one batch. That is, in one embodiment of the present disclosure, data processing and operation may be performed in units of batches.


Next, the processor 120 executes one or more instructions to input the input data ØECU to the artificial neural network model 20 and obtain an estimation value Pe of the output variable output by the artificial neural network model 20. More specifically, the processor 120 may input the input data ØEcu to the generator 21 of the artificial neural network model 20 and obtain an estimation value of the fluid pressure output by the generator 21.


In addition, the processor 120 executes one or more instructions to input the input data ØEcu and the actual measurement value Pm of the output variable to the artificial neural network model 20 and obtain a discrimination value DS. More specifically, the processor 120 may input the input data ØEcu and the actual measurement value of the fluid pressure to the discriminator 22 and obtain the discrimination value DS.


Next, the processor 120 outputs an error data DError. The error data DError may be output for each data batch.


More specifically, a plurality of data sets each including the input data, the estimation value Pe of the output variable, and the actual measurement value Pm of the output variable may be present.


In this case, the error data DError may include an average μE and a standard deviation OF of errors between the estimation values Pe and the actual measurement values Pm obtained from the plurality of data sets, a maximum absolute error MaxE between the estimation values Pe and the actual measurement values Pm obtained from the plurality of data sets, and discrimination values DS of the discriminator 22 for the input data ØECU and the actual measurement values Pm.


Finally, the processor 120 executes one or more instructions to input the error data DError to the abnormality detection model 30 and obtain an output of the abnormality detection model 30. As described above, the abnormality detection model 30 may include the OCSVM algorithm.


The processor 120 may detect a level of abnormality of the hydraulic pressure generation motor (performance degradation of the hydraulic pressure generation motor) on the basis of the output (feature) obtained from the abnormality detection model 30 which receives the error data DError.


As described above, the hydraulic pressure generation motor abnormality detection device 100 according to one embodiment of the present disclosure has been described in detail. Hereinafter, the hydraulic pressure generation motor abnormality detection method will be described.



FIG. 7 is a flowchart illustrating the hydraulic pressure generation motor abnormality detection method according to one embodiment of the present disclosure.


In a hydraulic pressure generation motor abnormality detection method S100 according to one embodiment of the present disclosure, abnormality of the hydraulic pressure generation motor for driving the pump to generate a hydraulic pressure for performing braking corresponding to an input displacement of the brake pedal of the vehicle is detected.


Referring to FIG. 7, the hydraulic pressure generation motor abnormality detection method S100 according to one embodiment of the present disclosure may be performed as follows.


First, the processor 120 executes one or more instructions to input input data to the artificial neural network model, which receives the input data and outputs an estimation value of an output variable related to an output of the hydraulic pressure generation motor, and obtain the estimation value of the output variable (S110).


The input data may include one or more of a speed of the vehicle, a lateral acceleration of the vehicle, an input displacement of the brake pedal, and a wheel speed of the vehicle. In one embodiment of the present disclosure, the input data may include the speed of the vehicle, the lateral acceleration of the vehicle, the input displacement of the brake pedal, and the wheel speed of the vehicle. In this case, the wheel speed of the vehicle may be obtained for each of the plurality of wheels of the vehicle.


Meanwhile, the input data may be obtained through the CAN 300 of the vehicle.


A fluid pressure generated by the pump may be the output variable. In other words, the estimation value of the output variable may be an estimation value of the fluid pressure.


The estimation value of the output variable is obtained to be compared with an actual measurement value of the output variable. In this case, the actual measurement value of the output variable may be an actual measurement value of the fluid pressure generated by the pump.


In one embodiment of the present disclosure, the artificial neural network model 20 may include the GAN. In other words, the artificial neural network model 20 may be the neural network model based on deep learning for the integrated dynamic brake system of the vehicle.


The artificial neural network model 20 may include the generator 21 and the discriminator 22. The generator 21 receives the input data and generates an estimation value of the output variable. The generator 21 may be provided as the multivariate transformer. In addition, the discriminator 22 may receive the input data and an actual measurement value of the output variable and output an actual measurement related discrimination value.


In operation S110 of obtaining an estimation value of the output variable, the processor 120 may input the input data to the generator 21 and obtain an estimation value of the output variable generated by the generator 21.


Meanwhile, the artificial neural network model 20 may be built by performing learning of the generator 21 and the discriminator 22, and input data and an actual measurement value used for the learning may be obtained in a normal state of each of the vehicle and the integrated dynamic brake system. Accordingly, an estimation value of the output variable may follow the actual measurement value of the output variable obtained in the normal state of the integrated dynamic brake system.


Next, the processor 120 detects whether the hydraulic pressure generation motor is abnormal by using a difference between the estimation value of the output variable and the actual measurement value of the output variable (S120).


The processor 120 may execute one or more instructions to input error data related to the difference between the estimation value and the actual measurement value to the abnormality detection model 30 to determine whether the hydraulic pressure generation motor is abnormal. For example, the abnormality detection model 30 may use the OCSVM algorithm.



FIG. 8 is a detailed flowchart illustrating detecting the abnormality of the hydraulic pressure generation motor in the hydraulic pressure generation motor abnormality detection method according to one embodiment of the present disclosure.


Referring to FIG. 8, operation S120 of detecting whether the hydraulic pressure generation motor is abnormal may be performed as follows.


First, the processor 120 inputs the input data and the actual measurement value of the output variable to the discriminator 22 and obtains an actual measurement related discrimination value generated by the discriminator 22 (S121).


In this case, a discrimination value output by the discriminator 22 for the input data and the actual measurement value of the output variable may be defined as the actual measurement related discrimination value.


Next, the processor 120 inputs the actual measurement related discrimination value and error data including values related to the difference between the estimation value of the output variable and the actual measurement value of the output variable to the abnormality detection model 30 and obtains an output of the abnormality detection model 30 (S122).


There are a plurality of data sets each including the input data, the actual measurement value of the output variable, and the estimation value of the output variable. The error data may include an average and a standard deviation of errors between the actual measurement values of the output variables and the estimation values of the output variables obtained from the plurality of data sets, a maximum absolute error between the actual measurement values of the output variables and the estimation values of the output variables obtained from the plurality of data sets, and discrimination values of the discriminator for the input data and the actual measurement values of the output variables included in the plurality of data sets.


In one embodiment of the present disclosure, the error data may include an average and a standard deviation of errors between the estimation values and the actual measurement values of the fluid pressures obtained from the plurality of data sets, a maximum absolute error between the estimation values and the actual measurement values of the fluid pressures obtained from the plurality of data set, and discrimination values of the discriminator 22 for the input data and the actual measurement values of the fluid pressures included in the plurality of data sets.


A plurality of data sets each including the input data, an estimation value of the output variable, and an actual measurement value of the output variable may be processed as one batch. For example, 100 to 150 data sets (specifically, 128 data sets) may be processed as one batch, and the error data may be obtained for each batch.


The error data obtained for one batch may be input to the abnormality detection model 30. In addition, a level of the abnormality of the hydraulic pressure generation motor (performance degradation of the hydraulic pressure generation motor) may be detected on the basis of the output (feature) obtained from the abnormality detection model 30 which receives the error data.


Meanwhile, the present disclosure further provides the non-transitory computer-readable storage medium in which a program for performing the hydraulic pressure generation motor abnormality detection method is stored. Specifically, the present disclosure may provide the non-transitory computer-readable storage medium in which the program including at least one instruction for performing the hydraulic pressure generation motor abnormality detection method is stored.


In this case, the instruction may include a machine language code generated by a compiler. In addition, the instruction may also include a high-level language code which may be executed by a computer.


The storage medium may include a hardware device, such as magnetic media including a hard disk, a floppy disk, and a magnetic tape, optical media including a CD-ROM and a DVD, a magneto-optical media including a floptical disk, a ROM, a RAM, and a flash memory, configured to store and execute a program instruction.


The above described F1 score may be considered in relation to an anomaly detection metric showing an effect of the present disclosure.


As a result of simulation, in the case of the present disclosure, it is seen that performance of the hydraulic pressure generation motor can be predicted to be degraded to 10% of a maximum output power thereof with a F1 score of 0.85 or more. In other words, in the case of the present disclosure, a degradation level of the hydraulic pressure generation motor, which drives the pump to generate a braking hydraulic pressure corresponding to an input displacement of the brake pedal of the vehicle, can be accurately predicted before the hydraulic pressure generation motor fails.


According to the present disclosure, output degradation of the hydraulic pressure generation motor can be sensitively detected. As a result, a proactive measure can be performed before the hydraulic pressure generation motor fails, and a failure of the hydraulic pressure generation motor can be effectively prevented.


As described above, the present disclosure provides estimation of future operation and remaining effective lifetime of a system and prognostic proper to a predictive maintenance application. Accordingly, a proactive measure and maintenance can be guided before the hydraulic pressure generation motor fails.


According to the above described configuration, the hydraulic pressure generation motor abnormality detection device and method and the non-transitory computer-readable storage medium in which a program for performing the method according to one aspect of the present disclosure is stored can prognose performance degradation of the hydraulic pressure generation motor which drives the pump to generate a hydraulic pressure for performing braking corresponding to an input displacement of the brake pedal of the vehicle on the basis of a digital twin algorithm based on artificial intelligence before the hydraulic pressure generation motor fails.


In addition, the hydraulic pressure generation motor abnormality detection device and method and the non-transitory computer-readable storage medium in which a program for performing the method is stored can detect performance degradation of the hydraulic pressure generation motor using CAN signals without an additional sensor.


It should be understood that the effects of the present disclosure are not limited to the above-described effects, and include all effects inferable from a configuration of the invention described in detailed descriptions or claims of the present disclosure.


Although embodiments of the present disclosure have been described, the spirit of the present disclosure is not limited by the embodiments presented in the specification. Those skilled in the art who understand the spirit of the present disclosure will be able to easily suggest other embodiments by adding, changing, deleting, or adding components within the scope of the same spirit, but this will also be included within the scope of the spirit of the present disclosure.

Claims
  • 1. A device comprising: a memory configured to store one or more instructions; anda processor configured to execute the one or more instructions comprising:inputting input data, associated with one or more operations of a vehicle, to an artificial neural network model configured to, in response to the input data, output an estimation value of an output variable related to an output of a hydraulic pressure generation motor configured to drive a pump configured to generate a hydraulic pressure for performing a brake;obtaining the estimation value of the output variable from the artificial neural network model; anddetecting whether the hydraulic pressure generation motor is in an abnormal state based on a difference between the estimation value of the output variable and an actual measurement value of the output variable.
  • 2. The device of claim 1, wherein the input data, associated with the one or more operations of the vehicle, includes one or more of a speed of the vehicle, a lateral acceleration of the vehicle, the input displacement of the brake pedal, or a wheel speed of the vehicle.
  • 3. The device of claim 1, wherein the output variable includes a fluid pressure generated by the pump.
  • 4. The device of claim 1, wherein the processor is configured to obtain the input data, associated with the one or more operations of the vehicle, through a controller area network (CAN) of the vehicle.
  • 5. The device of claim 1, wherein the artificial neural network model is comprised in a generative adversarial network (GAN) including a generator configured to receive the input data, associated with the one or more operations of the vehicle, and generate the estimation value of the output variable and a discriminator configured to, in response to the input data, associated with the one or more operations of the vehicle, and the actual measurement value of the output variable, output an actual measurement related discrimination value.
  • 6. The device of claim 5, wherein the discriminator is further configured to, in response to the input data, associated with the one or more operations of the vehicle, and the estimation value of the output variable, output an estimation related discrimination value.
  • 7. The device of claim 6, wherein: the artificial neural network model is configured to alternately perform learning of the generator and the discriminator; andthe input data and the actual measurement value of the output variable used for the learning of the generator and the discriminator are obtained when the vehicle and the hydraulic pressure generation motor are in a normal state.
  • 8. The device of claim 5, wherein the generator comprises a multivariate transformer.
  • 9. The device of claim 5, wherein the processor is configured to input error data related to the difference between the estimation value of the output variable and the actual measurement value of the output variable to an abnormality detection model, and determine whether the hydraulic pressure generation motor is the abnormal state based on an output of the abnormality detection model.
  • 10. The device of claim 9, wherein the abnormality detection model is configured to use a one-class support vector machine (OCSVM) algorithm.
  • 11. The device of claim 9, wherein: each of a plurality of data sets includes the input data, the actual measurement value of the output variable, and the estimation value of the output variable; andthe error data includes an average and a standard deviation of errors between actual measurement values of output variables of the plurality of data sets and estimation values of output variables of the plurality of data sets, a maximum absolute error between the actual measurement values of the output variables of the plurality of data sets and the estimation values of the output variables of the plurality of data sets, and discrimination values of the discriminator for the input data and the actual measurement values of the output variables of the plurality of data sets.
  • 12. A method comprising: inputting input date, associated with one or more operations of a vehicle, to an artificial neural network model configured to, in response to the input data, output an estimation value of an output variable related to an output of a hydraulic pressure generation motor configured to drive a pump configured to generate a hydraulic pressure for performing a brake, and obtaining the estimation value of the output variable from the artificial neural network model; anddetecting whether the hydraulic pressure generation motor is in an abnormal state based on a difference between the estimation value of the output variable and an actual measurement value of the output variable.
  • 13. The method of claim 12, wherein the input data, associated with the one or more operations of the vehicle, includes one or more of a speed of the vehicle, a lateral acceleration of the vehicle, the input displacement of the brake pedal, or a wheel speed of the vehicle.
  • 14. The method of claim 12, wherein the output variable includes a fluid pressure generated by the pump.
  • 15. The method of claim 12, wherein the artificial neural network model is comprised in a generative adversarial network (GAN) including a generator configured to receive the input data, associated with the one or more operations of the vehicle, and generate the estimation value of the output variable and a discriminator configured to, in response to the input data, associated with the one or more operations of the vehicle, the actual measurement value of the output variable, output an actual measurement related discrimination value.
  • 16. The method of claim 15, wherein: the artificial neural network model is configured to alternately perform learning of the generator and the discriminator; andthe method further comprises obtaining the input data and the actual measurement value of the output variable used for performing the learning of the generator and the discriminator when the vehicle and the hydraulic pressure generation motor are in a normal state.
  • 17. The method of claim 15, wherein the detecting of whether the hydraulic pressure generation motor is in the abnormal state includes: inputting the input data and the actual measurement value of the output variable to the discriminator and obtaining the actual measurement related discrimination value generated by the discriminator; andinputting the actual measurement related discrimination value and error data including values related to the difference between the estimation value of the output variable and the actual measurement value of the output variable to an abnormality detection model and obtaining an output of the abnormality detection model.
  • 18. The method of claim 17, wherein the abnormality detection model uses a one-class support vector machine (OCSVM) algorithm.
  • 19. The method of claim 17, wherein: each of a plurality of data sets includes the input data, the actual measurement value of the output variable, and the estimation value of the output variable; andthe error data includes an average and a standard deviation of errors between actual measurement values of output variables of the plurality of data sets and estimation values of output variables of the plurality of data sets, a maximum absolute error between the actual measurement values of the output variables of the plurality of data sets and the estimation values of the output variables of the plurality of data sets, and discrimination values of the discriminator for the input data and the actual measurement values of the output variables of the plurality of data sets.
  • 20. A non-transitory computer-readable storage medium configured to store instructions that when executed by one or more processors, cause the one or more processors to perform operations comprising: inputting input date, associated with one or more operations of a vehicle, to an artificial neural network model configured to, in response to the input data, output an estimation value of an output variable related to an output of a hydraulic pressure generation motor configured to drive a pump configured to generate a hydraulic pressure for performing a brake, and obtaining the estimation value of the output variable from the artificial neural network model; anddetecting whether the hydraulic pressure generation motor is in an abnormal state based on a difference between the estimation value of the output variable and an actual measurement value of the output variable.
Priority Claims (2)
Number Date Country Kind
10-2023-0117732 Sep 2023 KR national
10-2024-0060705 May 2024 KR national