Vehicle chassis drive-by-wire method and device, electronic equipment and readable storage medium

Abstract

The invention relates to the technical field of vehicle control, and provides a vehicle chassis drive-by-wire method and device, electronic equipment and a readable storage medium. The method comprises the steps of obtaining current actual state data of a vehicle at a current moment and historical state data of a plurality of continuous historical moments before the current moment; jointly inputting the current actual state data and the historical state data into a pre-trained prediction model to obtain a state change prediction value of the vehicle at the future moment relative to the current moment; calculating a plurality of state error amounts between the state change prediction value and a preset expected state change amount; inputting the plurality of state error amounts into a preset fuzzy controller, and performing fuzzy reasoning based on a preset fuzzy rule base in the fuzzy controller to generate a fuzzy control signal; performing conversion processing on the fuzzy control signal to obtain a control instruction for the vehicle chassis drive-by-wire system; and controlling the action of the vehicle chassis drive-by-wire execution mechanism according to the control instruction.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicle control, and in particular to a vehicle chassis-by-wire method and device, an electronic device, and a readable storage medium. BACKGROUND

[0002] As a core component of modern automotive electronic control systems, chassis-by-wire technology replaces traditional mechanical connections with electronic signals, enabling precise control of steering, braking, suspension, and other chassis systems. With the rapid development of autonomous driving technology, vehicles need to complete real-time perception, decision-making, and execution in complex and changing driving environments, which puts higher requirements on the intelligent level and response capability of the chassis-by-wire system.

[0003] In the prior art, chassis-by-wire methods based on traditional control theory often perform poorly when faced with the strong nonlinearity and time-varying characteristics of vehicle dynamics systems. These methods either rely on accurate mathematical models and lack adaptability to uncertain factors, or are difficult to meet real-time control requirements due to high computational complexity, or have inherent limitations in handling system delays and dynamic responses.

[0004] Therefore, it is currently difficult to effectively compensate for control delays while ensuring system real-time performance. SUMMARY

[0005] Therefore, the embodiments of the present application provide a vehicle chassis-by-wire method, device, electronic device, and readable storage medium to solve the problem of effectively compensating for control delays while ensuring system real-time performance in the prior art.

[0006] The first aspect of the embodiments of the present application provides a vehicle chassis-by-wire method, comprising: obtaining current actual state data of a vehicle at a current time and historical state data at a plurality of consecutive historical times before the current time; inputting the current actual state data and the historical state data into a pre-trained prediction model to obtain a state change prediction value of the vehicle at a future time relative to the current time; calculating a plurality of state error quantities between the state change prediction value and a preset expected state change quantity; inputting the plurality of state error quantities into a preset fuzzy controller, performing fuzzy reasoning based on a preset fuzzy rule base in the fuzzy controller, and generating a fuzzy control signal; performing conversion processing on the fuzzy control signal to obtain a control instruction for a vehicle chassis-by-wire system; controlling the action of a vehicle chassis-by-wire actuator according to the control instruction.

[0007] The second aspect of the embodiments of the present application provides a vehicle chassis-by-wire device, comprising: an acquisition module configured to acquire current actual state data of the vehicle at a current time and historical state data of the vehicle at a plurality of continuous historical times before the current time; a first input module configured to input the current actual state data and the historical state data into a pre-trained prediction model to obtain a state change prediction value of the vehicle at a future time relative to the current time; a calculation module configured to calculate a plurality of state error amounts between the state change prediction value and a preset expected state change amount; a second input module configured to input the plurality of state error amounts into a preset fuzzy controller to perform fuzzy reasoning based on a preset fuzzy rule base in the fuzzy controller to generate a fuzzy control signal; a conversion module configured to perform conversion processing on the fuzzy control signal to obtain a control instruction for a chassis-by-wire system of the vehicle; a control module configured to control actions of a chassis-by-wire actuating mechanism of the vehicle according to the control instruction.

[0008] In a third aspect, an electronic device is provided, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the steps of the above method when executing the computer program.

[0009] In a fourth aspect, a computer readable storage medium is provided, which stores a computer program, and the computer program implements the steps of the above method when executed by a processor.

[0010] Compared with the prior art, the embodiments of the present application have the beneficial effects that: the current actual state data of the vehicle at the current time and the historical state data of the vehicle at the plurality of continuous historical times before the current time are acquired and input into a prediction model to predict a state change prediction value of the vehicle at the future time relative to the current time, so that the control system can perceive the change in the motion trend of the vehicle in advance. The plurality of state error amounts between the state change prediction value and a preset expected state change amount are calculated, the plurality of state error amounts are input into a preset fuzzy controller, fuzzy reasoning is performed based on a preset fuzzy rule base in the fuzzy controller to generate a fuzzy control signal, and accurate description and control of the nonlinear dynamic characteristics of the vehicle are realized. Conversion processing is performed on the fuzzy control signal to obtain a control instruction for a chassis-by-wire system of the vehicle, and actions of a chassis-by-wire actuating mechanism of the vehicle are directly controlled according to the control instruction, which guarantees the control precision and significantly improves the response speed and adaptive ability of the system. BRIEF DESCRIPTION OF DRAWINGS

[0011] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 This is a schematic flowchart of a vehicle chassis drive-by-wire method provided in an embodiment of this application; Figure 2 This is a schematic flowchart of another vehicle chassis drive-by-wire method provided in an embodiment of this application; Figure 3 This is a schematic diagram of the structure of a vehicle chassis drive-by-wire device provided in an embodiment of this application; Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0013] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0014] A vehicle chassis drive-by-wire method and apparatus according to embodiments of this application will now be described in detail with reference to the accompanying drawings.

[0015] Figure 1 This is a schematic flowchart of a vehicle chassis drive-by-wire method provided in an embodiment of this application. Figure 1 The vehicle chassis drive-by-wire method can be executed by the server. For example... Figure 1 As shown, the vehicle chassis drive-by-wire method includes: Step S101: Obtain the current actual state data of the vehicle at the current moment, as well as the historical state data of multiple consecutive historical moments before the current moment; Current actual state data refers to the vehicle motion parameters collected in real time by vehicle sensors within the current control cycle. Historical state data refers to the sequence of vehicle state data collected at multiple consecutive sampling times prior to the current moment.

[0016] The vehicle state data is obtained through the vehicle-mounted sensors. These data include but are not limited to the lateral position, longitudinal position, longitudinal speed and longitudinal acceleration of the vehicle. For example, the system will collect these parameters at the current time and several previous sampling times to form a complete time series data. This provides a sufficient data basis for subsequent prediction and ensures the integrity of the prediction input.

[0017] Step S102: input the current actual state data and the historical state data into the pre-trained prediction model to obtain a state change prediction value of the vehicle at a future time relative to the current time; The prediction model refers to a long short-term memory network model trained to predict future state changes based on historical state data. The state change prediction value refers to the change amount of the vehicle state at the future time relative to the current time output by the prediction model.

[0018] The obtained state data sequence is input into the pre-trained prediction model. The prediction model adopts a long short-term memory network structure and can effectively learn the time sequence characteristics of the vehicle state change through its unique gating mechanism. For example, the model can predict the change amount of the vehicle relative to the current position and speed at the next time based on the vehicle motion trajectory in the past period. By using the powerful modeling capability of the deep learning model on time series data, accurate prediction of the future motion state of the vehicle is realized, which provides a basis for forward-looking control.

[0019] Step S103: calculate a plurality of state error amounts between the state change prediction value and a preset expected state change amount; The expected state change amount refers to the ideal state change amount preset according to the vehicle driving demand. The fuzzy controller is an intelligent controller based on fuzzy logic reasoning, and the core is the preset fuzzy rule base.

[0020] The predicted state change amount is compared with the expected state change amount. The expected state change amount can be determined according to the preset driving trajectory or driving strategy. For example, when driving on a curve, the expected lateral position change amount is calculated in real time according to the road curvature. By calculating the difference between the two amounts, a plurality of state error amounts are obtained, including lateral displacement error, longitudinal displacement error, etc. The abstract prediction result is converted into a specific control bias, which provides a quantitative basis for subsequent control decision.

[0021] Step S104: input the plurality of state error amounts into the preset fuzzy controller, and perform fuzzy reasoning based on the preset fuzzy rule base in the fuzzy controller to generate a fuzzy control signal; The fuzzy controller first converts precise error values ​​into fuzzy linguistic values, such as "positive large" or "negative small," and then performs inference based on a pre-set fuzzy rule base. For example, when the lateral displacement error is positive and the longitudinal velocity error is negative and small, the controller may output a large steering correction command. By utilizing the ability of fuzzy logic to handle uncertainty and nonlinearity, intelligent control of complex vehicle dynamics characteristics is achieved.

[0022] Step S105: Convert the fuzzy control signal to obtain the control command for the vehicle chassis drive-by-wire system; The fuzzy control signals output by the fuzzy controller are defuzzified and converted into precise control commands. These commands include steering angle commands for the steering system and acceleration commands for the drive and braking system. For example, defuzzification methods such as the center of gravity method or the maximum membership method are used to convert fuzzy steering correction commands into specific steering angle values. Transforming the fuzzy inference results into executable control signals ensures the accuracy of the control.

[0023] Step S106: Control the action of the vehicle chassis drive-by-wire actuator according to the control command.

[0024] The generated control commands are sent to the corresponding drive-by-wire actuators. Steering commands are sent to the steer-by-wire system to control the front wheel steering angle, and acceleration commands are sent to the brake-by-wire system or drive-by-wire system to control vehicle acceleration and deceleration. This achieves precise control of the vehicle's chassis system, ensuring that the vehicle travels stably along the intended trajectory.

[0025] In the embodiments of this application, by acquiring the current actual state data of the vehicle at the current moment, as well as the historical state data of multiple consecutive historical moments before the current moment, and inputting them into the prediction model, the predicted value of the vehicle's state change relative to the current moment at a future moment is predicted, enabling the control system to perceive changes in the vehicle's motion trend in advance. Multiple state error quantities are calculated between the predicted state change value and the preset expected state change quantity; these multiple state error quantities are input into a preset fuzzy controller, and fuzzy inference is performed based on the preset fuzzy rule base in the fuzzy controller to generate fuzzy control signals, achieving accurate description and control of the vehicle's nonlinear dynamic characteristics. The fuzzy control signals are converted and processed to obtain control commands for the vehicle chassis drive-by-wire system; the actions of the vehicle chassis drive-by-wire actuators are directly controlled according to the control commands, significantly improving the system's response speed and adaptive capability while ensuring control accuracy.

[0026] In some embodiments, step S101 includes: Motion parameters, including lateral position, longitudinal position, longitudinal velocity, and longitudinal acceleration, are collected through vehicle sensors. The collected motion parameters are filtered and preprocessed to obtain the current actual state data.

[0027] Raw motion parameters are collected by a multi-source sensor network deployed on the vehicle, including key state information such as lateral position coordinates, longitudinal position coordinates, longitudinal motion speed, and longitudinal acceleration. In practical applications, the lateral position and longitudinal position are usually obtained by fusing global positioning system and inertial navigation system data, the longitudinal speed can be calculated by wheel speed sensors, and the longitudinal acceleration can be directly measured by accelerometers.

[0028] Considering the noise interference and data anomalies in the sensor measurement process, the system will perform necessary filtering and data preprocessing on the collected raw motion parameters. For example, Kalman filter algorithm is used to smooth the position and speed data, and sliding window method is used to remove outliers of acceleration data. These preprocessing operations can effectively eliminate random errors caused by measurement noise and improve the accuracy and reliability of state data. The clean data obtained after these processes is the current actual state data, which provides high-quality input guarantee for the subsequent prediction model.

[0029] Through data fusion of multi-source sensors, the integrity and accuracy of state information are ensured; through advanced filtering algorithms, the interference of measurement noise is effectively suppressed; and the preprocessed data ultimately lays a reliable foundation for subsequent state prediction and control decision, ensuring the performance and stability of the entire control system from the source.

[0030] In some embodiments, the state change prediction value includes a lateral displacement change amount, a longitudinal displacement change amount, a longitudinal speed change amount, and a longitudinal acceleration change amount. The expected state change amount includes an expected lateral displacement change amount, an expected longitudinal displacement change amount, an expected longitudinal speed change amount, and an expected longitudinal acceleration change amount. The plurality of state error amounts includes a lateral displacement error amount, a longitudinal displacement error amount, a longitudinal speed error amount, and a longitudinal acceleration error amount.

[0031] Among them, the state change prediction value contains four key components: the lateral displacement change amount represents the position change trend of the vehicle in the lateral direction, the longitudinal displacement change amount reflects the position change of the vehicle in the forward direction, the longitudinal speed change amount reflects the rate of change of speed, and the longitudinal acceleration change amount describes the dynamic characteristics of acceleration. Correspondingly, the expected state change amount also contains four corresponding components: the expected lateral displacement change amount is determined according to the predetermined driving trajectory, the expected longitudinal displacement change amount is provided by the path planning module, the expected longitudinal speed change amount is set based on energy saving and efficiency principles, and the expected longitudinal acceleration change amount considers comfort and safety requirements.

[0032] By comparing the state change prediction value with each component of the expected state change amount item by item, the system can accurately calculate four key error amounts: the lateral displacement error amount reflects the deviation of the actual lateral movement from the expected trajectory, the longitudinal displacement error amount represents the longitudinal position tracking accuracy, the longitudinal velocity error amount embodies the accuracy of the speed control, and the longitudinal acceleration error amount reflects the acceleration tracking performance. For example, when the vehicle needs to perform a lane changing action, the expected lateral displacement change amount is calculated in real time according to the target lane position, and the error generated by comparing the actual output lateral displacement change prediction value with it will be directly used to control the action amplitude of the steering system.

[0033] By decomposing the overall state change into four independent and related movement components, the control system can accurately describe the vehicle movement state in multiple dimensions; secondly, the independent calculation of each error amount provides a clear input dimension for subsequent fuzzy control, enabling the controller to adopt corresponding control strategies for different movement characteristics; finally, this decomposition method effectively improves the system's adaptability to complex driving conditions, especially in scenarios that require simultaneous consideration of lateral and longitudinal movement coordination control, ensuring the smoothness and stability of vehicle movement.

[0034] In some embodiments, step S104 comprises: The lateral displacement error amount, the longitudinal displacement error amount, the longitudinal velocity error amount, and the longitudinal acceleration error amount are taken as input fuzzy variables; The fuzzy control signal is taken as an output fuzzy variable, and the output fuzzy variable includes a steering angle control instruction and an acceleration control instruction; The values of each input fuzzy variable are converted into fuzzy language values through corresponding membership functions, and the fuzzy language values include at least three of negative large, negative small, zero, positive small, and positive large; A preset fuzzy rule base is applied for reasoning, and the fuzzy rule base contains multiple rules, and each rule is in the form of: if the lateral displacement error amount is a first language value, the longitudinal displacement error amount is a second language value, the longitudinal velocity error amount is a third language value, and the longitudinal acceleration error amount is a fourth language value, then the steering angle control instruction is a fifth language value and the acceleration control instruction is a sixth language value.

[0035] The lateral displacement error amount, the longitudinal displacement error amount, the longitudinal velocity error amount, and the longitudinal acceleration error amount are taken as input fuzzy variables, and the steering angle control instruction and the acceleration control instruction in the control instruction are defined as output fuzzy variables, establishing a complete mapping relationship from the vehicle state error to the control instruction and providing a structural foundation for intelligent decision-making.

[0036] In the fuzzification stage, the precise values of each input fuzzy variable are converted into fuzzy linguistic values by specially designed membership functions. These linguistic values are divided into five levels, including negative large, negative small, zero, positive small and positive large. For example, when the lateral displacement error reaches the maximum allowable deviation value, its membership function classifies it as the "positive large" level; when the error is close to zero, it is classified as the "zero" level. The technical advantage of this fuzzification process is that it can convert precise numerical errors into semantic descriptions suitable for logical reasoning, creating conditions for subsequent rule-based reasoning.

[0037] The pre-set rule base contains a series of uniformly formatted inference rules. Each rule adopts the form of "if-then" conditional statements, establishing a corresponding relationship between the language values of the four input variables and the language values of the two output variables. For example, a rule may stipulate that if the lateral displacement error is positive large, the longitudinal displacement error is positive small, the longitudinal velocity error is zero, and the longitudinal acceleration error is negative small, then the steering angle control command is positive large and the acceleration control command is negative small.

[0038] Through the multi-input and multi-output rule structure, the coupling problem of lateral and longitudinal control in vehicle motion is effectively solved. Secondly, the semantic reasoning method is adopted, which avoids the dependence on complex dynamic models and enhances the practicality of the system. Finally, the scalability of the rule base enables the control system to adapt to more special working conditions by supplementing new rules, significantly improving the adaptability of the algorithm. This design ensures that the control system can maintain the reliability of traditional methods while possessing intelligent characteristics for handling complex nonlinear problems.

[0039] In some embodiments, the prediction model is trained by: obtaining a training data set, the training data set containing multiple groups of samples, each group of samples including a sample state data sequence collected at multiple sample time points, and a corresponding actual state change amount at a future sample time point relative to the last sample time point of the sequence; training the initial prediction model by taking the sample state data sequence as input and the actual state change amount as expected output until the model converges.

[0040] Each training sample consists of two parts: one part is a sequence of vehicle state data collected at consecutive multiple sample time points, including key parameters such as lateral position, longitudinal position, longitudinal velocity and longitudinal acceleration; the other part is the corresponding actual state change amount, which calculates the state difference value at a future sample time relative to the last sample time of the current sequence.

[0041] In the data preparation phase, the quality of the training data directly affects the prediction accuracy of the model. For example, vehicle driving data collected from actual road tests need to be strictly screened and cleaned to remove outliers and noise interference, ensuring that the samples truly reflect the normal operating state of the vehicle. These sample data also need to be normalized to unify different dimensional state parameters into the same numerical range, in order to improve the training efficiency and convergence speed of the model.

[0042] In the model training phase, supervised learning is used, with the sample state data sequence as the input feature and the corresponding actual state change as the training label. The root mean square error is used as the loss function during training, and the model parameters are continuously adjusted through the gradient descent algorithm, so that the prediction output gradually approaches the true label value. For example, when a sequence containing the vehicle motion state at the past few time points is input, the model should be able to accurately predict the state change trend at the future time relative to the current time through learning.

[0043] By using the actual state change as the training target, the model focuses on learning the change rule of the state rather than the absolute state value, which is more in line with the actual needs of the control system. Secondly, the sequence-to-value training method fully utilizes the advantages of long short-term memory networks in handling time series data. Finally, training based on real road data ensures that the model can adapt to various complex driving conditions. The well-trained prediction model can accurately capture the dynamic characteristics of vehicle motion and provide reliable prediction information for subsequent control decisions, thereby significantly improving the performance and stability of the entire control system.

[0044] In an embodiment of the present application, the prediction model is implemented using a long short-term memory network (LSTM). As a special type of recurrent neural network, LSTM has a unique gating mechanism, including an input gate, a forget gate, and an output gate, which can effectively control the flow of information, selectively remember important information and forget unimportant information, thereby solving the problem of gradient explosion or disappearance that often occurs in the training process of traditional recurrent neural networks. This characteristic makes it particularly suitable for handling prediction tasks such as vehicle motion state, which has strong temporal correlation.

[0045] In this embodiment, the LSTM network adopts a four-layer hidden layer structure, with the network input being a historical state sequence containing vehicle trajectory coordinates, longitudinal speed, and longitudinal acceleration, and the output being a state change prediction value at the next time step. The network training uses the Adam optimizer based on gradient optimization, which minimizes the root mean square error between the prediction value and the true value through the backpropagation algorithm. To improve the generalization ability of the model, the ten-fold cross-validation method is used during training, with the data set randomly divided into training and test sets to ensure that the model maintains stable prediction performance under different data distributions.

[0046] In the actual prediction process, the LSTM network obtains the vehicle state data at the current time and multiple consecutive times before the current time in a sliding window manner, and predicts the state change amount at the future time relative to the current time based on the historical information. For example, when the system needs to predict the vehicle state after 0.5 seconds in the future, the network will take the vehicle motion trajectory in the past 2 seconds as input, extract the time sequence features through the internal gating mechanism, and finally output accurate state change prediction. This rolling prediction method can fully utilize the modeling capability of LSTM for long time series, and accurately capture the future vehicle motion trend.

[0047] The LSTM network automatically learns the complex characteristics of the vehicle dynamics system through data-driven methods, avoiding the tedious parameter tuning process in traditional control methods. Secondly, the natural robustness of the network to noise and uncertainty makes the prediction results more reliable, reducing the control deviation caused by sensor errors. Finally, the LSTM model after sufficient training has high computational efficiency in the inference stage, which can meet the strict real-time requirements of the chassis-by-wire system. These characteristics enable the LSTM-based prediction model to provide accurate and reliable forward-looking information for subsequent fuzzy control, thereby significantly improving the performance and adaptability of the entire chassis-by-wire system.

[0048] In some embodiments, step S106 comprises: Step S201, sending the steering control instruction in the control instruction to the steer-by-wire system to control the front wheel steering angle; Step S202, sending the acceleration control instruction in the control instruction to the brake-by-wire system or drive-by-wire system to control the vehicle acceleration or braking.

[0049] The control instructions generated through the foregoing steps are parsed and distributed, wherein the steering control instruction is sent to the steer-by-wire system for precise control of the front wheel steering angle; at the same time, the acceleration control instruction is sent to the brake-by-wire system or drive-by-wire system to realize precise control of the vehicle acceleration or braking operation.

[0050] In terms of steering control, the steer-by-wire system receives the steering control instruction, which is parsed by the control unit, and then drives the steering execution motor to generate a corresponding steering torque. For example, when the system needs the vehicle to perform lane keeping, it will control the front wheel to generate an accurate deflection angle according to the real-time calculated steering angle requirement. The technical advantage of this process is to realize fast response and precise control of the steering system, providing good path tracking capability for the vehicle.

[0051] In terms of acceleration control, the system intelligently selects the control object according to the driving demand. When acceleration is needed, the acceleration control instruction is sent to the drive-by-wire driving system to achieve smooth acceleration by adjusting the output torque of the driving motor; when deceleration is needed, the instruction is sent to the drive-by-wire braking system to achieve precise braking by controlling the pressure of the brake caliper. For example, in the following vehicle driving scenario, the system calculates the safety distance and relative speed to generate appropriate acceleration instructions to maintain a reasonable vehicle distance.

[0052] Through the coordinated work of the steering system and the driving and braking system, decoupling control of vehicle lateral and longitudinal motion is achieved, and the control precision is improved; secondly, the control mode based on electrical signal transmission greatly shortens the system response time and improves the control real-time performance; finally, this design enables the control system to adapt to various complex driving conditions, whether it is highway cruising or urban congested road conditions, and ensures the smoothness and safety of vehicle operation. When the stability monitoring system issues an adjustment instruction, the two execution systems can respond cooperatively to ensure the driving stability of the vehicle.

[0053] In some embodiments, before step S106, the method further comprises: obtaining the real-time yaw rate and the center of mass side slip angle of the vehicle; adjusting the control instruction based on the real-time yaw rate and the center of mass side slip angle.

[0054] By obtaining the yaw rate and the center of mass side slip angle of the vehicle in real time, two key dynamic parameters, the control instruction to be executed is optimized and verified. The yaw rate reflects the rotation rate of the vehicle around the vertical axis, and the center of mass side slip angle represents the deviation of the actual driving direction of the vehicle from the heading direction. These two parameters are the core indicators for evaluating the driving stability of the vehicle.

[0055] The yaw rate of the vehicle is monitored in real time by the inertial measurement unit, and the center of mass side slip angle is estimated by the state observer. These real-time data are compared with the preset stability range threshold. For example, when the yaw rate is close to the critical value or the center of mass side slip angle exceeds the safe range, the system will dynamically adjust the control instruction according to the preset optimization algorithm. This adjustment may be manifested as a moderate reduction in the steering angle instruction, a reduction in the acceleration instruction value, or the introduction of appropriate compensation control.

[0056] On the basis of maintaining the original control performance, a layer of safety protection is added to effectively prevent the vehicle from losing stability in extreme conditions; secondly, through real-time optimization and adjustment, the control system can fully exert its performance while ensuring that it always works in a safe area; finally, this design enhances the adaptability to different road conditions and driving conditions, especially in low adhesion road conditions or emergency obstacle avoidance, which can significantly improve the active safety of the vehicle.

[0057] All the optional technical solutions described above can be combined to form optional embodiments of the present application, which will not be described here.

[0058] The following is an embodiment of the device of the present application, which can be used to execute the method embodiments of the present application. For details not disclosed in the device embodiments of the present application, please refer to the method embodiments of the present application.

[0059] Figure 3 is a schematic diagram of a vehicle chassis wire control device provided by an embodiment of the present application. As shown in Figure 3 , the vehicle chassis wire control device comprises: The acquisition module 301 is configured to acquire current actual state data of the vehicle at a current time and historical state data at a plurality of continuous historical times before the current time. The first input module 302 is configured to input the current actual state data and the historical state data into a pre-trained prediction model to obtain a state change prediction value of the vehicle at a future time relative to the current time. The calculation module 303 is configured to calculate a plurality of state error quantities between the state change prediction value and a preset expected state change quantity. The second input module 304 is configured to input the plurality of state error quantities into a preset fuzzy controller to perform fuzzy reasoning based on a preset fuzzy rule base in the fuzzy controller to generate a fuzzy control signal. The conversion module 305 is configured to perform conversion processing on the fuzzy control signal to obtain a control instruction for the vehicle chassis wire control system. The control module 306 is configured to control the action of the vehicle chassis wire control actuator according to the control instruction.

[0060] According to the technical solution provided by the embodiment of the present application, by acquiring the current actual state data of the vehicle at the current time and the historical state data at the plurality of continuous historical times before the current time and inputting them into the prediction model, the state change prediction value of the vehicle at the future time relative to the current time is predicted, so that the control system can perceive the change of the vehicle motion trend in advance. The plurality of state error quantities between the state change prediction value and the preset expected state change quantity are calculated; the plurality of state error quantities are input into the preset fuzzy controller to perform fuzzy reasoning based on the preset fuzzy rule base in the fuzzy controller to generate the fuzzy control signal, thereby realizing accurate description and control of the nonlinear dynamic characteristics of the vehicle. The fuzzy control signal is converted to obtain the control instruction for the vehicle chassis wire control system; the action of the vehicle chassis wire control actuator is directly controlled according to the control instruction, which guarantees the control precision while significantly improving the response speed and adaptive ability of the system. In some embodiments, the acquisition module 301 is specifically configured to: collect, by a vehicle sensor, motion parameters including a lateral position, a longitudinal position, a longitudinal speed, and a longitudinal acceleration; The collected motion parameters are filtered and preprocessed to obtain current actual state data.

[0061] In some embodiments, the acquisition module 301 is further configured to: acquire a real-time yaw rate and a center of mass side slip angle of the vehicle. The vehicle chassis drive-by-wire device further comprises: The adjustment module 307 is configured to adjust the control instruction based on the real-time yaw rate and the center of mass side slip angle.

[0062] In some embodiments, the state change prediction value includes a lateral displacement change amount, a longitudinal displacement change amount, a longitudinal speed change amount, and a longitudinal acceleration change amount. The expected state change amount includes an expected lateral displacement change amount, an expected longitudinal displacement change amount, an expected longitudinal speed change amount, and an expected longitudinal acceleration change amount. The plurality of state error amounts include a lateral displacement error amount, a longitudinal displacement error amount, a longitudinal speed error amount, and a longitudinal acceleration error amount.

[0063] In some embodiments, the second input module 304 is specifically configured to: take the lateral displacement error amount, the longitudinal displacement error amount, the longitudinal speed error amount, and the longitudinal acceleration error amount as input fuzzy variables. The fuzzy control signal is taken as an output fuzzy variable, and the output fuzzy variable includes a steering angle control instruction and an acceleration control instruction. The values of the input fuzzy variables are converted into fuzzy language values through corresponding membership functions, and the fuzzy language values include at least three of negative large, negative small, zero, positive small, and positive large. A preset fuzzy rule base is applied for reasoning, the fuzzy rule base contains a plurality of rules, and each rule is in the form of: if the lateral displacement error amount is a first language value, the longitudinal displacement error amount is a second language value, the longitudinal speed error amount is a third language value, and the longitudinal acceleration error amount is a fourth language value, then the steering angle control instruction is a fifth language value and the acceleration control instruction is a sixth language value.

[0064] In some embodiments, the acquisition module 301 is specifically configured to: Acquire a training data set, the training data set contains a plurality of samples, each sample includes a sample state data sequence collected at a plurality of sample moments, and a corresponding actual state change amount at a future sample moment relative to a last sample moment of the sequence; The vehicle chassis drive-by-wire device further comprises: The training module 308 is configured to train the initial prediction model by taking the sample state data sequence as input and taking the actual state change amount as expected output until the model converges.

[0065] In some embodiments, the control module 306 is specifically configured to: send the steering control instruction in the control instruction to the steer-by-wire system to control the front wheel steering angle; send the acceleration control instruction in the control instruction to the brake-by-wire system or drive-by-wire system to control the vehicle acceleration or braking.

[0066] It should be understood that the size of the serial number of each step in the above embodiments does not mean the order of execution, and the execution order of each process should be determined according to its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0067] Figure 4 is a schematic diagram of an electronic device 4 provided by an embodiment of the present application. As shown in the figure, the electronic device 4 of this embodiment includes a processor 401, a memory 402, and a computer program 403 stored in the memory 402 and executable on the processor 401. The processor 401 implements the steps in each of the above method embodiments when executing the computer program 403. Alternatively, the processor 401 implements the functions of each module / unit in each of the above device embodiments when executing the computer program 403. Figure 4

[0068] The electronic device 4 can be a desktop computer, a notebook computer, a palm computer, a cloud server, and the like. The electronic device 4 can include but is not limited to the processor 401 and the memory 402. Those skilled in the art can understand that the electronic device 4 can include more or fewer components or different components than those shown in the figure. Figure 4 The electronic device 4 is merely an example and does not constitute a limitation on the electronic device 4, which can include more or fewer components or different components than those shown in the figure.

[0069] The processor 401 can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gates or transistor logic components, discrete hardware components, etc.

[0070] ​The memory 402 can be an internal storage unit of the electronic device 4, for example, a hard disk or a memory of the electronic device 4. The memory 402 can also be an external storage device of the electronic device 4, for example, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc. equipped on the electronic device 4. The memory 402 can also include both the internal storage unit and the external storage device of the electronic device 4. The memory 402 is used to store computer programs and other programs and data required by the electronic device.

[0071] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the above division of functional units and modules is exemplified, and in actual application, the above functions can be completed by different functional units and modules according to needs, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiment can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0072] The integrated module / unit, if realized in the form of a software functional unit and sold or used as an independent product, can be stored in a readable storage medium. Based on this understanding, all or part of the processes in the above-mentioned embodiment methods can also be instructed by a computer program to related hardware to complete, and the computer program can be stored in a readable storage medium. When the processor executes the computer program, the steps of each method embodiment described above can be implemented. The computer program can include computer program code, which can be in the form of source code, object code, executable files or some intermediate forms, etc. The computer readable medium can include any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content included in the computer readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction, for example, in some jurisdictions, according to legislation and patent practice, the computer readable medium does not include electric carrier signals and telecommunication signals.

[0073] The above examples are only used to illustrate the technical solutions of the present application, but not limit the same; although the present application has been described in detail with reference to the foregoing examples, those skilled in the art should understand that the technical solutions recorded in the foregoing examples can be modified, or some technical features can be replaced by equivalent ones; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A vehicle chassis drive-by-wire method, characterized in that, The method includes: Obtain the vehicle's current actual state data at the current moment, as well as historical state data from multiple consecutive historical moments prior to the current moment; The current actual state data and the historical state data are input together into a pre-trained prediction model to obtain the predicted value of the vehicle's state change relative to the current time at a future time. Calculate multiple state error values ​​between the predicted state change value and the preset expected state change amount; The multiple state error quantities are input to a preset fuzzy controller, and fuzzy inference is performed based on the preset fuzzy rule base in the fuzzy controller to generate a fuzzy control signal. The fuzzy control signal is converted and processed to obtain control commands for the vehicle chassis drive-by-wire system; The vehicle chassis drive-by-wire actuator is controlled according to the control command.

2. The method according to claim 1, characterized in that, The acquisition of the vehicle's current actual state data at the current moment includes: Motion parameters, including lateral position, longitudinal position, longitudinal velocity, and longitudinal acceleration, are collected through vehicle sensors. The collected motion parameters are filtered and preprocessed to obtain the current actual state data.

3. The method according to claim 1, characterized in that, Before controlling the action of the vehicle chassis drive-by-wire actuator according to the control command, the method further includes: Obtain the vehicle's real-time yaw rate and sideslip angle; The control commands are adjusted based on the real-time yaw rate and the sideslip angle of the center of mass.

4. The method according to claim 1, characterized in that, The predicted values ​​of the state change include the change in lateral displacement, the change in longitudinal displacement, the change in longitudinal velocity, and the change in longitudinal acceleration; The desired changes in state include the desired changes in lateral displacement, the desired changes in longitudinal displacement, the desired changes in longitudinal velocity, and the desired changes in longitudinal acceleration. The multiple state error quantities include lateral displacement error, longitudinal displacement error, longitudinal velocity error, and longitudinal acceleration error.

5. The method according to claim 4, characterized in that, The step of inputting the multiple state error quantities to a preset fuzzy controller and performing fuzzy inference based on a preset fuzzy rule base in the fuzzy controller includes: The lateral displacement error, the longitudinal displacement error, the longitudinal velocity error, and the longitudinal acceleration error are used as input fuzzy variables. The fuzzy control signal is used as the output fuzzy variable, which includes steering angle control command and acceleration control command; The values ​​of each input fuzzy variable are converted into fuzzy language values ​​through the corresponding membership function. The fuzzy language values ​​include at least three of the following: negative large, negative small, zero, positive small, and positive large. Reasoning is performed using a preset fuzzy rule base, which contains multiple rules. Each rule is in the form of: if the lateral displacement error is a first language value, the longitudinal displacement error is a second language value, the longitudinal velocity error is a third language value, and the longitudinal acceleration error is a fourth language value, then the steering angle control command is a fifth language value and the acceleration control command is a sixth language value.

6. The method according to claim 1, characterized in that, The prediction model is trained in the following way: Obtain a training dataset, which contains multiple sets of samples. Each set of samples includes a sequence of sample state data collected at multiple sample times, and the corresponding actual state change at future sample times relative to the last sample time of the sequence. The sample state data sequence is used as input, and the actual state change is used as the expected output to train the initial prediction model until the model converges.

7. The method according to claim 1, characterized in that, The control of the vehicle chassis drive-by-wire actuator according to the control command includes: The steering control command in the control command is sent to the steer-by-wire system to control the front wheel steering angle; The acceleration control command in the control command is sent to the brake-by-wire system or drive-by-wire system to control the vehicle's acceleration or braking.

8. A vehicle chassis drive-by-wire device, characterized in that, Applied to servers, including: The acquisition module is used to acquire the current actual state data of the vehicle at the current moment, as well as the historical state data of multiple consecutive historical moments before the current moment; The first input module is used to input the current actual state data and the historical state data into a pre-trained prediction model to obtain the predicted value of the vehicle's state change at a future time relative to the current time. The calculation module is used to calculate multiple state error quantities between the predicted state change value and the preset expected state change amount; The second input module is used to input the multiple state error quantities to a preset fuzzy controller, perform fuzzy inference based on the preset fuzzy rule base in the fuzzy controller, and generate a fuzzy control signal. The conversion module is used to convert and process the fuzzy control signal to obtain control commands for the vehicle chassis drive-by-wire system; The control module is used to control the action of the vehicle chassis drive-by-wire actuator according to the control commands.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 7.

Images