Rear wheel steering motor anomaly detection device and method, and computer readable storage medium

Abstract

The invention discloses a rear wheel steering motor anomaly detection device and method and a computer readable storage medium. The rear wheel steering motor abnormality detection apparatus may include: a memory for storing one or more instructions; and a processor that executes the one or more instructions, the processor detecting an abnormality of a rear wheel steering motor for generating a driving force for driving a rack for performing rear wheel steering of the vehicle by executing the one or more instructions, the processor detecting the abnormality of the rear wheel steering motor by executing the one or more instructions, and the processor detecting the abnormality of the rear wheel steering motor by executing the one or more instructions. Input data related to rear wheel steering of the vehicle is input to an artificial neural network model, estimated data related to rear wheel steering output by the artificial neural network model is acquired, and whether the rear wheel steering motor is abnormal is detected by comparing the estimated data with measured data.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to and the benefit of Korean Patent Application No. 10-2023-0106182, filed on August 14, 2023, and Korean Patent Application No. 10-2024-0089573, filed on July 8, 2024, which are hereby incorporated by reference into this application in their entirety. Technical Field

[0003] The present invention relates to a rear-wheel steering motor abnormality detection device and method and a non-transitory computer-readable storage medium storing a program for executing the method, and more specifically, to a rear-wheel steering motor abnormality detection device and method and a non-transitory computer-readable storage medium storing a program for executing the method, which detects abnormality of a rear-wheel steering motor that generates a driving force for driving a rack that performs rear wheel steering of a vehicle. Background Art

[0004] The rear-wheel steering system actively controls the rear-wheel steering angle in coordination with the steering angle of the front wheels according to the vehicle's driving conditions. The rear-wheel steering system reduces the vehicle's turning radius and improves driving stability and ride quality.

[0005] The rear-wheel steering system includes a rack that adjusts the angle of the rear wheels of the vehicle and a rear-wheel steering actuator that drives the rack. Generally, the rear-wheel steering actuator is composed of a motor. That is, the rear-wheel steering actuator can be a rear-wheel steering motor.

[0006] The output torque of the rear-wheel steering motor may decrease due to demagnetization caused by degradation, corrosion of insulating materials, etc. When the output torque of the rear-wheel steering motor decreases, the driver will receive heterogeneous ride feedback that is different from his or her steering intention. This heterogeneous feedback may cause an accident caused by steering error.

[0007] Generally, the output torque of the rear-wheel steering motor decreases gradually. Therefore, in order to improve the safety of the rear-wheel steering system, it is preferred to predict the abnormality of the motor in advance. However, the conventional method for detecting the abnormality of the rear-wheel steering motor still remains at the stage of post-fault diagnosis.

[0008] Patent document: Korean authorized patent No. 2523442, "Steering control device and steering control method", authorized on April 14, 2023. Summary of the invention

[0009] The present invention aims to solve the above-mentioned problems. The purpose of the present invention is to provide a rear-wheel steering motor abnormality detection device and method and a non-temporary computer-readable storage medium storing a program for executing the method, which can detect the performance degradation of the rear-wheel steering motor before a failure occurs. The rear-wheel steering motor is used to generate a driving force for driving a rack that performs rear-wheel steering of a vehicle.

[0010] Another object of the present invention is to provide a rear-wheel steering motor abnormality detection device and method and a non-temporary computer-readable storage medium storing a program for executing the method, which can effectively detect the reduction in the output of the rear-wheel steering motor in advance without using additional sensors in the vehicle other than the sensors configured on the existing vehicle.

[0011] The objects of the present invention are not limited to the above objects, and those skilled in the art to which the present invention pertains will be able to clearly understand other objects not mentioned through the following description.

[0012] According to one aspect of the present invention, there is provided a rear-wheel steering motor abnormality detection device, comprising: a memory for storing one or more instructions; and a processor for executing the one or more instructions, wherein the processor detects an abnormality of the rear-wheel steering motor by executing the one or more instructions, wherein the rear-wheel steering motor is used to generate a driving force for driving a rack for performing rear-wheel steering of a vehicle, wherein the processor inputs input data related to the rear-wheel steering of the vehicle into an artificial neural network model, obtains estimated data related to the rear-wheel steering output by the artificial neural network model, and detects whether the rear-wheel steering motor is abnormal by comparing the estimated data with measured data.

[0013] In the rear-wheel steering motor abnormality detection device according to one aspect of the present invention, the input data may include one or more of the vehicle speed, the shift lever position information of the vehicle, the vehicle steering angle and the vehicle steering angular velocity.

[0014] In a rear-wheel steering motor abnormality detection device according to one aspect of the present invention, the estimated data may include a first estimated value for a first residual determined based on the input data, wherein the first residual is the difference between an instruction and a measured value related to the position of the rack, and the measured data may include a first measured value, which is the measured value of the first residual.

[0015] In a rear-wheel steering motor abnormality detection device according to one aspect of the present invention, the estimated data may also include a second estimated value for a second residual determined based on the input data, wherein the second residual is the difference between an instruction and a measured value related to the motor torque to be generated by the rear-wheel steering motor, and the measured data may also include a second measured value, which is the measured value of the second residual.

[0016] In the rear wheel steering motor abnormality detection device according to one aspect of the present invention, the input data can be acquired through a controller area network (CAN) of the vehicle.

[0017] In the rear wheel steering motor abnormality detection device according to one aspect of the present invention, the artificial neural network model can be composed of a generative adversarial network (GAN), and the generative adversarial network may include: a generator, which receives the input data and generates the estimated data; and a discriminator, which receives the input data and the measured data and outputs a measured related identification value.

[0018] In the rear-wheel steering motor abnormality detection device according to one aspect of the present invention, the generator may be composed of a multivariate transformer.

[0019] In a rear-wheel steering motor abnormality detection device according to one aspect of the present invention, the processor can input error data related to the difference between the estimated data and the measured data into an abnormality detection model and determine whether the rear-wheel steering motor is abnormal by executing one or more instructions.

[0020] In the rear-wheel steering motor abnormality detection device according to one aspect of the present invention, the abnormality detection model may use a one-class support vector machine (OCSVM) algorithm.

[0021] In the rear-wheel steering motor abnormality detection device according to one aspect of the present invention, there may be multiple data sets including the input data, the estimated data and the measured data, and the error data may include the average error and standard deviation between the measured data and the estimated data of the multiple data sets, the maximum absolute error between the measured data and the estimated data of the multiple data sets, and the measured related identification value of the identifier related to the input data and the measured data included in the multiple data sets.

[0022] In the rear-wheel steering motor abnormality detection device according to one aspect of the present invention, the discriminator may additionally receive the input data and the estimation data, and additionally output an estimation-related discrimination value.

[0023] In a rear-wheel steering motor abnormality detection device according to one aspect of the present invention, the artificial neural network model is constructed by alternating learning of the generator and the discriminator, and the input data and the measured data used in the learning may be data obtained when the vehicle and the rear-wheel steering motor are in normal states.

[0024] According to another aspect of the present invention, a method for detecting abnormality of a rear-wheel steering motor can be provided, comprising the following steps: a processor inputs input data related to the rear-wheel steering of a vehicle into an artificial neural network model, and obtains estimated data related to the rear-wheel steering output by the artificial neural network model; and the processor detects whether the rear-wheel steering motor is abnormal by comparing the estimated data with measured data, wherein the rear-wheel steering motor is used to generate a driving force for driving a rack for performing the rear-wheel steering of the vehicle.

[0025] In the rear wheel steering motor abnormality detection method according to one aspect of the present invention, the input data may include one or more of the vehicle speed, the shift lever position information of the vehicle, the vehicle steering angle and the vehicle steering angular velocity.

[0026] In a rear-wheel steering motor abnormality detection method according to one aspect of the present invention, the estimated data may include a first estimated value for a first residual determined based on the input data, wherein the first residual is the difference between an instruction and a measured value related to the position of the rack, and the measured data may include a first measured value, which is the measured value of the first residual.

[0027] In a rear-wheel steering motor abnormality detection method according to one aspect of the present invention, the estimated data may also include a second estimated value for a second residual determined based on the input data, wherein the second residual is the difference between an instruction and a measured value related to the motor torque to be generated by the rear-wheel steering motor, and the measured data may also include a second measured value, which is the measured value of the second residual.

[0028] In a rear-wheel steering motor abnormality detection method according to one aspect of the present invention, the artificial neural network model can be composed of a generative adversarial network, which includes: a generator, which receives the input data and generates the estimated data; and a discriminator, which receives the input data and the measured data and outputs a measured related identification value.

[0029] In a rear-wheel steering motor abnormality detection method according to one aspect of the present invention, the step of detecting whether the rear-wheel steering motor is abnormal may include the following steps: the processor inputs the input data and the measured data into the discriminator and obtains the measured related identification value generated by the discriminator; and the processor inputs the error data into the abnormality detection model and obtains the output of the abnormality detection model, the error data including the measured related identification value and a numerical value related to the difference between the estimated data and the measured data.

[0030] In the rear-wheel steering motor abnormality detection method according to one aspect of the present invention, there may be multiple data sets including the input data, the estimated data and the measured data, and the error data may include the average error and standard deviation between the measured data and the estimated data of the multiple data sets, the maximum absolute error between the measured data and the estimated data of the multiple data sets, and the measured related identification value of the identifier related to the input data and the measured data included in the multiple data sets.

[0031] According to still another aspect of the present invention, there is provided a non-transitory computer-readable storage medium storing a program including at least one instruction for executing the rear-wheel steering motor abnormality detection method. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The above and other objects, features and advantages of the present invention will become more apparent to those skilled in the art by describing in detail exemplary embodiments of the present invention with reference to the accompanying drawings.

[0033] Figure 1 It is a diagram schematically showing the configuration of a rear-wheel steering system of a vehicle.

[0034] Figure 2 is a diagram showing the detailed configuration and operation of a rear-wheel steering motor controller of a rear-wheel steering system.

[0035] Figure 3 : is a diagram showing a configuration of a rear-wheel steering motor abnormality detection device according to an embodiment of the present invention.

[0036] Figure 4 FIG. 1 is a diagram showing a model used to obtain an estimated value or detect abnormality in a rear-wheel steering motor abnormality detection device according to an embodiment of the present invention.

[0037] Figure 5 is a diagram showing a detailed configuration of an artificial neural network model.

[0038] Figure 6 is a diagram showing a detailed configuration of a generator of an artificial neural network model.

[0039] Figure 7 1 is a diagram illustrating the operation of the rear-wheel steering motor abnormality detection device according to an embodiment of the present invention.

[0040] Figure 8 is a sequence diagram of a method for detecting abnormality of a rear wheel steering motor according to an embodiment of the present invention.

[0041] Fig. 9 Detailed sequence diagram of the steps of detecting whether a rear-wheel steering motor is abnormal in a rear-wheel steering motor abnormality detection method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0042] Embodiments of the present invention will be described in detail below so that those skilled in the art to which the present invention belongs can easily implement the embodiments. The present invention can be implemented in a variety of different forms and is not limited to the embodiments described herein. In order to clearly describe the present invention, parts not related to the description are omitted in the accompanying drawings, and the same or similar parts are represented by the same reference numerals throughout the specification.

[0043] The words and terms used in the specification and claims should not be restrictively interpreted as their ordinary meanings or dictionary meanings, but should be interpreted as meanings and concepts consistent with the technical spirit of the present invention in accordance with the principle that the inventor is able to define terms and concepts to best describe his invention.

[0044] In the specification, it should be understood that terms such as “including” or “having” are intended to specify the existence of the features, numbers, steps, actions, components, parts or their combinations described in the specification, and do not exclude the possibility of the existence or addition of one or more other features, numbers, steps, actions, components, parts or their combinations.

[0045] Figure 1 It is a diagram schematically showing the configuration of a rear-wheel steering system of a vehicle.

[0046] The rear-wheel steering system 200 actively controls the steering angle of the rear wheels RW in coordination with the steering angle of the front wheels according to the driving conditions of the vehicle. The rear-wheel steering system 200 can control the steering angle of the rear wheels RW to be in the same direction as the steering angle of the front wheels of the vehicle or in the opposite direction to the steering angle of the front wheels of the vehicle according to the driving conditions.

[0047] More specifically, when the vehicle is traveling at low or medium speeds, the rear wheels RW are controlled to be opposite to the front wheels, thereby reducing the turning radius of the vehicle. And when the vehicle is traveling at high speeds, the rear wheels RW are controlled to be in the same direction as the front wheels, thereby improving the stability of turning.

[0048] Reference Figure 1The rear-wheel steering system 200 includes a sensing unit 210 that senses information related to the driving state of the vehicle. The sensing unit 210 may include: a vehicle speed sensor 211 that detects the speed of the vehicle; a gear lever position sensor 212 that detects the position information of the gear lever of the vehicle; and a steering angle sensor 213 that detects the steering angle and steering angle velocity of the steering wheel of the vehicle.

[0049] The sensing information of the sensing unit 210 is transmitted to the electronic control unit 250. For example, the speed of the vehicle from the vehicle speed sensor 211, the gear lever position information of the vehicle obtained by the gear lever position sensor 212, and the steering angle and steering angular velocity of the vehicle from the steering angle sensor 213 can be transmitted to the electronic control unit 250.

[0050] The electronic control unit 250 generates input data for rear wheel steering based on the sensing information. The input data generated by the electronic control unit 250 may be transmitted to a rear wheel steering motor controller 260 for controlling a rear wheel steering motor 270 that generates a driving force for rear wheel steering.

[0051] The rear-wheel steering motor controller 260 controls the rear-wheel steering motor 270 based on the input data. The rear-wheel steering motor 270 allows the rack 280 connected to the rear wheels RW of the vehicle to move. When the rack 280 moves according to the driving of the rear-wheel steering motor 270, the angle of the rear wheels RW of the vehicle changes, so that the rear-wheel steering can be achieved.

[0052] Figure 2 is a diagram showing the detailed configuration and operation of a rear-wheel steering motor controller of a rear-wheel steering system.

[0053] Reference Figure 2 The rear wheel steering motor controller 260 may include a command generator 261 , a rack position controller 262 , a rear wheel steering motor torque controller 263 and a rear wheel steering motor current controller 264 .

[0054] The command generator 261 may generate a command rack position Pd, which is a position of the rack 280 that needs to be set according to input data. At this time, the input data may include any one or more of the vehicle speed, the vehicle shift lever position information, the vehicle steering angle, and the vehicle steering angular velocity.

[0055] The rack position controller 262 receives the difference between the command rack position Pd and the measured rack position Pm and generates a command rear-wheel steering motor torque Td, which is a torque that the rear-wheel steering motor 270 should generate.

[0056] The difference between the command rack position Pd and the measured rack position Pm is the difference between the command and the measured value related to the rack position, which can be defined as a first residual. In other words, the rack position controller 262 can receive the first residual and generate a command rear wheel steering motor torque Td.

[0057] The rear-wheel steering motor torque controller 263 receives the difference between the command rear-wheel steering motor torque Td and the measured rear-wheel steering motor torque Tm measured by the rear-wheel steering motor 270 and generates a command current Id which is a current that should be supplied to the rear-wheel steering motor 270 .

[0058] The difference between the command rear wheel steering motor torque Td and the measured rear wheel steering motor torque Tm is the difference between the command and the measured value related to the motor torque, and can be defined as a second residual. In other words, the rear wheel steering motor torque controller 263 can receive the second residual and generate a command current Id.

[0059] The rear-wheel steering motor current controller 264 receives an input current reflecting feedback of the command current Id and the measured current Im and controls the current supplied to the rear-wheel steering motor 270. Thus, the voltage Vd input to the rear-wheel steering motor 270 can be determined.

[0060] On the other hand, the measured current Im of the motor can be calculated based on the angle θm (position) of the rear-wheel steering motor 270. Also, the measured rear-wheel steering motor torque Tm can be estimated based on the measured current Im of the rear-wheel steering motor 270.

[0061] Figure 3 : is a diagram showing the configuration of a rear wheel steering motor abnormality detection device according to an embodiment of the present invention. And, Figure 4 FIG. 1 is a diagram showing a model used to obtain an estimated value or detect abnormality in a rear-wheel steering motor abnormality detection device according to an embodiment of the present invention.

[0062] The rear-wheel steering motor abnormality detection device 100 according to an embodiment of the present invention detects an abnormality of a rear-wheel steering motor 270 that generates a driving force for driving a rack 280 for performing rear-wheel steering of a vehicle. More specifically, the rear-wheel steering motor abnormality detection device 100 can detect a decrease in output before the rear-wheel steering motor 270 enters a fault state in which it is difficult to operate.

[0063] According to an embodiment of the present invention, the rear-wheel steering motor abnormality detection device 100 can predict the functional degradation of the rear-wheel steering motor 270 before the rear-wheel steering motor 270 fails by using an artificial intelligence-based digital twin algorithm. The digital twin algorithm can virtually construct a rear-wheel steering system of a vehicle.

[0064] Reference Figure 3 According to an embodiment of the present invention, a rear wheel steering motor abnormality detection device 100 may include a memory 110 and a processor 120 .

[0065] The memory 110 stores one or more instructions that can be executed by the processor 120 .

[0066] The memory 110 may include a hardware device configured to store and execute program instructions. For example, the memory 110 may include storage media such as a read-only memory (ROM), a random access memory (RAM), a flash memory, etc. In addition, the memory 110 may also include magnetic media such as floppy disks and magnetic tapes, optical recording media such as compact disk read only memory (CD-ROM), digital video disks (DVD), and magneto-optical media such as floptical disks.

[0067] The processor 120 executes the one or more instructions. For example, the processor 120 may be a hardware unit that performs calculations and controls in a computer. The processor 120 may include at least one arithmetic logic unit (ALU) and a register.

[0068] The processor 120 inputs the input data related to the steering of the vehicle into the artificial neural network model 20 by executing the one or more instructions and obtains the rear wheel steering related estimation data output by the artificial neural network model 20.

[0069] The input data may include any one or more of the speed of the vehicle, the position information of the gear lever of the vehicle, the steering angle of the vehicle, and the steering angular velocity of the vehicle. For example, the input data may include the speed of the vehicle, the position information of the gear lever of the vehicle, the steering angle of the vehicle, and the steering angular velocity of the vehicle.

[0070] The input data may be acquired from a controller area network (CAN) 300 of the vehicle. For example, the input data may be transmitted to the memory 110 via the CAN 300.

[0071] As described above, the input data may be acquired through the vehicle CAN 300. Thus, according to the present invention, there is no need to use an additional sensor associated with detecting abnormality of the rear-wheel steering motor 270.

[0072] 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 from the CAN 300. Thus, the processor 120 may obtain the input data from the memory 110.

[0073] The estimated data may include a first estimated value for a first residual determined based on the input data, the first residual being a difference between a command and a measured value related to the position of the rack, and a second estimated value for a second residual determined based on the input data, the second residual being a difference between a command and a measured value related to the motor torque to be generated by the rear wheel steering motor 270.

[0074] The artificial neural network model 20 receives the input data and outputs the estimated data. More specifically, the artificial neural network model 20 may receive the input data and output a first estimated value and a second estimated value.

[0075] The estimated data is used for comparison with measured data related to rear wheel steering. At this time, the measured data may include a first measured value as a measured value of the first residual and a second measured value as a measured value of the second residual.

[0076] In one embodiment of the present invention, the artificial neural network model 20 may be composed of a generative adversarial network.

[0077] In one embodiment of the present invention, the artificial neural network model 20 can be a deep learning-based neural twin model of the rear-wheel steering system of the vehicle. In other words, the artificial neural network model 20 can be used as a virtual twin model of the rear-wheel steering system of the vehicle.

[0078] Figure 5 is a diagram showing a detailed configuration of an artificial neural network model.

[0079] Reference Figure 5 , the artificial neural network model 20 may include a generator 21 and a discriminator 22.

[0080] The artificial neural network model 20 is constructed by alternately learning the generator 21 and the discriminator 22, and the data used in the learning may be obtained when the vehicle and the rear-wheel steering system are in a normal state. Thus, the first estimated value and the second estimated value included in the estimated data may follow the first measured value and the second measured value obtained when the vehicle and the rear-wheel steering system are in a normal state.

[0081] The generator 21 receives the input data and generates the estimated data. The generator 21 may include a neural network. The generator 21 may be composed of a multivariable transformer. That is, the artificial neural network model 20 may be a generative adversarial network based on a multivariable transformer.

[0082] In one embodiment of the present invention, the generator 21 may receive input data related to the rear wheel steering of the vehicle. As described above, the input data may include the speed of the vehicle, the steering angle of the steering wheel, and the driving force of the rack 280 that receives the driving force from the rear wheel steering motor 270 and allows the rear wheels RW of the vehicle to move.

[0083] The generator 21 may calculate the input data and output estimated data. More specifically, the generator 21 may output a first estimated value of the first residual and a second estimated value of the second residual.

[0084] The estimated data output by the generator 21 follows the first measured value of the first residual and the second measured value of the second residual output according to the input data when the rear-wheel steering system of the vehicle is in a normal state.

[0085] Figure 6 is a diagram showing a detailed configuration of a generator of an artificial neural network model.

[0086] Reference Figure 6 , the generator 21 may include an encoder 21a and a decoder 21b.

[0087] The encoder 21a and the decoder 21b constitute a network for estimating estimation data related to the rear wheel steering of the vehicle. In other words, the generator 21 can be composed of an encoder 21a-decoder 21b structure network trained in an end-to-end manner to estimate the estimation data.

[0088] The encoder 21a may be designed to extract high-level features from the input data using a transformer block and linear operations. In this case, the transformer block may include an embedding layer, a self-attention layer, and a feed-forward layer.

[0089] First, the embedding layer in the Transformer block can encode spatiotemporal information into input sequences using a trainable embedding matrix.

[0090] Next, deep features can be extracted through a self-attention layer, and refinement can be performed using a feed-forward layer.

[0091] At this point, the self-attention layer can utilize the scaled dot-product attention mechanism. The self-attention layer can be the main building block in the transformer-based deep neural network (DNN) architecture to enable the model to process and understand dependencies in the input sequence.

[0092] The scaled dot-product attention first applies layer normalization to the input sequence and processes the result into three unique vectors, namely the query (Q) vector, the key (K) vector, and the value (V) vector, through linear operations.

[0093] Next, the attention score (AS) of each position in the input sequence can be calculated based on the derived vector. The calculation formula of the attention score (AS) can be as follows:

[0094] Formula 1:

[0095]

[0096] Among them, σ(·) represents the softmax operation, ∥x input ∥ dim represents the feature dimension of the input, and (·) represents the dot product.

[0097] The attention score quantifies the relative importance of each position in the input sequence with respect to a given query. That is, higher scores are assigned to more appropriate positions.

[0098] The output of the transformer block can be provided to two different linear operations to compute the query, key vector and initial input value of decoder 21b

[0099] The decoder 21b may use the acquired values ​​to estimate the one or more estimated values ​​in an autoregressive manner similar to Sequence-to-Sequence Recurrent Neural Networks.

[0100] Similar to the encoder 21a, the decoder 21b may include a transformer block and a linear operation. In addition, the decoder 21b may also include a cross-attention layer within the transformer block.

[0101] The criss-cross attention layer can perform scaled dot-product attention using the query vector and key vector of encoder 21a to learn the dynamic correlation relationship between the input signal and the estimated signal.

[0102] The discriminator 22 receives the input data and the measured data and outputs a discrimination value. The discrimination value output by the discriminator 22 for the input data and the measured data may be defined as a measured-related discrimination value.

[0103] On the other hand, the discriminator 22 may additionally receive the input data and the estimated data and output a discriminant value. The discriminant value output by the discriminator 22 for the input data and the estimated data may be defined as an estimated correlation discriminant value. This process may be completed when the generator 21 and the discriminator 22 are learning.

[0104] As described above, the discriminator 22 outputs discrimination values ​​for the input data and the measured data in the operation for detecting abnormality. On the other hand, in the operation for learning, the discriminator 22 may output discrimination values ​​for the input data and the estimated data.

[0105] The discriminator 22 can be configured as a discriminator network for adversarial training of a generative adversarial network. In other words, the generator 21 and the discriminator 22 can be trained by adversarial training to improve the estimation performance and establish a generative adversarial network model.

[0106] More specifically, the discriminator 22 may perform optimization alternately with the generator 21 to solve a Wasserstein min-max problem such as Equation 2,

[0107] Formula 2:

[0108]

[0109] Among them, x D,Real represents the actual data set consisting of input data and measured data related to rear wheel steering, x D,Fake represents a virtual data sample consisting of input data and rear-wheel steering related estimation data.

[0110] In the artificial neural network model 20, the generator 21 is trained to deceive the discriminator 22 for distinguishing the spatiotemporal characteristics of xD, Real and xD, Fake. Thus, the generator 21 can generate an estimated value of the rear wheel steering system that follows the normal state.

[0111] In this regard, the loss function for training the generative adversarial network can be defined as the following equation 3:

[0112] Formula 3:

[0113]

[0114] Where N is the batch size, L G is the loss function of the generator, L D is the loss function of the discriminator.

[0115] The first term of the loss of generator 21 corresponds to the mean absolute error loss of supervised training (L MAE ). The remaining terms of the loss of the generator 21 together with the loss of the discriminator 22 constitute the generative adversarial network loss function (L GAN ).

[0116] Furthermore, the processor 120 detects whether the rear wheel steering motor 270 is abnormal by comparing the estimated data with the measured data. As described above, the estimated data may include a first estimated value of the first residual and a second estimated value of the second residual. Furthermore, the measured data may include a first measured value as a measured value of the first residual and a second measured value as a measured value of the second residual.

[0117] The processor 120 inputs the error data related to the difference between the estimated data and the measured data into the abnormality detection model 30 by executing the one or more instructions and determines whether the rear wheel steering motor is abnormal. For example, the abnormality detection model 30 can use a single-class support vector machine algorithm.

[0118] There are a plurality of data sets including the input data, the estimated data and the measured data. The error data may include an average error and a standard deviation between the measured data and the estimated data of the plurality of data sets, a maximum absolute error between the measured data and the estimated data of the plurality of data sets, and a measured related identification value of the identifier 22 related to the input data and the measured data included in the plurality of data sets.

[0119] In one embodiment of the present invention, the error data may include the average error between the first estimated value and the first measured value, the standard deviation between the first estimated value and the first measured value, the maximum absolute error between the first estimated value and the first measured value, the average error between the second estimated value and the second measured value, the standard deviation between the second estimated value and the second measured value, the maximum absolute error between the second estimated value and the second measured value, and an identification value of an identifier associated with the input data, the first measured value, and the second measured value.

[0120] A plurality of data sets including the input data, the estimated data, and the measured data may be considered as a batch. For example, 100 to 150 data sets (as a specific example, 128 data sets) may be considered as a batch, and the error data of each batch may be obtained.

[0121] The error data obtained for one batch may be input to the abnormality detection model 30. And, the abnormality level of the rear-wheel steering motor (performance degradation of the rear-wheel steering motor) may be detected based on the output (feature) obtained from the abnormality detection model 30 that received the error data.

[0122] Regarding the anomaly detection metric, we can consider the F1 score given by equation 4 below:

[0123] Formula 4:

[0124]

[0125] Among them, TP is a true positive, FP is a false positive, FN is a false negative, and a positive example indicates that the performance of the rear wheel steering motor is reduced.

[0126] On the other hand, as described above, the discriminator 22 can calculate not only the discriminant value associated with the input data and measured data included in the data set, but also the discriminant value associated with the input data and estimated data included in the data set. The discriminant value of the discriminator 22 for the estimated data of the data set can be provided as feedback to the generator 21 of the artificial neural network model 20 for training.

[0127] So far, the configuration of the rear wheel steering motor abnormality detection device 100 according to an embodiment of the present invention has been described in detail. Next, the operation of the rear wheel steering motor abnormality detection device 100 will be described in detail.

[0128] Figure 7 1 is a diagram illustrating the operation of the rear-wheel steering motor abnormality detection device according to an embodiment of the present invention.

[0129] Reference Figure 7 , the rear wheel steering motor abnormality detection device 100 according to an embodiment of the present invention can work as follows.

[0130] First, the memory 110 stores input data At this time, the measured data may include the first measured value R1 of the measured first residual m and the second measured value R2 of the measured second residualm .

[0131] As mentioned above, input data The input data may include any one or more of the vehicle speed, the gear lever position information of the vehicle, the steering angle of the vehicle, and the steering angular velocity of the vehicle. The first measured value R1 m and the second measured value R2 m The measured data can be regarded as a batch. That is, in one embodiment of the present invention, data processing and calculation can be performed in units of data batches.

[0132] Next, the processor 120 executes one or more instructions to convert the input data The processor 120 inputs the data into the artificial neural network model 20, and obtains the estimated data output by the artificial neural network model 20. More specifically, the processor 120 The first estimated value R1 outputted by the generator 21 as the estimated value of the first residual is obtained. e and the second estimate R2 which is the estimate of the second residual e estimated data.

[0133] In addition, the processor 120 executes one or more instructions to convert the input data The first measured value R1 m and the second measured value R2 m The input data is input into the artificial neural network model 20 and the identification value DS is obtained. More specifically, the processor 120 can The first measured value R1 m and the second measured value R2 m It is input to the discriminator 22 and obtains the discrimination value DS.

[0134] Next, the processor 120 outputs the error data D Error . Error data D can be output for each data batch Error .

[0135] More specifically, there may be a plurality of data sets including the input value, the one or more estimated values, and the one or more measured values.

[0136] At this time, the error data D Error The first estimated value R1 may be included e and the first measured value R1 m The average error between E(R1) , the first estimated value R1 e and the first measured value R1 m The standard deviation betweenE(R1) , the first estimated value R1 e and the first measured value R1 m The maximum absolute error between E(R1) , the second estimated value R2 e and the second measured value R2 m The average error between E(R2) , the second estimated value R2 e and the second measured value R2 m The standard deviation between E(R2) , the second estimated value R2 e and the second measured value R2 m The maximum absolute error between E(R2) And for input data The first measured value R1 m and the second measured value R2 m The discriminator's discriminator value DS.

[0137] Finally, the processor 120 executes one or more instructions to convert the error data D Error The anomaly detection model 30 is inputted, and the output of the anomaly detection model 30 is obtained. As described above, the anomaly detection model 30 may include a single-class support vector machine algorithm.

[0138] Based on the received error data D Error Based on the output (features) obtained by the abnormality detection model 30, the processor 120 can detect the abnormality (performance reduction of the rear-wheel steering motor) level of the rear-wheel steering system's rear-wheel steering motor.

[0139] The above is a detailed description of the rear wheel steering motor abnormality detection device 100 according to an embodiment of the present invention. Next, a rear wheel steering motor abnormality detection method will be described in detail.

[0140] Figure 8 is a sequence diagram of a method for detecting abnormality of a rear wheel steering motor according to an embodiment of the present invention.

[0141] In the rear-wheel steering motor abnormality detection method S100 according to an embodiment of the present invention, abnormality of a rear-wheel steering motor that generates a driving force for driving a rack for performing rear-wheel steering of a vehicle is detected.

[0142] Reference Figure 8 , the rear wheel steering motor abnormality detection method S100 according to an embodiment of the present invention can be performed as follows.

[0143] First, the processor 120 inputs input data related to the rear wheel steering of the vehicle into the artificial neural network model 20, and obtains rear wheel steering related estimation data output by the artificial neural network model 20 (step S110).

[0144] The input data may include any one or more of the speed of the vehicle, the position information of the gear lever of the vehicle, the steering angle of the vehicle, and the steering angular velocity of the vehicle. In one embodiment of the present invention, the input data may include the speed of the vehicle, the position information of the gear lever of the vehicle, the steering angle of the vehicle, and the steering angular velocity of the vehicle.

[0145] Alternatively, the input data may be acquired from the CAN 300 of the vehicle.

[0146] The estimated data may include a first estimated value for a first residual determined based on the input data, the first residual being a difference between a command and a measured value related to the position of the rack 280. Furthermore, the estimated data may include a second estimated value for a second residual determined based on the input data, the second residual being a difference between a command and a measured value related to the motor torque to be generated by the rear wheel steering motor 270.

[0147] The estimated data is obtained for comparison with steering-related measured data. At this time, the measured data may include a first measured value as a measured value of the first residual and a second measured value as a measured value of the second residual.

[0148] In one embodiment of the present invention, the artificial neural network model 20 may include a generative adversarial network. In other words, the artificial neural network model 20 may be a neural twin model based on deep learning of the rear wheel steering system of the vehicle.

[0149] The artificial neural network model 20 may include a generator 21 and a discriminator 22. The generator 21 receives the input data and generates the estimated data. The generator 21 may be composed of a multivariable transformer. And, the discriminator 22 may receive the input data and the measured data and output a measured related discriminant value.

[0150] In the step S110 of acquiring the one or more estimated values, the processor 120 may input the input data into the generator 21 and acquire the estimated data generated by the generator 21 .

[0151] On the other hand, the artificial neural network model 20 is constructed by learning through the generator 21 and the discriminator 22, and the input data and measured data used in the learning can be obtained when the vehicle and the rear-wheel steering system are in a normal state. Thus, the estimated data can follow the rear-wheel steering related measured data obtained when the rear-wheel steering system is in a normal state.

[0152] Next, the processor 120 compares the estimated data with the measured data and detects whether the rear wheel steering motor is abnormal (step S120).

[0153] The processor 120 inputs the error data related to the difference between the estimated data and the measured data into the abnormality detection model 30 by executing one or more instructions and determines whether the rear wheel steering motor is abnormal. For example, the abnormality detection model 30 can use a single-class support vector machine algorithm.

[0154] Fig. 9 Detailed sequence diagram of the steps of detecting whether a rear-wheel steering motor is abnormal in a rear-wheel steering motor abnormality detection method according to an embodiment of the present invention.

[0155] Reference Fig. 9 , step 120 of detecting whether the rear wheel steering motor is abnormal can be performed as follows.

[0156] First, the processor 120 inputs the input data and the measured data into the discriminator 22, and obtains the measured correlation discrimination value generated by the discriminator 22 (step S121).

[0157] Next, the processor 120 inputs the error data including the measured related identification value and a numerical value related to the difference between the estimated data and the measured data into the abnormality detection model 30 and obtains the output of the abnormality detection model 30 (step S122).

[0158] There may be a plurality of data sets including the input data, the estimated data and the measured data. The error data may include an average error and a standard deviation between the measured data and the estimated data of the plurality of data sets, a maximum absolute error between the measured data and the estimated data of the plurality of data sets, and a discrimination value of the discriminator 22 associated with the input data and the measured data obtained from the plurality of data sets.

[0159] In one embodiment of the present invention, the error data may include the average error between the first estimated value and the first measured value, the standard deviation between the first estimated value and the first measured value, the maximum absolute error between the first estimated value and the first measured value, the average error between the second estimated value and the second measured value, the standard deviation between the second estimated value and the second measured value, the maximum absolute error between the second estimated value and the second measured value, and an identification value of an identifier associated with the input data, the first measured value, and the second measured value.

[0160] A plurality of data sets including the estimated data and the measured data may be regarded as a batch. For example, 100 to 150 data sets (as a specific example, 128 data sets) may be regarded as a batch, and the error data of each batch may be obtained.

[0161] The error data obtained for one batch may be input to the abnormality detection model 30. And, the abnormality level of the rear-wheel steering motor (performance degradation of the rear-wheel steering motor) may be detected based on the output (feature) obtained from the abnormality detection model 30 that received the error data.

[0162] On the other hand, the present invention additionally provides a non-transitory computer-readable storage medium storing a program for executing the rear-wheel steering motor abnormality detection method. Specifically, the present invention may provide a non-transitory computer-readable storage medium storing a program including at least one instruction for executing the rear-wheel steering motor abnormality detection method.

[0163] At this time, the instructions may include machine codes generated by a compiler and high-level language codes executable by a computer.

[0164] The storage medium may include hardware devices configured to store and execute program instructions such as magnetic media (Magnetic Media) such as hard disks, floppy disks and magnetic tapes, optical recording media (Optical Media) such as compact disk read only memory (CD-ROM) and digital video disk (Digital Video Disk, DVD), magneto-optical media (Magneto-Optical Media) such as floptical disks (Floptical Disk), read-only memory (ROM), random access memory (RAM), flash memory, etc.

[0165] Regarding the anomaly detection metric, one may consider the F1 score as described above.

[0166] As a result of the simulation, it was found that according to the present invention, it was possible to predict the performance of the F1 score of 0.85 or more until the rear-wheel steering motor is reduced to 10% relative to the maximum output. In other words, according to the present invention, before the rear-wheel steering motor that is configured in the rear-wheel steering system and generates the driving force for rear-wheel steering completely fails, the degradation level of the rear-wheel steering motor can be accurately predicted.

[0167] According to the present invention, before the rear wheel steering motor of the rear wheel steering system fails, it can be predicted. As a result, active measures can be taken before the rear wheel steering motor fails, and the rear wheel steering motor failure can be prevented.

[0168] As described above, the present invention provides an estimation of the future behavior and remaining service life of the system and a predictive diagnosis suitable for predictive maintenance applications, thereby effectively taking proactive measures and maintenance before the rear wheel steering motor of the rear wheel steering system fails.

[0169] According to the above configuration, a rear-wheel steering motor abnormality detection device and method according to one aspect of the present invention and a non-temporary computer-readable storage medium storing a program for executing the method can predict performance degradation in advance before a rear-wheel steering motor that generates a driving force for driving a rack that performs rear-wheel steering of a vehicle fails based on an artificial intelligence-based digital twin algorithm.

[0170] In addition, a rear-wheel steering motor abnormality detection device and method according to one aspect of the present invention and a non-transitory computer-readable storage medium storing a program for executing the method can use a controller area network signal of a vehicle to predict performance degradation of the rear-wheel steering motor without any additional sensors.

[0171] It should be understood that the effects of the present invention are not limited to the above-mentioned effects, and include all effects inferable from the inventive configuration described in the detailed description of the present invention or the claims.

[0172] Although the embodiments of the present invention have been described, the spirit of the present invention is not limited by the embodiments set forth in the specification. As long as those skilled in the art understand the spirit of the present invention, they can easily propose other embodiments within the scope of the same spirit by adding, changing, deleting or adding components, but this is also included in the scope of the spirit of the present invention.

Claims

1. A rear wheel steering motor abnormality detection device, wherein: include: a memory for storing one or more instructions; as well as a processor, executing said one or more instructions, The processor detects an abnormality of a rear-wheel steering motor by executing the one or more instructions, the rear-wheel steering motor being used to generate a driving force for driving a rack for performing rear-wheel steering of the vehicle, The processor inputs input data related to the rear wheel steering of the vehicle into an artificial neural network model, obtains estimated data related to the rear wheel steering output by the artificial neural network model, and detects whether the rear wheel steering motor is abnormal by comparing the estimated data with the measured data.

2. The rear wheel steering motor abnormality detection device according to claim 1, wherein: The input data includes one or more of a speed of the vehicle, position information of a shift lever of the vehicle, a steering angle of the vehicle, and a steering angular velocity of the vehicle.

3. The rear wheel steering motor abnormality detection device according to claim 2, wherein: the estimated data comprising a first estimate of a first residual determined from the input data, the first residual being a difference between a command and a measurement associated with a position of the rack, The measured data includes a first measured value, which is a measured value of the first residual.

4. The rear wheel steering motor abnormality detection device according to claim 3, wherein: The estimated data also includes a second estimated value for a second residual determined based on the input data, the second residual being a difference between a command and a measurement related to a motor torque to be generated by the rear wheel steering motor, The measured data further includes a second measured value, which is a measured value of the second residual.

5. The rear wheel steering motor abnormality detection device according to claim 1, wherein: The input data is obtained via a controller area network of the vehicle.

6. The rear wheel steering motor abnormality detection device according to claim 1, wherein: The artificial neural network model is composed of a generative adversarial network. The generative adversarial network comprises: a generator receiving the input data and generating the estimated data; and The discriminator receives the input data and the measured data and outputs a measured related discriminant value.

7. The rear wheel steering motor abnormality detection device according to claim 6, wherein: The processor, by executing the one or more instructions, inputs error data related to the difference between the estimated data and the measured data into an abnormality detection model and determines whether the rear wheel steering motor is abnormal.

8. The rear wheel steering motor abnormality detection device according to claim 7, wherein: There are multiple data sets including the input data, the estimated data and the measured data, The error data includes an average error and a standard deviation between measured data and estimated data of the plurality of data sets, a maximum absolute error between measured data and estimated data of the plurality of data sets, and a measured-related identification value of the discriminator associated with input data and measured data included in the plurality of data sets.

9. A method for detecting abnormality of a rear wheel steering motor, wherein: The steps include: The processor inputs input data related to rear wheel steering of the vehicle into the artificial neural network model, and obtains estimated data related to rear wheel steering output by the artificial neural network model; as well as The processor detects whether a rear-wheel steering motor for generating a driving force for driving a rack for performing rear-wheel steering of the vehicle is abnormal by comparing the estimated data with actual measurement data.

10. The rear wheel steering motor abnormality detection method according to claim 9, wherein: The input data includes one or more of a speed of the vehicle, position information of a shift lever of the vehicle, a steering angle of the vehicle, and a steering angular velocity of the vehicle.

11. The rear wheel steering motor abnormality detection method according to claim 10, wherein: the estimated data comprising a first estimate of a first residual determined from the input data, the first residual being a difference between a command and a measurement associated with a position of the rack, The measured data includes a first measured value, which is a measured value of the first residual.

12. The rear wheel steering motor abnormality detection method according to claim 11, wherein: The estimated data also includes a second estimated value for a second residual determined based on the input data, the second residual being a difference between a command and a measurement related to a motor torque to be generated by the rear wheel steering motor, The measured data further includes a second measured value, which is a measured value of the second residual.

13. The rear wheel steering motor abnormality detection method according to claim 9, wherein: The artificial neural network model is composed of a generative adversarial network. The generative adversarial network comprises: a generator receiving the input data and generating the estimated data; and The discriminator receives the input data and the measured data and outputs a measured related discriminant value.

14. The rear wheel steering motor abnormality detection method according to claim 13, wherein: The step of detecting whether the rear wheel steering motor is abnormal comprises the following steps: The processor inputs the input data and the measured data into the discriminator and obtains a measured-related discriminator value generated by the discriminator; and The processor inputs error data into an anomaly detection model and obtains an output of the anomaly detection model, wherein the error data includes the measured related identification value and a value related to a difference between the estimated data and the measured data. 15 . A non-transitory computer-readable storage medium storing a program including at least one instruction for executing the rear-wheel steering motor abnormality detection method according to any one of claims 9 to 12.

Images