Steer-by-wire rack force estimation method and system based on fuzzy neural network

Abstract

The invention provides a steering-by-wire rack force estimation method and system based on a fuzzy neural network, and the method comprises the steps: capturing a long-time dependence relation among a motor torque, a motor angular velocity and a motor rotation angle through a long-short-time memory network, carrying out the fuzzy reasoning through a fuzzy neural network model, and enabling the fuzzy neural network model to obtain a steering-by-wire rack force estimation model through the learning capability of a neural network; and associating the fuzzy value of the input variable with the output fuzzy value of the estimated rack force to complete reasoning of rack force estimation. According to the scheme of the invention, high-precision estimation of the rack force can be realized in a complex non-linear and multi-working-condition environment, and the real-time requirement of a steer-by-wire system is met.

Description

Technical Field

[0001] The present invention belongs to the technical field related to automobile wire-controlled steering, and in particular relates to a wire-controlled steering rack force estimation method and system based on fuzzy neural network. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] The steer-by-wire system is a new type of steering technology that transmits the driver's steering intention through electronic signals and adjusts the vehicle's driving state at the same time. It is widely used in modern intelligent driving cars. However, the steer-by-wire system needs to obtain rack force information in real time to assist in driving feedback and system control. Rack force is a key physical quantity in the steer-by-wire system, which directly affects the steering performance and driving experience of the vehicle. Rack force is the combined force acting on the steering rack, including the input force from the driver's steering wheel, the road reaction force and the output force of the power-assist motor. Its main functions include: Driving feedback: simulate the road feel in the traditional steering system and provide a real driving experience. Vehicle stability: maintain the power balance of the steering system to ensure the stability of the vehicle under different working conditions. Steering power control: assist in determining the size and direction of the motor power assistance to improve steering performance.

[0004] In a steer-by-wire system, due to the lack of mechanical connection, the rack force cannot be directly sensed by the mechanical path and must be acquired in real time through an accurate estimation model.

[0005] At present, the main rack force acquisition methods include:

[0006] 1. Physical modeling-based method: A dynamic model of the vehicle steering system is established to estimate the rack force. However, this method is difficult to accurately model complex nonlinear characteristics, especially in multiple working conditions, and there are significant errors.

[0007] 2. Method based on direct measurement: The rack force is directly measured by a sensor, but its cost is high and the reliability of the sensor is poor in complex environments.

[0008] Rack force is affected by many factors, including driver input, road characteristics, and vehicle motion state. The relationship between them is highly nonlinear and strongly coupled, making it difficult to effectively model using traditional methods. Summary of the invention

[0009] In order to overcome the above-mentioned deficiencies of the prior art, the present invention provides a method and system for estimating rack force of a wire-controlled steering system based on a fuzzy neural network, which can achieve high-precision estimation of rack force in a complex nonlinear and multi-operating environment and meet the real-time requirements of the wire-controlled steering system.

[0010] In order to achieve the above object, the present invention adopts the following technical solution:

[0011] In a first aspect, the present invention provides a method for estimating rack force of a wire-controlled steering system based on a fuzzy neural network, comprising:

[0012] Acquire vehicle motor related data; wherein the vehicle motor related data at least includes motor torque, motor angular velocity and motor rotation angle;

[0013] Using a long short-term memory network to capture the long-term dependency of the vehicle motor related data, and obtain the motor time series characteristics;

[0014] A fuzzy neural network model is used to perform fuzzy reasoning on the motor time series characteristics to obtain a rack force estimation result; wherein the fuzzy neural network model uses the selected membership function and fuzzy rules corresponding to different working conditions, utilizes the learning ability of the neural network, and adjusts the fuzzy value of the output estimated rack force according to the current working condition of the vehicle based on the fuzzy value of the input variable to complete the reasoning of the rack force estimation.

[0015] Preferably, before the vehicle motor related data is processed by using the long short-term memory network, the vehicle motor related data is also preprocessed, and the preprocessing includes:

[0016] Normalizing the motor torque, the motor angular velocity and the motor rotation angle respectively;

[0017] The normalized motor torque, the motor angular velocity and the motor rotation angle are subjected to step sampling processing using a preset sliding window.

[0018] Preferably, in the fuzzy neural network model, a Gaussian function is used as a membership function to convert the input variables of the fuzzy neural network model into fuzzy values.

[0019] Preferably, in the fuzzy neural network model, the activation intensity of the fuzzy rule is:

[0020]

[0021] in, is the activation strength, n is the number of input variables, is the input variable x i For the fuzzy set A i The degree of membership.

[0022] Preferably, in the fuzzy neural network model, the rack force estimation result is obtained, specifically:

[0023]

[0024] Among them, F r is the rack force estimation result, m is the number of rules, is the activation strength of the jth rule, y j is the inferred value of the rack force obtained by the jth rule based on the input.

[0025] Preferably, in the training of the fuzzy neural network model, mean square error is used as the loss function.

[0026] In a second aspect, the present invention provides a steer-by-wire rack force estimation system based on a fuzzy neural network, comprising:

[0027] An acquisition module is configured to: acquire vehicle motor related data; wherein the vehicle motor related data at least includes motor torque, motor angular velocity and motor rotation angle;

[0028] An extraction module is configured to: use a long short-term memory network to capture the long-term dependency of the vehicle motor-related data to obtain motor time series features;

[0029] A fuzzy reasoning module is configured to: use a fuzzy neural network model to perform fuzzy reasoning on the motor time series characteristics to obtain a rack force estimation result; wherein the fuzzy neural network model uses the selected membership function and fuzzy rules corresponding to different working conditions, utilizes the learning ability of the neural network, and adjusts the fuzzy value of the output estimated rack force according to the current working condition of the vehicle based on the fuzzy value of the input variable to complete the reasoning of the rack force estimation.

[0030] In a third aspect, the present invention provides an electronic device comprising a memory and a processor, and computer instructions stored in the memory and executed on the processor, wherein when the computer instructions are executed by the processor, the method described in the first aspect is performed.

[0031] In a fourth aspect, the present invention provides a computer-readable storage medium for storing computer instructions, wherein when the computer instructions are executed by a processor, the method described in the first aspect is performed.

[0032] In a fifth aspect, the present invention provides a computer program product, comprising a computer program, which, when executed by a processor, implements the method described in the first aspect.

[0033] One or more of the above technical solutions have the following beneficial effects:

[0034] In the present invention, a long short-term memory network is used to capture the long-term dependency of the motor torque, the motor angular velocity and the motor rotation angle, and a fuzzy neural network model is used for fuzzy reasoning. The fuzzy neural network model uses the learning ability of the neural network to associate the fuzzy value of the input variable with the fuzzy value of the output estimated rack force to complete the reasoning of the rack force estimation. The scheme of the present invention can achieve high-precision estimation of the rack force in a complex nonlinear and multi-operating environment, and meet the real-time requirements of the wire control steering system.

[0035] Advantages of additional aspects of the present invention will be given in part in the following description, and in part will become obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0037] Figure 1 A schematic diagram of the network structure of a fuzzy neural network in an embodiment of the present invention;

[0038] Figure 2 The flowchart of the rack force estimation method of the wire control steering based on fuzzy neural network in the embodiment of the present invention. DETAILED DESCRIPTION

[0039] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0040] It should be noted that the terms used herein are for describing specific embodiments only and are not intended to be limiting of exemplary embodiments according to the present invention.

[0041] In the absence of conflict, the embodiments of the present invention and the features of the embodiments may be combined with each other.

[0042] As described in the background technology, the steer-by-wire system requires the rack force estimation to have extremely high real-time performance, but the existing complex modeling methods have high computational costs and are difficult to meet the requirements of embedded systems. The vehicle driving environment is complex and changeable, such as high-speed driving, low-speed turning, bumpy roads, etc., which requires the rack force estimation model to have strong generalization and robustness. The computing resources of the embedded controller are limited, and the complexity of the model needs to be optimized. Based on the above problems, this application proposes a steer-by-wire rack force estimation method based on fuzzy neural network, which can effectively overcome the above shortcomings and achieve high-precision and real-time estimation of rack force.

[0043] This embodiment discloses a method for estimating rack force of a wire-controlled steering system based on a fuzzy neural network, comprising:

[0044] Acquire vehicle motor related data; wherein the vehicle motor related data at least includes motor torque, motor angular velocity and motor rotation angle;

[0045] Using a long short-term memory network to capture the long-term dependency of the vehicle motor related data, and obtain the motor time series characteristics;

[0046] A fuzzy neural network model is used to perform fuzzy reasoning on the motor time series characteristics to obtain a rack force estimation result; wherein the fuzzy neural network model uses the selected membership function and fuzzy rules corresponding to different working conditions, utilizes the learning ability of the neural network, and adjusts the fuzzy value of the output estimated rack force according to the current working condition of the vehicle based on the fuzzy value of the input variable to complete the reasoning of the rack force estimation.

[0047] The present invention uses a long short-term memory network to capture the long-term dependency of motor torque, motor angular velocity and motor rotation angle, and uses a fuzzy neural network model for fuzzy reasoning. The fuzzy neural network model uses the learning ability of the neural network to associate the fuzzy value of the input variable with the fuzzy value of the output estimated rack force to complete the reasoning of the rack force estimation. The solution of the present invention can achieve high-precision estimation of the rack force in a complex nonlinear and multi-operating environment, meeting the real-time requirements of the wire control steering system.

[0048] like Figure 2 As shown, a method for estimating rack force of wire-controlled steering based on fuzzy neural network specifically includes:

[0049] Step 21: Obtain and process vehicle motor related data.

[0050] Get the signal of the motor in the wire-controlled steering system: motor torque T m 、Motor angular velocity w m and the motor angle θ m .

[0051] Among them, the motor torque T m , reflects the magnitude of the steering force input by the driver and is a parameter directly related to the rack force;

[0052] Motor angular velocity w m , the steering motor rotation speed reflects the steering frequency of the steering wheel and the road surface information;

[0053] Motor rotation angle θ m , showing the steering wheel angle, which is related to the steering target.

[0054] The above signal is normalized, and the normalization formula is:

[0055]

[0056] Among them, x norm is the normalized value, x min is the minimum value of the signal, x max is the maximum value of the signal, x is the signal (motor torque T m Or motor angular velocity w m Or motor angle θ m ).

[0057] In order to capture the temporal correlation of the signal, the continuous time step signal is used as the input of the long short-term memory network, and a sliding window of fixed length is used to input the data of the past set time step into the long short-term memory network:

[0058]

[0059] Among them, T ,,t 、w m,t ,θ m,t They are the motor torque, angular velocity and rotation angle at time t respectively.

[0060] The signal matrix X after sliding window processing is input into the LSTM network with a dimension of T×F, where T is the time step, which is set to 10 in this embodiment, and F is the number of features in each time step, which is set to 3 in this embodiment.

[0061] The LSTM network is responsible for extracting the time series features of the input signal. Its main structure includes:

[0062] Input Gate:

[0063] i t =σ(W i ·[h t-1 ,x t ]+b i )

[0064] Among them, i t is the input gate output at time t, σ is the Sigmoid function, W i is the input gate weight matrix, h t-1 is the hidden state at time t-1, x t is the input at time t, b i is the input gate bias.

[0065] Forget Gate:

[0066] f t =σ(W f ·[h t-1 ,x t ]+b f )

[0067] Among them, f t is the output of the forget gate at time t, W f is the forget gate weight matrix, b f is the forget gate bias.

[0068] Candidate memory cells:

[0069]

[0070] in, is the candidate memory unit at time t, W C is the candidate memory unit weight matrix, b C is the candidate memory cell bias.

[0071] Current memory unit update:

[0072]

[0073] Among them, C t is the state of the memory unit at time t, C t-1 is the state of the memory unit at time t-1.

[0074] Output gate and hidden state:

[0075] o t =σ(W o ·[h t-1 ,x t ]+b o )

[0076] h t =o t tanh(C t )

[0077] Among them, t is the output of the output gate at time t, W o is the output gate weight matrix, b o is the output gate bias, h t is the hidden state at time t.

[0078] The output layer LSTM outputs the hidden state sequence H = [h t-9 ,h t-8 ,…,h t ], the comprehensive representation of the features of the time series is extracted through the fully connected layer as the input of the fuzzy neural network model.

[0079] like Figure 1As shown in the figure, the fuzzy neural network model includes an input layer, a fuzzification layer, a fuzzy reasoning layer, a normalization layer, and an output layer. The input layer is the starting point of the model and is used to receive the original data of the motor parameters. The fuzzification layer fuzzifies the data of the motor parameters in the input layer and converts the precise input values ​​into fuzzy sets to better handle uncertainty and ambiguity. The fuzzy reasoning layer uses fuzzy rules to reason and calculate based on the fuzzified input to obtain intermediate results. The normalization layer normalizes the results of the fuzzy reasoning layer to keep them within a certain range. Finally, the output layer gives the final result or decision.

[0080] The fuzzy neural network model is responsible for completing the reasoning of the rack force based on the high-dimensional features extracted by LSTM.

[0081] The main steps of the fuzzy neural network model are as follows:

[0082] Fuzzification: The input variable is converted into a fuzzy value through a membership function, and the membership function adopts a Gaussian function:

[0083]

[0084] Among them, μ(x) is the membership degree, x is the input variable, c is the center of the Gaussian function, and σ is the standard deviation.

[0085] Rule reasoning: The activation strength of fuzzy rules is:

[0086]

[0087] in, is the activation strength, n is the number of input variables, is the input variable x i For the fuzzy set A i The degree of membership.

[0088] The setting of fuzzy rules needs to be based on the intrinsic relationship between motor torque, motor angular velocity, motor rotation angle and rack force, and combined with the actual operating conditions of the automotive steer-by-wire system, such as:

[0089] Normal steering conditions:

[0090] Rule 1: If the motor torque is small, it means the driver inputs less steering force, the motor angular velocity is low, which means the steering wheel turns slowly, and the motor angle is in a small range, which means the steering amplitude is small, then the rack force is small. In this case, when the vehicle is driving in a straight line or making small adjustments, the driver only needs to apply less force, and the rack force required by the system is also small.

[0091] Rule 2: If the motor torque is moderate, the motor angular velocity is moderate, and the motor angle is in the medium range, then the rack force is moderate. This is consistent with the scenario where the driver applies appropriate force during normal turning and the system generates a corresponding moderate rack force to achieve steering.

[0092] Emergency steering conditions:

[0093] Rule 3: When the motor torque is large, it means the driver is making a strong steering operation for emergency avoidance, etc., the motor angular velocity is high, which means the steering wheel is turning quickly, and the motor angle is large, which means the steering amplitude is large, then the rack force is large. In an emergency, the driver quickly turns the steering wheel and the system needs a large rack force to quickly change the direction of the vehicle.

[0094] Special road conditions:

[0095] Rule 4: If the motor torque fluctuates greatly, it means that the bumpy road will make the motor torque unstable. If the motor angular velocity fluctuates greatly, it means that the steering wheel is shaken by the bumpy road. If the motor angle fluctuates greatly, then the rack force fluctuates greatly. When driving on a bumpy road, the change in the road reaction force causes the motor parameters to fluctuate, which in turn causes the rack force to fluctuate.

[0096] Rule 5: If the motor torque continues to increase within a specific range, such as when the vehicle is turning on a hill, a greater steering force is required, the motor angular velocity is relatively stable, that is, a certain steering frequency is maintained, and the motor angle gradually increases, indicating continuous steering, then the rack force gradually increases. This applies to the situation where the rack force changes according to actual needs when the vehicle is turning under special road conditions.

[0097] Fuzzy rules are the basis for calculating the activation strength of fuzzy rules. In actual calculations, the fuzzy values ​​of the fuzzified input variables, namely motor torque, motor angular velocity, and motor rotation angle, are substituted into the corresponding rules after being transformed by the membership function. Taking the rule "If the motor torque is large, the motor angular velocity is high, and the motor rotation angle is large, then the rack force is large" as an example, the membership values ​​corresponding to the motor torque, motor angular velocity, and motor rotation angle are calculated according to the rule reasoning formula. Calculate the activation strength. Through such calculation, we can get the activation strength of the rule under the current input situation. This value reflects the applicability of this rule under the current working conditions.

[0098] The setting of fuzzy rules enables the fuzzy neural network model to better adapt to complex and changeable driving conditions. Different driving scenarios such as normal driving, emergency steering, and driving on bumpy roads correspond to different fuzzy rule combinations. When the vehicle is in different working conditions, the corresponding rules are activated, allowing the model to adjust the estimation of the rack force according to the actual situation.

[0099] Reasoning result calculation: Calculate the output by weighted average method:

[0100]

[0101] Among them, F r is the rack force estimation result, m is the number of rules, is the activation strength of the jth rule, y j is the inferred value of the rack force obtained by the jth rule based on the input.

[0102] Training of the fuzzy neural network model: 1000 sets of actual measurement data under different working conditions are selected as training samples, of which 800 sets are used for training and 200 sets are used for verification. Initialize the weights and fuzzy rule parameters of the fuzzy neural network, set the initial learning rate to 0.1, and the momentum term coefficient to 0.9. During the training process, the learning rate is automatically adjusted according to the training error. When the error drops less than 0.001 for 5 consecutive iterations, the learning rate is halved. After 5000 iterations of training, the network reaches the preset training accuracy, and the mean square error is less than 0.01. Use the known motor signal and rack force data to train the model, and use the mean square error (MSE) as the loss function:

[0103]

[0104] Among them, F r is the real rack force value, is the rack force value predicted by the fuzzy neural network model, and N is the number of samples

[0105] The parameters of the fuzzy neural network model are updated using the Adam optimization algorithm:

[0106]

[0107] Among them, θ t is the fuzzy neural network model parameter at time t, α is the learning rate, It is the gradient of the loss function with respect to the parameters of the fuzzy neural network model at time t.

[0108] This embodiment also provides a steer-by-wire rack force estimation system based on a fuzzy neural network, comprising:

[0109] An acquisition module is configured to: acquire vehicle motor related data; wherein the vehicle motor related data at least includes motor torque, motor angular velocity and motor rotation angle;

[0110] An extraction module is configured to: use a long short-term memory network to capture the long-term dependency of the vehicle motor-related data to obtain motor time series features;

[0111] A fuzzy reasoning module is configured to: use a fuzzy neural network model to perform fuzzy reasoning on the motor time series characteristics to obtain a rack force estimation result; wherein the fuzzy neural network model uses the learning ability of the neural network to associate the fuzzy value of the input variable with the fuzzy value of the output estimated rack force to complete the reasoning of the rack force estimation.

[0112] In further embodiments, there is also provided:

[0113] An electronic device includes a memory and a processor, and computer instructions stored in the memory and executed on the processor, wherein when the computer instructions are executed by the processor, the method described in Embodiment 1 is performed. For the sake of brevity, no further description is given here.

[0114] It should be understood that in this embodiment, the processor may be a central processing unit CPU, and the processor may also be other general-purpose processors, digital signal processors DSP, application-specific integrated circuits ASIC, off-the-shelf programmable gate arrays FPGA or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0115] The memory may include a read-only memory and a random access memory, and provide instructions and data to the processor. A portion of the memory may also include a non-volatile random access memory. For example, the memory may also store information about the device type.

[0116] A computer-readable storage medium is used to store computer instructions. When the computer instructions are executed by a processor, the method described in embodiment 1 is completed.

[0117] The method in the first embodiment can be directly embodied as a hardware processor, or a combination of hardware and software modules in the processor. The software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware. To avoid repetition, it will not be described in detail here.

[0118] A computer program product includes a computer program, and when the computer program is executed by a processor, the method described in the first embodiment is implemented.

[0119] The present invention also provides at least one computer program product tangibly stored on a non-transitory computer-readable storage medium. The computer program product includes computer executable instructions, such as instructions included in a program module, which are executed in a device on a real or virtual processor of the target to perform the process / method as described above. Typically, program modules include routines, programs, libraries, objects, classes, components, data structures, etc. that perform specific tasks or implement specific abstract data types. In various embodiments, the functions of program modules can be combined or divided between program modules as needed. Machine executable instructions for program modules can be executed in local or distributed devices. In distributed devices, program modules can be located in local and remote storage media.

[0120] The computer program code for implementing the method of the present invention can be written in one or more programming languages. These computer program codes can be provided to the processor of a general-purpose computer, a special-purpose computer or other programmable data processing device, so that the program code, when executed by the computer or other programmable data processing device, causes the function / operation specified in the flow chart and / or block diagram to be implemented. The program code can be executed completely on a computer, partially on a computer, as an independent software package, partially on a computer and partially on a remote computer or completely on a remote computer or server.

[0121] In the context of the present invention, computer program codes or related data may be carried by any appropriate carrier to enable a device, apparatus or processor to perform the various processes and operations described above. Examples of carriers include signals, computer readable media, etc. Examples of signals may include electrical, optical, radio, acoustic or other forms of propagation signals, such as carrier waves, infrared signals, etc.

[0122] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with this embodiment can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0123] Although the above describes the specific implementation mode of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without creative work are still within the scope of protection of the present invention.

Claims

1. A method for estimating rack force of wire control steering based on fuzzy neural network, characterized in that: include: Acquire vehicle motor related data; wherein the vehicle motor related data at least includes motor torque, motor angular velocity and motor rotation angle; Using a long short-term memory network to capture the long-term dependency of the vehicle motor related data, and obtain the motor time series characteristics; A fuzzy neural network model is used to perform fuzzy reasoning on the motor time series characteristics to obtain a rack force estimation result; wherein the fuzzy neural network model uses the selected membership function and fuzzy rules corresponding to different working conditions, utilizes the learning ability of the neural network, and adjusts the fuzzy value of the output estimated rack force according to the current working condition of the vehicle based on the fuzzy value of the input variable to complete the reasoning of the rack force estimation.

2. A fuzzy neural network based steering-by-wire rack force estimation method as claimed in claim 1, characterized in that: Before the vehicle motor related data is processed by the long short-term memory network, the vehicle motor related data is preprocessed, and the preprocessing includes: Normalizing the motor torque, the motor angular velocity and the motor rotation angle respectively; The normalized motor torque, the motor angular velocity and the motor rotation angle are subjected to step sampling processing using a preset sliding window.

3. The method for estimating rack force of a steer-by-wire system based on a fuzzy neural network as claimed in claim 1, characterized in that: In the fuzzy neural network model, a Gaussian function is used as a membership function to convert the input variables of the fuzzy neural network model into fuzzy values.

4. The method for estimating rack force of a steer-by-wire system based on a fuzzy neural network as claimed in claim 1, characterized in that: In the fuzzy neural network model, the activation intensity of the fuzzy rule is: in, is the activation strength, n is the number of input variables, is the input variable x i For the fuzzy set A i The degree of membership.

5. A method for estimating rack force of a steer-by-wire system based on a fuzzy neural network as claimed in any one of claims 1 to 4, characterized in that: In the fuzzy neural network model, the rack force estimation result is obtained, specifically: Among them, F r is the rack force estimation result, m is the number of rules, is the activation strength of the jth rule, y j is the inferred value of the rack force obtained by the jth rule based on the input.

6. A method for estimating rack force of a steer-by-wire system based on a fuzzy neural network as claimed in any one of claims 1, characterized in that: In the training of the fuzzy neural network model, mean square error is used as the loss function.

7. A steer-by-wire rack force estimation system based on fuzzy neural network, characterized in that: include: An acquisition module is configured to: acquire vehicle motor related data; wherein the vehicle motor related data at least includes motor torque, motor angular velocity and motor rotation angle; An extraction module is configured to: use a long short-term memory network to capture the long-term dependency of the vehicle motor-related data to obtain motor time series features; A fuzzy reasoning module is configured to: use a fuzzy neural network model to perform fuzzy reasoning on the motor time series characteristics to obtain a rack force estimation result; wherein the fuzzy neural network model uses the selected membership function and fuzzy rules corresponding to different working conditions, utilizes the learning ability of the neural network, and adjusts the fuzzy value of the output estimated rack force according to the current working condition of the vehicle based on the fuzzy value of the input variable to complete the reasoning of the rack force estimation.

8. An electronic device, characterized in that: The method comprises a memory and a processor and computer instructions stored in the memory and executed on the processor, wherein when the computer instructions are executed by the processor, the method according to any one of claims 1 to 6 is completed.

9. A computer-readable storage medium, characterized in that: Used to store computer instructions, which, when executed by a processor, complete the method described in any one of claims 1 to 6.

10. A computer program product, characterized in that The invention comprises a computer program, which, when executed by a processor, implements the method according to any one of claims 1 to 6.

Images