Vehicle PID control method and device, computer equipment and storage medium

Abstract

The invention provides a vehicle PID control method and device, computer equipment and a storage medium. The method comprises the following steps: if a target vehicle runs under a crawling working condition, acquiring current crawling working condition data of the target vehicle; inputting the current creeping working condition data into a back propagation neural network model to obtain a target creeping PID parameter output by the back propagation neural network model; and inputting the target creeping PID parameter into a vehicle PID controller to obtain target creeping control information of the target vehicle. Therefore, the PID parameters are enabled to better fit the actual crawling working condition of the target vehicle while the difficulty in setting the PID parameters of the vehicle PID controller can be reduced, and the drivability of the target vehicle running under the crawling working condition is improved.

Description

Technical Field

[0001] This application relates to the field of autonomous driving technology, and particularly relates to a vehicle PID control method, device, computer device, and storage medium. Background Art

[0002] With the development of technology, people's dependence on vehicles is increasing. At the same time, vehicles often drive under different driving conditions and pass through various road conditions during driving, which also puts higher requirements on the drivability and comfort of vehicles.

[0003] In related technologies, the vehicle is usually controlled by means of look-up table control or PID control. However, the applicable driving conditions of the look-up table control method or the PID control method are limited. Especially when the vehicle is driving in a creeping condition on roads with various road conditions, the drivability of the vehicle needs to be improved.

[0004] Therefore, there is an urgent need to provide a vehicle PID control method to improve the drivability of the vehicle when operating in a creeping condition. Summary of the Invention

[0005] In view of this, this application is committed to providing a vehicle PID control method, device, computer device, and storage medium to improve the drivability of the vehicle when operating in a creeping condition.

[0006] This application provides a vehicle PID control method, which includes: if the target vehicle is operating in a creeping condition, obtaining the current creeping condition data of the target vehicle; inputting the current creeping condition data into a backpropagation neural network model to obtain the target creeping PID parameters output by the backpropagation neural network model; and inputting the target creeping PID parameters into a vehicle PID controller to obtain the target creeping control information of the target vehicle.

[0007] In the above embodiment, on the one hand, when it is detected that the target vehicle is operating in a creeping condition, by inputting the current creeping condition data of the target vehicle into the backpropagation neural network model, the target creeping PID parameters output by the backpropagation neural network model are obtained. In this way, while reducing the tuning difficulty of the PID parameters of the vehicle PID controller, the PID parameters can be made more suitable for the actual creeping condition of the target vehicle; on the other hand, inputting the target creeping PID parameters that fit the actual creeping condition into the vehicle PID controller makes the target creeping control information output by the vehicle PID controller also more suitable for the actual creeping condition of the target vehicle. Furthermore, when the target vehicle is controlled according to the target creeping control information, the drivability of the target vehicle when operating in a creeping condition can be improved.

[0008] In some embodiments, the current crawling condition data can represent the vehicle state when controlling the target vehicle based on the current crawling control information; the current crawling control information is determined by the vehicle PID controller based on the current crawling PID parameters; if the current crawling PID parameters are initial PID parameters, the initial PID parameters are obtained by processing the initially simulated crawling condition data by the backpropagation neural network model.

[0009] In the above embodiments, the initial crawling condition data is generated by simulation, and the initial PID parameters are obtained by processing the initial crawling condition data by the backpropagation neural network model. Then, the initial PID parameter set is used as the current crawling PID parameters and input into the vehicle PID controller to obtain the current crawling control information for controlling the target vehicle. Further, the current crawling condition data of the target vehicle under the control of the current crawling control information can be obtained. In this way, a PID control basis can be provided for the subsequent operation of the target vehicle in the crawling condition.

[0010] In some embodiments, the current crawling condition data includes at least one of vehicle speed, acceleration, brake opening, motor speed, motor torque, maximum motor crawling torque, pitch slope, vehicle speed difference, and road surface adhesion coefficient; wherein, the vehicle speed difference is the difference between the vehicle speed and the maximum crawling vehicle speed; the initial crawling condition data includes at least one of vehicle speed, acceleration, brake opening, motor speed, motor torque, maximum motor crawling torque, pitch slope, vehicle speed difference, and road surface adhesion coefficient.

[0011] In the above embodiments, the initial crawling condition data or the current crawling condition data including at least one of vehicle speed, acceleration, brake opening, motor speed, motor torque, maximum motor crawling torque, pitch slope, vehicle speed difference, and road surface adhesion coefficient enables the accurate construction of crawling condition data conforming to the target vehicle through the initial crawling condition data, and enables the accurate reflection of the current state of the target vehicle running in the crawling condition through the current crawling condition data. Thus, the PID parameters determined based on the initial crawling condition data or the current crawling condition data can be more suitable for the crawling condition. Furthermore, when controlling the target vehicle through the crawling control information determined based on the PID parameters, the drivability of the target vehicle running in the crawling condition can be effectively improved.

[0012] In some embodiments, before inputting the current creep condition data into the backpropagation neural network model, the method further includes: receiving the model parameters of the backpropagation neural network model sent by the cloud platform; wherein the model parameters are obtained by training the backpropagation neural network model based on the creep condition data of at least one vehicle collected by the cloud platform; the at least one vehicle includes or does not include the target vehicle; determining the backpropagation neural network model based on the received model parameters.

[0013] In the above embodiments, by training the backpropagation neural network model based on the creep condition data of the target vehicle or non-target vehicle collected by the cloud platform, the model parameters of the backpropagation neural network model are determined. In this way, a trained backpropagation neural network model for tuning the creep PID parameters of the vehicle PID controller can be obtained, and the model parameters are sent to the target vehicle, so that the target vehicle can tune the creep PID parameters of the vehicle PID controller in real time based on the trained backpropagation neural network model.

[0014] In some embodiments, the method further includes: controlling the target vehicle based on the target creep control information of the target vehicle to obtain the new creep condition data of the target vehicle; uploading the new creep condition data to the cloud platform, so that a data sample can be constructed based on the new creep condition data to train the backpropagation neural network model.

[0015] In the above embodiments, by controlling the target vehicle based on the target creep control information of the target vehicle and uploading the obtained new creep condition data of the target vehicle to the cloud platform, a data sample can be constructed based on the new creep condition data of the target vehicle to train the backpropagation neural network model, and subsequently, the model parameters of the trained backpropagation neural network model sent by the cloud platform can be received to obtain an updated backpropagation neural network model. In this way, data support can be continuously provided for the continuous update and optimization of the backpropagation neural network model through the way of reverse feedback.

[0016] In some embodiments, the backpropagation neural network model is trained in the following manner, including: constructing a data sample based on historical creep condition data, wherein the data sample includes at least one of vehicle speed, acceleration, brake opening, motor speed, motor torque, maximum motor creep torque, pitch slope, vehicle speed difference, road surface adhesion coefficient; inputting the data sample into the backpropagation neural network model to be trained for forward calculation to obtain model output data; performing a backward error calculation based on the model output data and the expected output data of the data sample to update the model parameters of the backpropagation neural network model to be trained.

[0017] In the above embodiments, data samples are constructed based on historical creep condition data that can accurately reflect the state of the corresponding vehicle operating in the creep condition, and the data samples are input into the backpropagation neural network model to be trained for forward calculation to obtain model output data. Based on the model output data and the expected output data of the data samples, backward error calculation is performed to update the model parameters of the backpropagation neural network model to be trained. In this way, the trained backpropagation neural network model can set PID parameters that are more suitable for the creep condition, and further improve the drivability of the vehicle when controlling the vehicle operating in the creep condition subsequently.

[0018] In some embodiments, the method further includes: obtaining historical vehicle operation data; and performing data screening on the historical vehicle operation data according to creep condition detection conditions to obtain the historical creep condition data.

[0019] In the above embodiments, by performing data screening on the obtained historical vehicle operation data according to creep condition detection conditions, invalid data in the historical vehicle operation data can be cleaned, and real historical creep condition data related to the creep condition can be obtained. In this way, the interference of invalid data on the training process of the backpropagation neural network model can be reduced, the training effect of the backpropagation neural network model can be improved, the creep PID parameters set by the backpropagation neural network model can be more suitable for the actual creep condition, and further the drivability of the vehicle operating in the creep condition can be improved.

[0020] In some embodiments, the following method is used to determine that the target vehicle is operating in the creep condition: obtaining data required for creep judgment of the target vehicle; wherein, the data required for creep judgment at least includes creep enable state data, vehicle speed, and fault data; if the data required for creep judgment meets the creep condition detection conditions, it is determined that the target vehicle is operating in the creep condition.

[0021] In the above embodiments, by obtaining data required for creep judgment that at least includes creep enable state data, vehicle speed, and fault data, and judging whether the data required for creep judgment meets the creep condition detection conditions, and when the state data, vehicle speed, and fault data in the data required for creep judgment meet the creep condition detection conditions, it is determined that the target vehicle is operating in the creep condition, and the current creep condition data of the target vehicle operating in the creep condition is input into the backpropagation neural network, so that creep PID parameters that are suitable for the creep condition of the target vehicle can be set in a timely manner, and further the drivability of the target vehicle operating in the creep condition can be improved.

[0022] This application provides a vehicle PID control device, which includes: a data acquisition module for acquiring the current creep condition data of the target vehicle if the target vehicle is operating in the creep condition; a PID parameter determination module for inputting the current creep condition data into a backpropagation neural network model to obtain the target creep PID parameters output by the backpropagation neural network model; and a vehicle control module for inputting the target creep PID parameters into a vehicle PID controller to obtain the target creep control information of the target vehicle.

[0023] This application provides a computer device, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the PID control method described in any of the above embodiments.

[0024] This application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the PID control method described in any of the above embodiments.

[0025] This application provides a vehicle, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the PID control method described in any of the above embodiments.

[0026] The above description is only an overview of the technical solution of this application. In order to be able to understand the technical means of this application more clearly, it can be implemented according to the content of this application. And in order to make the above and other purposes, features and advantages of this application more obvious and understandable, the specific embodiments of this application are specifically listed below. Description of the Drawings

[0027] By reading the detailed description of the following alternative embodiments, various other advantages and benefits will become clear to those of ordinary skill in the art. The drawings are only for the purpose of showing the alternative embodiments and are not considered to be a limitation of this application. And in all the drawings, the same reference numerals are used to represent the same components. In the drawings:

[0028] Figure 1 It is a schematic diagram of the application environment of the PID control method provided by the embodiment of this application;

[0029] Figure 2 It is a schematic flowchart of the PID control method provided by the embodiment of this application;

[0030] Figure 3 It is a schematic flowchart of the training method of the BP neural network model provided by the embodiment of this application;

[0031] Figure 4Schematic flowchart of the method for determining the operating condition of the target vehicle running in the creep mode provided by the embodiment of the present application;

[0032] Figure 5 Schematic diagram of the PID control device provided by the embodiment of the present application;

[0033] Figure 6 Schematic diagram of the computer device provided by the embodiment of the present application. Detailed implementation manners

[0034] The embodiments of the technical solution of the present application will be described in detail below with reference to the accompanying drawings. The following embodiments are only used to illustrate the technical solution of the present application more clearly, and therefore are only examples and cannot be used to limit the protection scope of the present application.

[0035] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present application belongs; the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit the present application; the terms "including" and "having" and any variations thereof in the embodiments of the present application and the claims and the above description of the drawings are intended to cover non-exclusive inclusion.

[0036] In the description of the embodiments of the present application, technical terms such as "first" and "second" are only used to distinguish different objects and cannot be understood as indicating or implying relative importance or implicitly indicating the quantity, specific order or primary-secondary relationship of the indicated technical features. In the description of the embodiments of the present application, the meaning of "a plurality" is two or more unless otherwise specifically defined.

[0037] Referring to "embodiments" herein means that specific features, structures or characteristics described in connection with the embodiments can be included in at least one embodiment. The phrase appears in various places in the present application does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art explicitly and implicitly understand that the embodiments described herein can be combined with other embodiments.

[0038] In the description of the embodiments of the present application, the term "and / or" is only a description of the association relationship of associated objects, indicating that there can be three relationships, for example, A and / or B, which can mean: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this article generally represents an "or" relationship between the front and rear associated objects.

[0039] With the development of technology, people's dependence on vehicles is increasing. During vehicle driving, vehicles often operate under different driving conditions and pass through various road conditions, which also puts higher requirements on vehicle performance. For example, vehicle drivability, handling, and comfort. Vehicle drivability, that is, the driveability of the vehicle, can be used to evaluate the power performance and braking performance of the vehicle, or rather, can be used to evaluate the longitudinal X-axis movement or straight-line driving performance of the vehicle. Handling can be used to evaluate the vehicle's ability to handle curves, or rather, can be used to evaluate the yaw movement performance of the vehicle's lateral Y-axis. And comfort can be used to evaluate the performance of the vehicle's vertical Z-axis.

[0040] In related technologies, usually, a look-up table is used for torque control to control the vehicle, or a Proportional Integral Derivative (PID) control method is used to control the vehicle. However, the applicable driving conditions of the look-up table control method or the PID control method are limited. Especially when the vehicle is driving in a creeping condition, or rather, when the vehicle is cruising at a low speed, it often drives on various complex road conditions, such as slopes, rain and snow roads, rock roads, sandy roads, etc., or rather, low-adhesion roads. It is difficult to drive through the look-up table control method. And the PID control method uses proportional, integral, and differential coefficients to adjust the response of the vehicle PID controller to control the vehicle. For a complex control object like a vehicle operating in a creeping condition with strong nonlinearity, time-varying uncertainty, and external interference, it is difficult for the PID control method to set appropriate PID parameters.

[0041] Therefore, it is necessary to provide a vehicle PID control method. First, detect whether the target vehicle is operating in a creeping condition. If the target vehicle is operating in a creeping condition, obtain the current creeping condition data of the target vehicle. Then, input the current creeping condition data into a Back Propagation Neural Network model to obtain the target creeping PID parameters output by the BP neural network model. Next, input the obtained target creeping PID parameters into the vehicle PID controller to obtain the target creeping control information output by the vehicle PID control, so as to enable the control of the target vehicle operating in the creeping condition based on the target creeping control information.

[0042] Please refer to Figure 1 , Figure 1 which is a schematic diagram of the application environment of the PID control method provided in this scenario example. This PID control method is applied to the target vehicle 110, or applied to other devices with the function of controlling the vehicle.

[0043] In this scenario example, the target vehicle 110 can be a fuel vehicle, a gas vehicle, or a new energy vehicle. The new energy vehicle can be a pure electric vehicle, a hybrid vehicle, an extended-range vehicle, etc. A vehicle controller is provided inside the target vehicle 110, and the vehicle controller is configured with a vehicle PID controller algorithm. Exemplarily, the vehicle controller can be a Vehicle Control Unit (VCU). The target vehicle 110 can communicate with the cloud platform 120.

[0044] In this scenario example, the cloud platform 120 can collect creep condition data, which is the historical creep condition data of the target vehicle 110 or other vehicles. The cloud platform 120 can train the BP neural network model to be trained based on the historical creep condition data, obtain the trained BP neural network model, and can send the model parameters of the trained BP neural network model to the target vehicle 110.

[0045] In this scenario example, the target vehicle 110 can receive the model parameters sent by the cloud platform, thereby obtaining the trained BP neural network model, and can input the current creep condition data into the BP neural network model under the creep condition to obtain the target creep PID parameters, and then input the target creep PID parameters into the vehicle PID controller to obtain the target creep control information for controlling the target vehicle 110, thereby controlling the target vehicle 110. And the new creep condition data obtained by controlling the target vehicle 110 based on the target creep control information can be uploaded to the cloud platform 120.

[0046] In this scenario embodiment, the cloud platform 120 can receive the new creep condition data uploaded by the target vehicle 110, use the new creep condition data of the target vehicle 110 as historical creep condition data to construct a data sample, update and optimize the BP neural network model, and send the model parameters of the updated and optimized BP neural network model to the target vehicle 110 again.

[0047] The implementation mode of this application provides a vehicle PID control method. Please refer to Figure 2 , Figure 2 is a schematic flowchart of a vehicle PID control method provided by this implementation mode. This implementation mode provides method operation steps as shown in the flowchart, but based on routine or non-creative labor, it can include more or fewer operation steps. The step order listed in the implementation mode is only one execution mode among the execution orders of numerous steps, and does not represent the only execution order. When the actual system or server product executes, it can be executed in the method order shown in the embodiment or executed in parallel (for example, in an environment of parallel processors or multi-threaded processing). Specifically, as Figure 2As shown, the PID control method may include the following steps.

[0048] Step S210: If the target vehicle is operating in a creeping condition, obtain the current creeping condition data of the target vehicle.

[0049] In some cases, when it is detected that the target vehicle is operating in a creeping condition, vehicle control adapted to the creeping condition in which the target vehicle is operating can be performed on the target vehicle, thereby improving the drivability of the target vehicle when it is operating in the creeping condition.

[0050] Among them, the current creeping condition data can describe the vehicle state of the target vehicle when it is operating in the creeping condition at the current moment.

[0051] Step S220: Input the current creeping condition data into the backpropagation neural network model to obtain the target creeping PID parameters output by the backpropagation neural network model.

[0052] Among them, the backpropagation neural network model, or the BP neural network model for short, is a trained model that can output the target creeping PID parameters and can realize the tuning of the PID parameters required by the vehicle PID controller.

[0053] Specifically, the target creeping PID parameters may include a proportional regulation parameter (Proportional), an integral regulation parameter (Integral), and a derivative regulation parameter (Derivative).

[0054] Step S230: Input the target creeping PID parameters into the vehicle PID controller to obtain the target creeping control information of the target vehicle.

[0055] Specifically, the vehicle PID controller may be configured in the vehicle controller of the target vehicle. Exemplarily, the vehicle PID controller may be a PID controller algorithm configured in the vehicle control unit VCU. The vehicle PID controller can output the target creeping control information based on the input target creeping PID parameters, so that the target vehicle operating in the creeping condition can be controlled based on the target creeping control information, or in other words, the target vehicle can be controlled to operate in the creeping condition.

[0056] In the above embodiments, on the one hand, when it is detected that the target vehicle is operating in a creep condition, by inputting the current creep condition data of the target vehicle into the BP neural network model, the target creep PID parameters output by the BP neural network model are obtained. In this way, while reducing the tuning difficulty of the PID parameters of the vehicle PID controller, the PID parameters can be made more suitable for the actual creep condition of the target vehicle. On the other hand, by inputting the target creep PID parameters that fit the actual creep condition into the vehicle PID controller, the target creep control information output by the vehicle PID controller is also more in line with the actual creep condition of the target vehicle. Furthermore, when controlling the target vehicle according to the target creep control information, the drivability of the target vehicle operating in the creep condition can be improved.

[0057] In some embodiments, the current creep condition data can represent the vehicle state when controlling the target vehicle based on the current creep control information. Among them, the current creep control information is determined by the vehicle PID controller based on the current creep PID parameters.

[0058] In this embodiment, if the current creep PID parameter is the initial PID parameter, the initial PID parameter is obtained by the backpropagation neural network model processing the initially simulated creep condition data. In other words, before the target vehicle operates in the creep condition, the initial creep condition data can be generated by simulation, and the initial PID parameter can be obtained by the BP neural network model processing the initial creep condition data, and this initial PID parameter can be used as the current creep PID parameter to determine the current creep control information.

[0059] Among them, the initial PID parameter can include a proportional regulation parameter, an integral regulation parameter, and a derivative regulation parameter.

[0060] In the above embodiments, the initial creep condition data is generated by simulation, and the initial PID parameter is obtained by the BP neural network model processing the initial creep condition data. Then, the initial PID parameter is used as the current creep PID parameter and input into the vehicle PID controller to obtain the current creep control information for controlling the target vehicle. Furthermore, the current creep condition data of the target vehicle under the control of the current creep control information can be obtained. In this way, a creep condition data basis and a creep PID parameter basis can be provided for the subsequent operation of the target vehicle in the creep condition.

[0061] In some embodiments, the current creep condition data can include at least one of vehicle speed, acceleration, brake opening, motor speed, motor torque, maximum motor creep torque, pitch slope, vehicle speed difference, and road surface adhesion coefficient.

[0062] Among them, the vehicle speed difference is the difference between the vehicle speed and the maximum creep vehicle speed. The maximum creep torque of the motor is the maximum motor torque when the vehicle is operating in the creep condition.

[0063] Correspondingly, the initial creep condition data may include at least one of vehicle speed, acceleration, brake opening, motor speed, motor torque, maximum creep torque of the motor, pitch slope, vehicle speed difference, and road surface adhesion coefficient.

[0064] In the above embodiments, the initial creep condition data or the current creep condition data including at least one of vehicle speed, acceleration, brake opening, motor speed, motor torque, maximum creep torque of the motor, pitch slope, vehicle speed difference, and road surface adhesion coefficient enables the accurate construction of creep condition data conforming to the target vehicle through the initial creep condition data, and enables the accurate reflection of the current state of the target vehicle operating in the creep condition through the current creep condition data. Thus, the PID parameters determined based on the initial creep condition data or the current creep condition data can be more suitable for the creep condition. Furthermore, when controlling the target vehicle through the creep control information determined based on the PID parameters, the drivability of the target vehicle operating in the creep condition can be effectively improved.

[0065] In some embodiments, before inputting the current creep condition data into the BP neural network model, the PID control method may further include the following steps: receiving the model parameters of the BP neural network model sent by the cloud platform; determining the BP neural network model based on the received model parameters; wherein, the model parameters are obtained by training the BP neural network model based on the creep condition data of at least one vehicle collected by the cloud platform.

[0066] Among them, the at least one vehicle may or may not include the target vehicle. The creep condition data of the at least one vehicle may be the historical creep condition data of the target vehicle or the historical creep condition data of non-target vehicles.

[0067] Specifically, the cloud platform may train the BP neural network model based on the historical creep condition data and send the model parameters of the trained BP neural network model to the target vehicle. The target vehicle may receive the model parameters sent by the cloud platform and obtain the trained BP neural network model based on the received model parameters, or update the local BP neural network model based on the received model parameters to obtain an updated or optimized BP neural network model.

[0068] Exemplarily, the model parameters of the BP neural network model sent by the cloud platform may be received by using the Over The Air technology (OTA).

[0069] In the above embodiments, the BP neural network model is trained based on the creep condition data of the target vehicle or non-target vehicle collected by the cloud platform to determine the model parameters of the BP neural network model. In this way, a trained BP neural network model for tuning the creep PID parameters of the vehicle PID controller can be obtained, and the model parameters are sent to the target vehicle, so that the target vehicle can tune the creep PID parameters of the vehicle PID controller in real time based on the trained or updated and optimized BP neural network model.

[0070] In some embodiments, the PID control method may further include the following steps: controlling the target vehicle based on the target creep control information of the target vehicle to obtain the new creep condition data of the target vehicle; uploading the new creep condition data to the cloud platform so that a data sample can be constructed based on the new creep condition data to train the backpropagation neural network model.

[0071] In the above embodiments, the target vehicle is controlled based on the target creep control information of the target vehicle, and the obtained new creep condition data of the target vehicle is uploaded to the cloud platform so that a data sample can be constructed based on the new creep condition data of the target vehicle to train the backpropagation neural network model, and subsequently, the model parameters of the trained backpropagation neural network model can be received from the cloud platform to obtain an updated backpropagation neural network model. In this way, data support can be continuously provided for the continuous update and optimization of the backpropagation neural network model through the reverse feedback method. Furthermore, when the target vehicle operates in the creep condition, the creep PID parameters of the vehicle PID controller can be tuned in real time based on the actual creep condition data, the matching degree between the creep PID parameters and the actual creep condition can be improved, that is to say, the ability of the target vehicle to cope with complex road conditions or road condition changes when operating in the creep condition can be improved.

[0072] In some embodiments, please refer to Figure 3 , training the BP neural network model may include the following steps.

[0073] Step S310: Construct a data sample based on the historical creep condition data, where the data sample includes at least one of vehicle speed, acceleration, brake opening, motor speed, motor torque, maximum creep torque of the motor, pitch slope, vehicle speed difference, and road surface adhesion coefficient.

[0074] In some cases, a BP neural network model can be composed of an input layer (Input), a hidden layer (Hidding), and an output layer (Output). Among them, the number of hidden layers can be at least 1 layer. The input layer, the hidden layer, and the output layer each have corresponding neurons and the number of neurons. Between the input layer and the hidden layer, between the previous hidden layer and the next hidden layer, and between the hidden layer and the output layer, there are corresponding connection weights or weight coefficients respectively.

[0075] In some other cases, the training of the BP neural network model, or rather, the learning process of the BP neural network model, can be composed of forward propagation and backpropagation of error. During forward propagation, data samples are input from the input layer, processed layer by layer through each hidden layer, or transmitted to the output layer through the weighted effect of the connection weights between layers. If the model output data output by the output layer does not match the expected output data or does not meet the set requirements, then the error between the model output data and the expected output data will be used as an adjustment signal to propagate backward layer by layer to adjust the connection weights in the BP neural network model to reduce the error until the error is reduced to an acceptable range.

[0076] In this embodiment, the initial model parameters of the BP neural network model can be initialized first. Specifically, for example, the number of hidden layers of the BP neural network model can be set, the number of neurons in the input layer, the output layer, and each hidden layer can be set, and the connection weights can be set, etc. Among them, the number of neurons in the output layer is 3, corresponding to the creeping PID parameters including the proportional adjustment parameter, the integral adjustment parameter, and the derivative adjustment parameter output by the BP neural network model.

[0077] Exemplarily, the number of iterations and the learning rate can also be set.

[0078] Exemplarily, an activation function can also be set. The activation function can be used for forward propagation with the connection weights, or rather, can be used for forward calculation with the connection weights. As an example, the activation function can be the hyperbolic tangent function. As an example, the activation function can be the non-negative hyperbolic tangent function.

[0079] Exemplarily, based on the constructed data samples, the number of neurons in the input layer can be determined.

[0080] Exemplarily, based on the number of neurons in the input layer set during the initialization of the BP neural network model, data samples can also be constructed.

[0081] Exemplarily, the expected output data of the data samples can also be set.

[0082] Step S320: Input the data samples into the BP neural network model to be trained for forward calculation to obtain the model output data.

[0083] Specifically, a data sample can be used as the input to the input layer, processed layer by layer through the hidden layer, and passed to the output layer to obtain the model output data output by the output layer.

[0084] Exemplarily, the input to the input layer can be expressed by the following formula. Here, j represents the input layer, and M is the number of neurons in the input layer j.

[0085] O j = x(j) (j = 1, 2,......M)

[0086] Exemplarily, the input to the hidden layer can be expressed by the following formula. Here, i represents the hidden layer, and w ij represents the connection weight between the input layer and the hidden layer.

[0087]

[0088] Exemplarily, the output of the hidden layer can be expressed by the following formula. Here, f represents the activation function as the hyperbolic tangent function. Q is the number of neurons in the hidden layer i. k represents the current moment, or rather, represents the current iteration number.

[0089] o i (k) = f(net i (k)) i = 1, 2,......Q

[0090] Exemplarily, the input to the output layer can be expressed by the following formula. Here, l represents the output layer, and w ij represents the connection weight between the hidden layer and the output layer.

[0091]

[0092] Exemplarily, the output of the output layer can be expressed by the following formula. Here, g represents the activation function as the non - negative hyperbolic tangent function.

[0093] o l (k) = g(net l (k)) i = 1, 2, 3

[0094] Exemplarily, the model output data can be obtained based on o l (k) = g(net l (k)).

[0095] In this way, through the data sample and the BP neural network model, by forward - calculating the input signals and output signals of the neurons in each layer of the BP neural network model, the current output signal o l (k) = g(net l(k)), that is, the model output data of the BP neural network model can be obtained.

[0096] Step S330: Perform backpropagation error calculation based on the model output data and the expected output data of the data sample to update the model parameters of the BP neural network model to be trained.

[0097] Specifically, if the model output data does not match the expected output data or does not meet the set requirements, the error between the model output data and the expected output data can be used as an adjustment signal to propagate backward layer by layer, and the connection weights in the BP neural network model can be adjusted to reduce the error until the error is reduced to an acceptable range.

[0098] Exemplarily, the error can be determined based on the model output data and the expected output data; in the case where the error does not meet the set requirements, if the current iteration number is less than the specified iteration number, based on the learning rate and the negative gradient rule, the connection weights between the layers in the BP neural network model are corrected backward until the error meets the set requirements or the current iteration number reaches the specified iteration number.

[0099] When the error meets the set requirements or the current iteration number reaches the specified iteration number, the update or training of the BP neural network model is completed.

[0100] In the above embodiments, by constructing a data sample based on the historical creep condition data that can accurately reflect the state of the corresponding vehicle running under the creep condition, and inputting the data sample into the BP neural network model to be trained for forward calculation to obtain the model output data, and performing backpropagation error calculation based on the model output data and the expected output data of the data sample to update the model parameters of the BP neural network model to be trained, in this way, the trained BP neural network model can tune out PID parameters that are more suitable for the creep condition, and further improve the drivability of the vehicle when controlling the vehicle running under the creep condition subsequently.

[0101] In some embodiments, the PID control method may further include the following steps: obtaining the historical operation data of the vehicle; screening the historical operation data of the vehicle according to the creep condition detection conditions to obtain the historical creep condition data.

[0102] In the above embodiments, by screening the acquired historical vehicle operation data according to the detection conditions of the crawling working condition, the invalid data in the historical vehicle operation data can be cleaned, and the real historical crawling working condition data related to the crawling working condition can be obtained. In this way, the interference of the invalid data on the training process of the BP neural network model can be reduced, the training effect of the BP neural network model can be improved, so that the crawling PID parameters set by the BP neural network model can better fit the actual crawling working condition, and thus the drivability of the vehicle running in the crawling working condition can be improved.

[0103] In some embodiments, please refer to Figure 4 , and the target vehicle can be determined whether it is running in the crawling working condition in the following way.

[0104] Step S410: Obtain the data required for crawling judgment of the target vehicle; wherein, the data required for crawling judgment includes at least the crawling enable status data, vehicle speed, and fault data.

[0105] Among them, the crawling enable status data is the command data used to indicate that the vehicle travels in the crawling working condition.

[0106] Exemplarily, the data required for crawling judgment may further include at least one of the gear data, brake pedal depth, and accelerator pedal depth.

[0107] Step S420: If the data required for crawling judgment meets the detection conditions of the crawling working condition, it is determined that the target vehicle is running in the crawling working condition.

[0108] In the above embodiments, by obtaining the data required for crawling judgment including at least the crawling enable status data, vehicle speed, and fault data, and judging whether the data required for crawling judgment meets the detection conditions of the crawling working condition, and when the status data, vehicle speed, and fault data in the data required for crawling judgment meet the detection conditions of the crawling working condition, it is determined that the target vehicle is running in the crawling working condition, and the current crawling working condition data of the target vehicle running in the crawling working condition is input into the backpropagation neural network, so that the crawling PID parameters that fit the crawling working condition of the target vehicle can be timely set, and thus the drivability of the target vehicle running in the crawling working condition is improved.

[0109] An embodiment of the present application provides a vehicle PID control method, and the PID control method may include the following steps.

[0110] Step S502: The cloud platform obtains the historical vehicle operation data.

[0111] Step S504: Screen the historical vehicle operation data according to the detection conditions of the crawling working condition to obtain the historical crawling working condition data.

[0112] Step S506: Construct data samples based on historical creeping condition data. Among them, the data samples include at least one of vehicle speed, acceleration, brake opening, motor speed, motor torque, maximum creeping torque of the motor, pitch slope, vehicle speed difference, and road surface adhesion coefficient.

[0113] Step S508: Input the data samples into the BP neural network model to be trained for forward calculation to obtain the model output data.

[0114] Step S510: Perform backward error calculation based on the model output data and the expected output data of the data samples to update the model parameters of the BP neural network model to be trained.

[0115] Step S512: The target vehicle receives the model parameters issued by the cloud platform. Among them, the model parameters are obtained by training the BP neural network model based on the creeping condition data collected by the cloud platform for at least one vehicle; the at least one vehicle includes the target vehicle or does not include the target vehicle.

[0116] Step S514: The target vehicle determines the BP neural network model based on the received model parameters.

[0117] Step S516: Obtain the data required for creeping judgment of the target vehicle; among them, the data required for creeping judgment at least includes creeping enable state data, vehicle speed, and fault data.

[0118] Step S518: If the data required for creeping judgment meets the creeping condition detection conditions, it is determined that the target vehicle is running in the creeping condition.

[0119] Step S520: If the target vehicle is running in the creeping condition, obtain the current creeping condition data of the target vehicle when the target vehicle is controlled based on the current creeping control information.

[0120] Specifically, the current creeping condition data includes at least one of vehicle speed, acceleration, brake opening, motor speed, motor torque, maximum creeping torque of the motor, pitch slope, vehicle speed difference, and road surface adhesion coefficient; among them, the vehicle speed difference is the difference between the vehicle speed and the maximum creeping vehicle speed.

[0121] Step S522: Input the current creeping condition data into the BP neural network model to obtain the target creeping PID parameters output by the BP neural network model.

[0122] Step S524: Input the target creeping PID parameters into the vehicle PID controller to obtain the target creeping control information of the target vehicle.

[0123] Step S526: Control the target vehicle based on the target creeping control information of the target vehicle to obtain the new creeping condition data of the target vehicle.

[0124] Step S528: Upload the new creep driving condition data to the cloud platform so that a data sample can be constructed based on the new creep driving condition data to train the backpropagation neural network model.

[0125] An embodiment of the present application provides a vehicle PID control device. Please refer to Figure 5 , the PID control device may include a data acquisition module 510, a PID parameter determination module 520, and a vehicle control module 530.

[0126] The data acquisition module 510 is configured to acquire the current creep driving condition data of the target vehicle if the target vehicle is operating in the creep driving condition;

[0127] The PID parameter determination module 520 is configured to input the current creep driving condition data into the backpropagation neural network model to obtain the target creep PID parameters output by the backpropagation neural network model;

[0128] The vehicle control module 530 is configured to input the target creep PID parameters into the vehicle PID controller to obtain the target creep control information of the target vehicle.

[0129] In some embodiments, the PID control device may further include: a parameter receiving module, configured to receive the model parameters of the backpropagation neural network model sent by the cloud platform; wherein, the model parameters are obtained by training the backpropagation neural network model based on the creep driving condition data collected by the cloud platform from at least one vehicle; the at least one vehicle includes the target vehicle or does not include the target vehicle; a model determination module, configured to determine the backpropagation neural network model based on the received model parameters.

[0130] In some embodiments, the vehicle control module 530 is further configured to control the target vehicle based on the target creep control information of the target vehicle to obtain the new creep driving condition data of the target vehicle; upload the new creep driving condition data to the cloud platform so that a data sample can be constructed based on the new creep driving condition data to train the backpropagation neural network model.

[0131] In some embodiments, the data acquisition module 510 is further configured to acquire the data required for creep judgment of the target vehicle; wherein, the data required for creep judgment at least includes creep enabling state data, vehicle speed, and fault data; if the data required for creep judgment meets the creep driving condition detection condition, it is determined that the target vehicle is operating in the creep driving condition.

[0132] Regarding the specific functions and effects achieved by the PID control device, reference may be made to other embodiments of this application for comparative explanation, which will not be elaborated here. Each module in the PID control device can be implemented in whole or in part by software, hardware, and their combinations. Each module can be embedded in the processor of the computer device in hardware form or be independent of it, or can be stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to each of the above modules.

[0133] An embodiment of this application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by the computer, the computer executes the PID control method in any of the above embodiments.

[0134] An embodiment of this application also provides a computer program product containing instructions. When the instructions are executed by the computer, the computer executes the PID control method in any of the above embodiments.

[0135] An embodiment of this application also provides a vehicle, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the PID control method in any of the above embodiments.

[0136] An embodiment of this application also provides a computer device, including a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, it implements the PID control method in the above embodiment.

[0137] In this embodiment, please refer to Figure 6 , the computer device may be a terminal, and its internal structure diagram may be as Figure 6 shown. The computer device includes a processor, a memory, and a communication interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, a carrier network, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements the PID control method in any of the above embodiments.

[0138] It can be understood that the specific examples in this article are only for helping those skilled in the art to better understand the embodiments of this application, rather than limiting the scope of this application.

[0139] It can be understood that in various embodiments of the present application, the magnitudes of the serial numbers of the various processes do not imply the sequence of execution, and the execution sequence of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0140] It can be understood that the various embodiments described in the present application can be implemented alone or in combination, and the embodiments of the present application do not limit this.

[0141] It can be understood that the processor in the embodiments of the present application can be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method embodiments can be completed by the integrated logic circuit in the hardware of the processor or instructions in the form of software. The above processor can be a general-purpose processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. It can implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of the present application. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in combination with the embodiments of the present application can be directly embodied as being executed by the hardware decoding processor, or executed by a combination of the hardware and software modules in the decoding processor. The software module can be located in a mature storage medium in the art 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. This storage medium is located in the memory, and the processor reads the information in the memory and combines its hardware to complete the steps of the above method.

[0142] It can be understood that the memory in the embodiments of the present application can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories. Among them, the non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory can be a random access memory (RAM). It should be noted that the memory of the systems and methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0143] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0144] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0145] In several embodiments provided in this application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling, direct coupling, or communication connection to each other can be through some interfaces. The indirect coupling or communication connection of the devices or units can be in an electrical, mechanical, or other form.

[0146] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place, or they can be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0147] In addition, the functional units in each embodiment of this application can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit.

[0148] When the above-mentioned functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of this application. The foregoing storage medium includes: various media that can store program codes such as USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs.

[0149] The above is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art within the technical scope disclosed by this application can easily think of changes or substitutions, which should all be covered by the protection scope of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claims.

Claims

1. A vehicle PID control method, characterized in that: The method comprises: If the target vehicle is running in a creeping condition, obtaining current creeping condition data of the target vehicle; Inputting the current creep working condition data into a back propagation neural network model to obtain target creep PID parameters output by the back propagation neural network model; The target creep PID parameters are input into a vehicle PID controller to obtain target creep control information of the target vehicle.

2. The method according to claim 1, characterized in that: The current creep operating condition data can represent the vehicle state when the target vehicle is controlled based on the current creep control information; the current creep control information is determined by the vehicle PID controller based on the current creep PID parameters; If the current creep PID parameters are initial PID parameters, the initial PID parameters are obtained by processing the initial creep operating condition data generated by simulation using the back propagation neural network model.

3. The method according to claim 2, characterized in that The current creeping working condition data includes at least one of vehicle speed, acceleration, brake opening, motor speed, motor torque, maximum motor creeping torque, pitch gradient, vehicle speed difference, and road adhesion coefficient; wherein the vehicle speed difference is the difference between the vehicle speed and the maximum creeping vehicle speed; The initial creep operating condition data includes at least one of vehicle speed, acceleration, brake opening, motor speed, motor torque, maximum motor creep torque, pitch slope, vehicle speed difference, and road adhesion coefficient.

4. The method according to claim 1, characterized in that: Before inputting the current creeping condition data into the back propagation neural network model, the method further includes: Receiving model parameters of the back propagation neural network model issued by the cloud platform; wherein the model parameters are obtained by training the back propagation neural network model based on creeping condition data of at least one vehicle collected by the cloud platform; the at least one vehicle includes the target vehicle or does not include the target vehicle; The back propagation neural network model is determined based on the received model parameters.

5. The method according to claim 4, characterized in that The method further comprises: Controlling the target vehicle based on the target creep control information of the target vehicle to obtain new creep operating condition data of the target vehicle; The new creeping operating condition data is uploaded to the cloud platform so that data samples can be constructed based on the new creeping operating condition data to train the back propagation neural network model.

6. The method according to any one of claims 1 to 5, characterized in that: The back propagation neural network model is trained in the following manner, including: Constructing a data sample based on historical creep condition data, wherein the data sample includes at least one of vehicle speed, acceleration, brake opening, motor speed, motor torque, maximum motor creep torque, pitch slope, vehicle speed difference, and road adhesion coefficient; Inputting the data sample into the back propagation neural network model to be trained for forward calculation to obtain model output data; A reverse error calculation is performed based on the model output data and the expected output data of the data sample to update the model parameters of the back propagation neural network model to be trained.

7. The method according to claim 6, characterized in that The method further comprises: Obtain vehicle historical operation data; The historical vehicle operation data is screened according to the creep condition detection condition to obtain the historical creep condition data.

8. The method according to claim 1, characterized in that: Determine that the target vehicle is running in creep condition by the following method: Acquire the data required for creeping judgment of the target vehicle; wherein the data required for creeping judgment at least includes creeping energy state data, vehicle speed and fault data; If the data required for the creep judgment meets the creep condition detection conditions, it is determined that the target vehicle is operating in the creep condition.

9. A vehicle PID control device, characterized in that: The device comprises: A data acquisition module, used for acquiring current creeping condition data of the target vehicle if the target vehicle is running in a creeping condition; A PID parameter determination module, used for inputting the current creep working condition data into a back propagation neural network model to obtain a target creep PID parameter output by the back propagation neural network model; The vehicle control module is used to input the target creep PID parameters into a vehicle PID controller to obtain target creep control information of the target vehicle.

10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the vehicle PID control method according to any one of claims 1 to 8 is implemented.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the vehicle PID control method according to any one of claims 1 to 8 is implemented.

12. A vehicle comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the vehicle PID control method according to any one of claims 1 to 8 is implemented.

Images