Vehicle suspension control method, electronic equipment and storage medium

Abstract

The invention provides a vehicle suspension control method, electronic equipment and a storage medium, and the method comprises the steps: obtaining a geometric structure parameter and a suspension parameter adjustment model of a vehicle, determining the current vehicle state, the current suspension parameter and the current road surface information of the vehicle, inputting the parameters into the suspension parameter adjustment model, obtaining a target suspension parameter of the vehicle, and obtaining the target suspension parameter of the vehicle; therefore, the suspension system of the vehicle is controlled, suspension adjustment based on the neural network model is achieved, the method can be suitable for various vehicle types, for different types of vehicle suspensions, only geometric structure parameters of the vehicle need to be input into the model, the universality and applicability of the model are improved, and in addition, the method is easy to implement. According to the method, the model can better adapt to different road conditions, the adaptability and the adjusting effect of suspension adjusting are improved, meanwhile, the model can be more accurate, a suspension system can be adjusted more accurately, and therefore vehicle body movement is reduced, vehicle bumping and inclination are eliminated and reduced, and driving comfort is improved.

Description

Technical Field

[0001] The present application relates to the field of automobile technology, and in particular to a vehicle suspension control method, electronic equipment and storage medium. Background Art

[0002] In modern automobile technology, the magic carpet function is an important comfort configuration that reduces body movement, eliminates and reduces bumps and tilts by adjusting the height and damping of the suspension. The traditional magic carpet function adjusts the height and damping of the suspension according to the ups and downs of the road and the relationship between the tire and the body. This technology improves driving comfort to a certain extent, but it also has some problems and limitations.

[0003] First, the vehicle suspension model cannot be reused for different types of vehicles, and each model needs to be designed and verified separately, which increases the complexity and cost of design and verification; second, the existing mathematical modeling methods may find it difficult to capture all relevant relationships, resulting in inaccurate adjustments and affecting driving comfort; in addition, the existing magic carpet function has poor adaptability to road conditions, and when road conditions change greatly, the adjustment effect may not be good. Summary of the invention

[0004] In view of the above-mentioned defects or deficiencies in the prior art, the present application aims to provide a vehicle suspension control method, an electronic device and a storage medium to solve the problems of high cost, inaccurate suspension adjustment and poor adjustment effect in the prior art.

[0005] The present application provides a vehicle suspension control method, the method comprising:

[0006] Acquiring geometric structure parameters of the vehicle and a suspension parameter adjustment model, wherein the suspension parameter adjustment model is a pre-trained neural network model;

[0007] Determining a current vehicle state, current suspension parameters, and current road surface information of the vehicle;

[0008] Inputting the geometric structure parameters, the current road surface information, the current suspension parameters and the current vehicle state into the suspension parameter adjustment model to obtain the target suspension parameters of the vehicle;

[0009] A suspension system of the vehicle is controlled according to the target suspension parameters.

[0010] Optionally, the current road surface information includes a current predicted contact trajectory between a tire of the vehicle and a road surface, and determining the current road surface information of the vehicle includes:

[0011] Acquiring current road surface point cloud data of the vehicle, and determining a current road surface fitting surface based on the current road surface point cloud data;

[0012] The predicted tire position of the vehicle is determined based on the current vehicle state, and the current predicted contact trajectory between the tire of the vehicle and the road surface is determined according to the current road surface fitting surface and the predicted tire position.

[0013] Optionally, after obtaining the current road surface point cloud data of the vehicle, the method further includes:

[0014] Dividing the current road surface point cloud data into regional point cloud data corresponding to a plurality of road surface regions;

[0015] For each of the regional point cloud data corresponding to the road surface area, a parameter mean corresponding to the regional point cloud data is determined based on the parameters of each point in the regional point cloud data, and the parameters of each point in the regional point cloud data are updated according to the parameter mean.

[0016] Optionally, determining the predicted tire position of the vehicle based on the current vehicle state includes:

[0017] Determining a current predicted driving trajectory of the vehicle based on a current speed, a current acceleration, and a current heading angle in the current vehicle state;

[0018] A predicted tire position of each tire in the vehicle is determined based on the current predicted driving trajectory of the vehicle.

[0019] Optionally, the training process of the suspension parameter adjustment model includes:

[0020] A training data set is collected based on a pre-built vehicle test bench, wherein the vehicle test bench is composed of a test vehicle fixed in parallel slide rails, the suspension height of the test vehicle is adjusted by the oil volume of the oil cylinder in the suspension height adjuster, and the test vehicle is always in contact with the road surface;

[0021] A neural network model is constructed, and the neural network model is trained based on the training data set to obtain the suspension parameter adjustment model.

[0022] Optionally, after controlling the suspension system of the vehicle according to the target suspension parameter, the method further includes:

[0023] Acquiring a body posture of the vehicle;

[0024] constructing a posture change curve of the vehicle according to each vehicle body posture, and judging whether the vehicle meets the model optimization condition based on the change rate of the posture change curve;

[0025] If so, the suspension parameter adjustment model is optimized and trained.

[0026] Optionally, the method further includes:

[0027] Determining reference suspension parameters based on the current road surface information, the current suspension parameters, and a vehicle suspension model of the vehicle, wherein the vehicle suspension model is a mathematical model describing a relationship between a road surface and a suspension system;

[0028] Accordingly, after obtaining the target suspension parameters of the vehicle, the method further includes:

[0029] The target suspension parameters are updated based on the reference suspension parameters.

[0030] Optionally, before controlling the suspension system of the vehicle according to the target suspension parameter, the method further includes:

[0031] determining whether the target suspension parameter is greater than a preset parameter threshold, and if so, updating the target suspension parameter according to the preset parameter threshold; and / or,

[0032] A parameter difference between the target suspension parameter and the current suspension parameter is determined, and if the parameter difference is greater than a preset parameter difference threshold, the target suspension parameter is updated according to the preset parameter difference threshold.

[0033] The present application also provides a vehicle suspension control device, the device comprising:

[0034] A model acquisition module, used to acquire geometric structure parameters of the vehicle and a suspension parameter adjustment model, wherein the suspension parameter adjustment model is a pre-trained neural network model;

[0035] An input determination module, used to determine a current vehicle state, current suspension parameters, and current road surface information of the vehicle;

[0036] A model output module, used for inputting the geometric structure parameters, the current road surface information, the current suspension parameters and the current vehicle state into the suspension parameter adjustment model to obtain the target suspension parameters of the vehicle;

[0037] A suspension control module is used to control a suspension system of the vehicle according to the target suspension parameters.

[0038] An embodiment of the present application further provides an electronic device, the electronic device comprising:

[0039] Processor and memory;

[0040] The processor is used to execute the steps of the vehicle suspension control method provided in any embodiment of the present application by calling the program or instruction stored in the memory.

[0041] An embodiment of the present application also provides a computer-readable storage medium, which stores a program or instruction, and the program or instruction enables a computer to execute the steps of the vehicle suspension control method provided by any embodiment of the present application.

[0042] In summary, the present application proposes a vehicle suspension control method, which obtains the geometric structure parameters and suspension parameter adjustment model of the vehicle, and determines the current vehicle state, current suspension parameters and current road surface information of the vehicle, and then inputs the geometric structure parameters, current road surface information, current suspension parameters and the current vehicle state into the suspension parameter adjustment model to obtain the target suspension parameters of the vehicle, so as to control the suspension system of the vehicle according to the target suspension parameters, and realizes suspension adjustment based on a pre-trained neural network model, wherein the suspension parameter adjustment model can be applicable to various vehicle models. For different types of vehicle suspensions, the same model can be used for suspension adjustment by simply inputting the geometric structure parameters of the vehicle into the model, thereby improving the versatility and applicability of the model and reducing the complexity and cost of model design. In addition, the method combines the current road surface information, the current vehicle state and the current suspension parameters to determine the target suspension parameters, so that the model can better adapt to different road conditions, improve the adaptability and adjustment effect of suspension adjustment, and at the same time, make the model more accurate and be able to more accurately adjust the suspension system, thereby reducing body movement, eliminating and reducing vehicle bumps and tilts, and improving driving comfort. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the specific implementation methods of the present application or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0044] Figure 1 A flow chart of a vehicle suspension control method provided in an embodiment of the present application;

[0045] Figure 2 A data collection schematic diagram of a vehicle test bench provided in an embodiment of the present application;

[0046] Figure 3 A schematic diagram of a hydraulic lifting suspension system in an embodiment of the present application;

[0047] Figure 4 A schematic diagram of a neural network model provided in an embodiment of the present application;

[0048] Figure 5 A schematic diagram of a fitting surface provided in an embodiment of the present application;

[0049] Figure 6 A schematic diagram of a magic carpet function provided in an embodiment of the present application;

[0050] Figure 7 A schematic structural diagram of a vehicle suspension control device provided in an embodiment of the present application;

[0051] Figure 8 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0052] The present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the relevant invention, rather than to limit the invention. It is also necessary to explain that, for ease of description, only the parts related to the invention are shown in the accompanying drawings.

[0053] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0054] As mentioned in the background technology, in response to the problems in the prior art, the present application proposes a vehicle suspension control method. Figure 1 A flow chart of a vehicle suspension control method provided in an embodiment of the present application. Figure 1 , the vehicle suspension control method specifically includes:

[0055] S110, obtaining geometric structure parameters of the vehicle and a suspension parameter adjustment model, wherein the suspension parameter adjustment model is a pre-trained neural network model.

[0056] The geometric structural parameters may be geometric parameters describing the vehicle structure, and specifically may be geometric parameters related to the vehicle's suspension system, such as the vehicle's wheelbase and the length of the vehicle's half-axle.

[0057] For example, the geometric structure parameters of the vehicle can be written into the local storage space of the vehicle in advance, and then the geometric structure parameters of the vehicle can be read from the local storage space. Alternatively, a structure parameter acquisition request can be sent to the cloud or remote server, and the request includes the vehicle model and vehicle identification number. The cloud or remote server can query the corresponding geometric structure parameters according to the vehicle model and vehicle identification number carried in the request and return them to the vehicle.

[0058] The suspension parameter adjustment model may be a pre-trained unified adjustment model applicable to various vehicle models. In a specific implementation, the training process of the suspension parameter adjustment model includes the following steps:

[0059] Step 11: Collect a training data set based on a pre-built vehicle test bench, wherein the vehicle test bench is composed of a test vehicle fixed in parallel slide rails, the suspension height of the test vehicle is adjusted by the oil volume of the oil cylinder in the suspension height adjuster, and the test vehicle is always in contact with the road surface;

[0060] Step 12: construct a neural network model, train the neural network model based on the training data set, and obtain a suspension parameter adjustment model.

[0061] The test vehicle can be fixed to the parallel rails (the parallel rails are fixed) in advance to ensure that the posture of the test vehicle remains unchanged, thus realizing the construction of the vehicle test bench. The test vehicle in the vehicle test bench can use a height-variable suspension system to ensure that the test vehicle is in contact with the road surface during driving. Figure 2 As shown, Figure 2 This is a data collection schematic diagram of a vehicle test bench provided in an embodiment of the present application, wherein the posture of a test vehicle in the vehicle test bench is fixed by parallel slide rails, and the test vehicle maintains contact with a customized road surface.

[0062] Figure 3 Schematic diagram of a hydraulic lifting suspension system in an embodiment of the present application. Figure 3 As shown, the test vehicle can use a hydraulic lifting suspension system, which is composed of an oil inlet, a suspension height adjuster, a spring, a damper, etc. Specifically, the suspension height can be changed by adjusting the oil cylinder in the suspension height adjuster through the oil inlet, so that the tires of the test vehicle in the vehicle test bench always maintain positive pressure with the road surface.

[0063] Specifically, in step 11, the test vehicle can be controlled to travel in the parallel slide rails, data during the travel process can be collected, and a training data set can be constructed. It should be noted that, considering the different deformation degrees of different road materials, the suspension adjustment under different road surfaces has certain differences. Therefore, the test vehicle can also be controlled to travel on customized road surfaces of different materials, such as asphalt, concrete, sand and gravel, etc., to collect data.

[0064] Among them, the training data set includes input data and corresponding output data. The input data includes vehicle status, geometric structure parameters, suspension parameters before adjustment and road surface information. The output data is the label corresponding to the input data, specifically the expected adjusted suspension parameters. In the embodiment of the present application, the suspension parameters after adjustment of the hydraulic lifting suspension system are taken.

[0065] The vehicle state may include the speed in all directions, the acceleration in all directions, the height of the vehicle's center of mass relative to the ground, the heading angle, the pitch angle, and the roll angle, which can be expressed as X = [v x ,v y ,v z ,a x ,ay ,a z ,h,θ,α,β], where v x ,v y ,v z are the speeds in the x, y, and z directions in the vehicle coordinate system, respectively, and a x ,a y ,a z are the accelerations in the x, y, and z directions in the vehicle coordinate system, respectively; h, θ, α, and β are the height, heading, pitch, and roll angles of the vehicle's center of mass relative to the ground, respectively.

[0066] The suspension parameters may include the height H of the four-wheel independent suspension t =[h rr ,h rl ,h fr ,h fl ], and the damping C of the four-wheel independent suspension t =[c rr ,c rl ,c fr ,c fl ]. The road surface information may include the road surface material, or the road surface information may include the road surface material and the predicted contact trajectory between the tire and the road surface.

[0067] Specifically, in step 12, a neural network model can be built. Exemplarily, the neural network model can adopt a three-layer structure, the first layer is a hidden layer, including 7 neurons, the second layer is a hidden layer, including 5 neurons, and the third layer is an output layer, including 2 neurons.

[0068] like Figure 4 As shown, Figure 4 Schematic diagram of a neural network model provided in an embodiment of the present application, the input of the neural network model is vehicle state, geometric structure parameters (wheelbase L and semi-axle length l), suspension parameters before adjustment (H t , C t ) and road surface information (road surface function parameter P ij , which can be understood as the predicted contact trajectory between the tire and the road), and the output is the desired suspension parameter (suspension height set H i , suspension damping set C i ).

[0069] In the embodiment of the present application, the data in the training data set can be normalized and screened. Normalization is to map the data to between 0 and 1 to ensure rapid convergence of the training model. Screening the data is to ensure the validity of the data and avoid contaminating the model. Further, the training data set can be divided into training samples, verification samples, and test samples, such as 70% as training samples, 15% as verification samples, and 15% as test samples.

[0070] Specifically, the neural network model can be trained using training samples. For example, the neural network model can be trained using the TensorFlow framework, with a learning rate of 0.001, a batch size of 100, and a training number of 500. Furthermore, the trained neural network model can be verified and tested using verification samples and test samples. If it fails, the neural network model can be trained again until the verification and test pass to obtain a suspension parameter adjustment model.

[0071] Through the above steps 11 and 12, the relationship between the vehicle suspension system and the road surface can be accurately modeled through the neural network model. Compared with the existing mathematical modeling methods, the neural network model can capture more relevant relationships and improve the accuracy and adaptability of the model. In addition, the construction of the training data set through the vehicle test bench can ensure the accuracy of the training data set, without the need to manually calibrate the suspension height, which can improve the efficiency of building the training data set. At the same time, collecting data on the vehicle test bench can ensure the accuracy and reliability of the data, improve the stability and reliability of the magic carpet function, and is not affected by environmental factors such as weather and light, thereby obtaining high-quality training data and improving the prediction accuracy and robustness of the model.

[0072] S120: Determine the current vehicle state, current suspension parameters, and current road surface information of the vehicle.

[0073] The current vehicle state may be the state of the vehicle at the current moment, including the speed in each direction, the acceleration in each direction, the height of the center of mass of the vehicle relative to the ground, the heading angle, the pitch angle, and the roll angle. The current suspension parameters may be the suspension parameters of the vehicle at the current moment, including the height of the four-wheel independent suspension and the damping of the four-wheel independent suspension.

[0074] The current road surface information may include road surface material, or the current road surface may include road surface material and a current predicted contact trajectory between the tire and the road surface, wherein the current predicted contact trajectory may be composed of contact points between the tire and the road surface from the current moment to multiple moments in the future.

[0075] In a specific implementation, the current road surface information includes a current predicted contact trajectory between a tire of the vehicle and the road surface, and determining the current road surface information of the vehicle includes the following steps:

[0076] Step 21, obtaining the current road surface point cloud data of the vehicle, and determining the current road surface fitting surface based on the current road surface point cloud data;

[0077] Step 22: Determine the predicted tire position of the vehicle based on the current vehicle state, and determine the current predicted contact trajectory between the tire of the vehicle and the road surface according to the current road surface fitting surface and the predicted tire position.

[0078] In step 21, the road surface information can be collected by the laser radar sensor mounted on the vehicle to obtain the current road surface point cloud data.

[0079] Considering that a large number of points in the current road surface point cloud data will affect the efficiency and accuracy of surface fitting, the current road surface point cloud data may also be pooled after being acquired.

[0080] In one example, after obtaining the current road surface point cloud data of the vehicle, the method further includes:

[0081] The current road surface point cloud data is divided into regional point cloud data corresponding to multiple road surface areas; for the regional point cloud data corresponding to each road surface area, based on the parameters of each point in the regional point cloud data, the parameter mean corresponding to the regional point cloud data is determined, and the parameters of each point in the regional point cloud data are updated according to the parameter mean.

[0082] In the embodiment of the present application, the average pooling method can be used to process the current road surface point cloud data. Specifically, the current road surface point cloud data can be divided into a plurality of equally divided road surface areas, and each road surface area has corresponding regional point cloud data.

[0083] Furthermore, in each road surface area, the parameter mean can be calculated based on the parameters of each point in the corresponding regional point cloud data, and then the parameter mean can be used to replace the parameters of each point in the regional point cloud data.

[0084]

[0085] In the formula, p i is the parameter of the i-th point in the road area after pooling, and n represents the road area R i The number of inliers, p n For R i points within.

[0086] Through the above example, the unordered point sequence scanned by the lidar can be first pooled into point clouds to avoid the influence of too many point clouds on the efficiency and accuracy of subsequent fitting.

[0087] Furthermore, in step 21, the road surface can be fitted by the current road surface point cloud data to obtain the current road surface fitting surface. For example, the x, y, z coordinates of each point in the current road surface point cloud data can be used to fit it into a three-dimensional surface using a spline to obtain the current road surface fitting surface, which can be expressed as:

[0088] Z = h(u(x,y),v(x,y));

[0089] Where u and v represent the direction vectors of the spline curve u and v.

[0090] For example, Figure 5 A schematic diagram of a fitting surface provided in an embodiment of the present application is shown as follows: Figure 5 As shown, the current road surface point cloud data can be first processed by average pooling, and then processed by surface fitting to obtain the current road surface fitting surface.

[0091] Furthermore, in step 22, the predicted tire position of the vehicle may be determined according to the current vehicle state, wherein the predicted tire position may be the predicted driving point of the tire from the current moment to multiple moments in the future.

[0092] Optionally, determining a predicted tire position of the vehicle based on the current vehicle state includes:

[0093] Based on the current speed, current acceleration and current heading angle in the current vehicle state, the current predicted driving trajectory of the vehicle is determined; and the predicted tire position of each tire in the vehicle is determined according to the current predicted driving trajectory of the vehicle.

[0094] Specifically, the driving trajectory of the vehicle may be predicted based on the current speed, current acceleration, and current heading angle in the current vehicle state to obtain the current predicted driving trajectory.

[0095] For example, the current speed, current acceleration and current heading angle can be input into a pre-trained trajectory prediction model to obtain the current predicted driving trajectory output by the trajectory prediction model, or the current predicted driving trajectory of the vehicle can be determined based on the current speed, current acceleration, current heading angle and vehicle kinematic model.

[0096] After the current predicted driving trajectory of the vehicle is obtained, the predicted tire position of each tire in the vehicle can be further obtained according to the wheelbase of the vehicle and the current predicted driving trajectory.

[0097] Through the above optional implementation, accurate prediction of the tire position can be achieved, which facilitates the subsequent accurate determination of the current predicted contact trajectory between the tire and the road surface.

[0098] Furthermore, in step 22, the predicted tire position may be projected onto the current road surface fitting surface, and then the current predicted contact trajectory between the vehicle's tire and the road surface may be determined based on each point obtained by the projection.

[0099] Exemplarily, the current predicted contact trajectory can be represented by a multi-segment Bezier curve:

[0100] B(t) i =(1-t) 3 P i0 +3t(1-t) 2 P i1 +3t 2 (1-t)P i2 +t 3 (1-t)P i3 ;

[0101] Where t∈[0,1], P ij represents the control node of the Bezier curve, i represents the Bezier segment sequence, B(t) i is a piecewise Bezier curve.

[0102] Through the above steps 21 and 22, the current road surface point cloud data can be fitted first, and then the predicted tire position can be projected onto the current road surface fitting surface obtained by fitting, so as to obtain the current predicted contact trajectory between the tire and the road surface, thereby ensuring the accuracy of the current predicted contact trajectory. Moreover, by inputting it into the model, the model can be combined with the current predicted contact trajectory to plan the suspension parameters according to the ups and downs of the road surface on which the tire is about to travel, thereby improving the accuracy of the suspension adjustment and being applicable to various types of road surfaces.

[0103] S130, inputting the geometric structure parameters, current road surface information, current suspension parameters and current vehicle state into a suspension parameter adjustment model to obtain target suspension parameters of the vehicle.

[0104] Specifically, the geometric structure parameters, current road surface information, current suspension parameters and current vehicle state can be input into the suspension parameter adjustment model to obtain the suspension parameter adjustment model. It should be noted that for different types of vehicle suspensions, the suspension parameter adjustment model can be used by simply adjusting the geometric structure parameters, so that the model can be reused.

[0105] In the embodiment of the present application, considering that some vehicles are already equipped with a mathematical model for suspension adjustment, in order to further improve the accuracy of suspension adjustment, the suspension adjustment can also be achieved by combining the mathematical model with the suspension parameter adjustment model.

[0106] Optionally, the method provided in the embodiment of the present application further includes: determining reference suspension parameters based on current road surface information, current suspension parameters and a vehicle suspension model of the vehicle, wherein the vehicle suspension model is a mathematical model that describes the relationship between the road surface and the suspension system;

[0107] Correspondingly, after obtaining the target suspension parameters of the vehicle, the method further includes: updating the target suspension parameters based on the reference suspension parameters.

[0108] Specifically, the current road surface information and current suspension parameters can be used as inputs and substituted into the vehicle suspension model for calculation to obtain reference suspension parameters determined based on the mathematical model, wherein the reference suspension parameters are the parameters that the suspension system is expected to be adjusted to.

[0109] Furthermore, after the target suspension parameters are obtained based on the suspension parameter adjustment model, the reference suspension parameters may be fused with the target suspension parameters, and the fusion results may be used as new target suspension parameters.

[0110] Exemplarily, a corresponding weight can be assigned to the vehicle suspension model in advance, and a corresponding weight can be assigned to the suspension parameter adjustment model. Then, the reference suspension parameters and the target suspension parameters can be weightedly calculated using the weight corresponding to the vehicle suspension model and the weight corresponding to the suspension parameter adjustment model to obtain new target suspension parameters.

[0111] Through the above implementation, after the target suspension parameters are obtained, the target suspension parameters can be optimized in combination with the reference suspension parameters calculated by the mathematical model, thereby further improving the accuracy of the target suspension parameters.

[0112] S140: Control the suspension system of the vehicle according to the target suspension parameters.

[0113] The target suspension parameters may include a target height and a target damping of the suspension system.

[0114] Specifically, the target suspension parameters may be sent to a controller of the suspension system, and the controller may execute control quantities according to the target suspension parameters to control the suspension system and adjust the height of the suspension system and / or the damping of the suspension system.

[0115] For example, Figure 6 A schematic diagram of a magic carpet function provided in an embodiment of the present application is shown in FIG. Figure 6 As shown, the chassis reference height and tire reference height can be determined before suspension adjustment, and then at the vehicle position at time t1, the suspension change amount C can be determined according to the target suspension parameters. t1 , and then execute C t1 At the vehicle position at time t2, the suspension change C can be determined based on the target suspension parameters. t2 , and then execute Ct2 .

[0116] Considering that if the gap between the target suspension parameters and the current suspension parameters is too large, the target suspension parameters may not be reached in a short period of time due to the limitations of the suspension system structure, or rapid adjustment to the target suspension parameters in a short period of time may cause damage to the suspension system, therefore, the adjustment rate of the suspension system can also be limited.

[0117] In addition, considering that if the target suspension parameter is greater than the threshold that the suspension system can reach, adjustment to the target suspension parameter will fail, and an error prompt message of the magic carpet function may appear, the target suspension parameter can also be limited.

[0118] In a specific implementation, before controlling the suspension system of the vehicle according to the target suspension parameters, the method further includes:

[0119] Determine whether the target suspension parameter is greater than a preset parameter threshold, and if so, update the target suspension parameter according to the preset parameter threshold;

[0120] And / or, determining a parameter difference between the target suspension parameter and the current suspension parameter, and if the parameter difference is greater than a preset parameter difference threshold, updating the target suspension parameter according to the preset parameter difference threshold.

[0121] The preset parameter threshold may be the maximum parameter that the suspension system can reach, i.e., the maximum height and the maximum damping. The preset parameter difference threshold may be the maximum parameter difference that the suspension system can adjust within a certain period of time, i.e., the maximum height difference and the maximum damping difference.

[0122] Specifically, before controlling the suspension system of the vehicle according to the target suspension parameters, it can be determined whether the target suspension parameters are greater than a preset parameter threshold. If so, it means that the suspension system cannot reach the target suspension parameters. At this time, the target suspension parameters can be updated according to the preset parameter threshold, such as updating the target suspension parameters to the preset parameter threshold.

[0123] In addition, before controlling the suspension system of the vehicle according to the target suspension parameters, the parameter difference between the target suspension parameters and the current suspension parameters, such as height difference and damping difference, can be calculated first, and it can be determined whether the parameter difference is greater than a preset parameter difference threshold. If so, it means that the suspension system cannot reach the target suspension parameters or reaching the target suspension parameters will cause certain damage. At this time, the target suspension parameters can be updated according to the preset parameter difference threshold, for example, the target suspension parameters can be updated to the sum of the current suspension parameters and the preset parameter difference threshold.

[0124] Through the above implementation, the target suspension parameters can be limited before the suspension system is controlled according to the target suspension parameters, thereby avoiding adjustment failure caused by excessively large target suspension parameters and avoiding suspension damage caused by excessively large differences between the target suspension parameters and the current suspension parameters.

[0125] In the embodiment of the present application, considering that after the vehicle has traveled for a period of time or a certain mileage, the suspension system may undergo certain changes due to wear and tear, resulting in reduced applicability of the suspension parameter adjustment model, the suspension parameter adjustment model can also be evaluated and optimized during the process of the vehicle using the suspension parameter adjustment model.

[0126] In a specific implementation, after controlling the suspension system of the vehicle according to the target suspension parameters, the following steps are further included:

[0127] Step 31, obtaining the vehicle body posture;

[0128] Step 32: construct a posture change curve of the vehicle according to each body posture, and determine whether the vehicle meets the model optimization condition based on the change rate of the posture change curve;

[0129] Step 33: If yes, then optimize the suspension parameter adjustment model.

[0130] Specifically, the vehicle body posture may be acquired after the suspension system of the vehicle is controlled according to the target suspension parameters, and then a posture change curve of the vehicle may be constructed according to each body posture.

[0131] Furthermore, the rate of change of posture can be calculated based on the posture change curve. If the rate of change of posture is greater than the preset rate threshold, it means that the body posture has changed significantly and there may be certain errors in the adjustment of the suspension system. At this time, it can be determined that the vehicle meets the model optimization conditions. Further, the suspension parameter adjustment model can be optimized and trained.

[0132] Exemplarily, abnormal postures can be screened out from the posture change curve, and then the vehicle state, suspension parameters, road surface information, etc. corresponding to the abnormal postures can be used as negative samples and input into the suspension parameter adjustment model for optimization training to achieve the purpose of updating the model parameters in the suspension parameter adjustment model.

[0133] Through the above steps 31 to 33, the suspension parameter adjustment model can be evaluated and optimized during the process of the vehicle using the suspension parameter adjustment model to adjust the suspension system, ensuring that the suspension parameter adjustment model can always be applied to the vehicle, thereby improving the accuracy and robustness of the vehicle suspension adjustment.

[0134] The vehicle suspension control method provided in the embodiment of the present application obtains the geometric structure parameters and the suspension parameter adjustment model of the vehicle, and determines the current vehicle state, current suspension parameters and current road surface information of the vehicle, and then inputs the geometric structure parameters, current road surface information, current suspension parameters and the current vehicle state into the suspension parameter adjustment model to obtain the target suspension parameters of the vehicle, so as to control the suspension system of the vehicle according to the target suspension parameters, and realizes suspension adjustment based on a pre-trained neural network model, wherein the suspension parameter adjustment model can be applicable to various vehicle models. For different types of vehicle suspensions, the same model can be used for suspension adjustment by simply inputting the geometric structure parameters of the vehicle into the model, thereby improving the versatility and applicability of the model and reducing the complexity and cost of model design. In addition, the method determines the target suspension parameters by combining the current road surface information, the current vehicle state and the current suspension parameters, so that the model can better adapt to different road conditions, improve the adaptability and adjustment effect of suspension adjustment, and at the same time, make the model more accurate and be able to more accurately adjust the suspension system, thereby reducing body movement, eliminating and reducing vehicle bumps and tilts, and improving driving comfort.

[0135] The vehicle suspension control method provided in the embodiment of the present application has the following technical effects:

[0136] 1. Improve model accuracy and adaptability: The embodiment of the present application uses neural network technology to build the magic carpet function. By collecting training data on the test bench, the relationship between the vehicle suspension system and the road surface can be modeled more accurately. Compared with the existing mathematical modeling method, the neural network model can capture more relevant relationships and improve the accuracy and adaptability of the model;

[0137] 2. Simplify model design and verification: The bench data acquisition solution and platform model training solution of the embodiment of the present application make the establishment of different types of vehicle suspension models reusable, greatly simplifying the complexity and cost of model design and verification. At the same time, due to the self-learning ability of the neural network model, the model verification process is also simpler and more efficient;

[0138] 3. Improve road adaptability: In the model training scheme of the embodiment of the present application, the system input includes polynomial tire predicted contact trajectory, road material, current suspension height, damping, vehicle wheelbase, center of mass, half-axle length, vehicle speed, acceleration, etc. These parameters enable the model to better adapt to different road conditions and improve the adaptability and adjustment effect of the magic carpet function;

[0139] 4. Improve driving comfort: Since the model of the embodiment of the present application is more accurate and adaptable, the height and damping of the suspension can be adjusted more accurately, thereby reducing body movement, eliminating and reducing bumps and tilts, and improving driving comfort.

[0140] In summary, the method provided in the embodiment of the present application, compared with the prior art, not only improves the accuracy and adaptability of the model, simplifies model design and verification, improves road adaptability and driving comfort, but also has higher practical value in actual applications.

[0141] The method provided by the embodiment of the present application can be widely used in the fields of automobile manufacturing, intelligent transportation system, smart home, etc. In the field of automobile manufacturing, the method provided by the embodiment of the present application, based on the magic carpet function of the neural network, can significantly improve the driving comfort. By accurately modeling the relationship between the vehicle suspension system and the road surface, the height and damping of the suspension can be accurately adjusted, thereby reducing the movement of the vehicle body, eliminating and reducing bumps and tilts. In addition, due to the use of neural networks, different vehicle suspension models can be quickly adapted without separate design and verification, which greatly reduces the complexity and cost of design and verification. In the field of intelligent transportation systems, the method provided by the embodiment of the present application can improve the safety and efficiency of traffic. By accurately sensing and responding to road conditions, the suspension system of the vehicle can be adjusted in advance, thereby reducing the occurrence of traffic accidents. In the field of smart home, the method provided by the embodiment of the present application can be used for intelligent home transportation tools, such as smart electric vehicles and smart scooters, etc., and automatic adjustment of these transportation tools can be achieved through accurate perception and response to road conditions, thereby improving driving comfort and safety.

[0142] Figure 7 The schematic diagram of the structure of a vehicle suspension control device provided in an embodiment of the present application includes a model acquisition module 710 , an input determination module 720 , a model output module 730 and a suspension control module 740 .

[0143] A model acquisition module 710, used to acquire geometric structure parameters of the vehicle and a suspension parameter adjustment model, wherein the suspension parameter adjustment model is a pre-trained neural network model;

[0144] An input determination module 720 is used to determine the current vehicle state, current suspension parameters and current road surface information of the vehicle;

[0145] A model output module 730, for inputting the geometric structure parameters, the current road surface information, the current suspension parameters and the current vehicle state into the suspension parameter adjustment model to obtain target suspension parameters of the vehicle;

[0146] The suspension control module 740 is used to control the suspension system of the vehicle according to the target suspension parameters.

[0147] Based on the above-mentioned embodiments, optionally, the input determination module 720 is also used to obtain current road surface point cloud data of the vehicle, determine the current road surface fitting surface based on the current road surface point cloud data; determine the predicted tire position of the vehicle based on the current vehicle state, and determine the current predicted contact trajectory between the tire of the vehicle and the road surface according to the current road surface fitting surface and the predicted tire position.

[0148] On the basis of the above-mentioned embodiments, optionally, the input determination module 720 is also used to divide the current road surface point cloud data into regional point cloud data corresponding to multiple road surface areas; for each regional point cloud data corresponding to the road surface area, based on the parameters of each point in the regional point cloud data, the parameter mean corresponding to the regional point cloud data is determined, and the parameters of each point in the regional point cloud data are updated according to the parameter mean.

[0149] Based on the above-mentioned embodiments, optionally, the input determination module 720 is also used to determine the current predicted driving trajectory of the vehicle based on the current speed, current acceleration and current heading angle in the current vehicle state; and determine the predicted tire position of each tire in the vehicle according to the current predicted driving trajectory of the vehicle.

[0150] On the basis of the above-mentioned embodiments, optionally, the device further includes a model training module, the model training module being used to collect a training data set based on a pre-built vehicle test bench, wherein the vehicle test bench is composed of a test vehicle fixed in parallel slide rails, the suspension height of the test vehicle is adjusted by the oil volume in the cylinder in the suspension height adjuster, and the test vehicle is always in contact with the road surface; constructing a neural network model, training the neural network model based on the training data set, and obtaining the suspension parameter adjustment model.

[0151] Based on the above-mentioned embodiments, optionally, the device also includes a model optimization module, which is used to obtain the body posture of the vehicle; construct a posture change curve of the vehicle according to each body posture, and judge whether the vehicle meets the model optimization conditions based on the change rate of the posture change curve; if so, optimize the suspension parameter adjustment model.

[0152] Based on the above-mentioned embodiments, optionally, the model output module 730 is also used to determine reference suspension parameters based on the current road surface information, the current suspension parameters and the vehicle suspension model of the vehicle, wherein the vehicle suspension model is a mathematical model that describes the relationship between the road surface and the suspension system; and update the target suspension parameters based on the reference suspension parameters.

[0153] On the basis of the above-mentioned embodiments, optionally, the suspension control module 740 is also used to determine whether the target suspension parameter is greater than a preset parameter threshold, and if so, update the target suspension parameter according to the preset parameter threshold; and / or determine the parameter difference between the target suspension parameter and the current suspension parameter, and if the parameter difference is greater than the preset parameter difference threshold, update the target suspension parameter according to the preset parameter difference threshold.

[0154] The vehicle suspension control device provided in the embodiment of the present application can execute the steps in the vehicle suspension control method provided in the method embodiment of the present application, and the execution steps and beneficial effects are no longer repeated here.

[0155] Figure 8 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Figure 8 As shown, the electronic device 400 includes one or more processors 401 and a memory 402 .

[0156] The processor 401 may be a central processing unit (CPU) or other forms of processing units having data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device 400 to perform desired functions.

[0157] The memory 402 may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory (cache), etc. The non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor 401 may run the program instructions to implement the vehicle suspension control method of any embodiment of the present application described above and / or other desired functions. Various contents such as initial external parameters, thresholds, etc. may also be stored in the computer-readable storage medium.

[0158] In one example, the electronic device 400 may further include: an input device 403 and an output device 404, which are interconnected via a bus system and / or other forms of connection mechanisms (not shown). The input device 403 may include, for example, a keyboard, a mouse, etc. The output device 404 may output various information to the outside, including early warning information, braking force, etc. The output device 404 may include, for example, a display, a speaker, a printer, a communication network and a remote output device connected thereto, etc.

[0159] Of course, to simplify, Figure 8Only some of the components related to the present application in the electronic device 400 are shown, and components such as a bus, an input / output interface, etc. are omitted. In addition, according to specific application conditions, the electronic device 400 may also include any other appropriate components.

[0160] In addition to the above-mentioned methods and devices, an embodiment of the present application may also be a computer program product, which includes computer program instructions, which, when executed by a processor, enable the processor to execute the steps of the vehicle suspension control method provided in any embodiment of the present application.

[0161] The computer program product may be written in any combination of one or more programming languages ​​to write program codes for performing the operations of the embodiments of the present application, including object-oriented programming languages, such as Java, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as an independent software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0162] In addition, an embodiment of the present application may also be a computer-readable storage medium on which computer program instructions are stored. When the computer program instructions are executed by a processor, the processor executes the steps of the vehicle suspension control method provided in any embodiment of the present application.

[0163] The computer readable storage medium can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can include, for example, but is not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0164] It should be noted that the terms used in this application are only for describing specific embodiments, rather than limiting the scope of this application. As shown in the specification and claims of this application, unless the context clearly indicates an exception, the words "one", "a", "a kind of" and / or "the" do not specifically refer to the singular, but may also include the plural. The terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method or device. In the absence of further restrictions, the elements defined by the sentence "including one..." do not exclude the presence of other identical elements in the process, method or device including the elements.

[0165] It should also be noted that the terms "center", "up", "down", "left", "right", "vertical", "horizontal", "inside", "outside", etc., indicating the orientation or position relationship, are based on the orientation or position relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present application. Unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", etc. should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be an indirect connection through an intermediate medium, or it can be a connection between the two elements. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0166] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and its core ideas of this application. The above is only the preferred implementation method of this application. It should be pointed out that due to the limitations of textual expression and the objective existence of infinite specific structures, ordinary technicians in this technical field can make several improvements, modifications or changes without departing from the principles of this application, and can also combine the above technical features in an appropriate manner; these improvements, modifications, changes or combinations, or the direct application of the concept and technical solution of the invention to other occasions without improvement, should be regarded as the scope of protection of this application.

Claims

1. A vehicle suspension control method, characterized in that: include: Acquiring geometric structure parameters of the vehicle and a suspension parameter adjustment model, wherein the suspension parameter adjustment model is a pre-trained neural network model; Determining a current vehicle state, current suspension parameters, and current road surface information of the vehicle; Inputting the geometric structure parameters, the current road surface information, the current suspension parameters and the current vehicle state into the suspension parameter adjustment model to obtain the target suspension parameters of the vehicle; A suspension system of the vehicle is controlled according to the target suspension parameters.

2. The method according to claim 1, characterized in that The current road surface information includes a current predicted contact trajectory between the tire of the vehicle and the road surface, and determining the current road surface information of the vehicle includes: Acquiring current road surface point cloud data of the vehicle, and determining a current road surface fitting surface based on the current road surface point cloud data; The predicted tire position of the vehicle is determined based on the current vehicle state, and the current predicted contact trajectory between the tire of the vehicle and the road surface is determined according to the current road surface fitting surface and the predicted tire position.

3. The method according to claim 2, characterized in that After obtaining the current road surface point cloud data of the vehicle, the method further includes: Dividing the current road surface point cloud data into regional point cloud data corresponding to a plurality of road surface regions; For each of the regional point cloud data corresponding to the road surface area, a parameter mean corresponding to the regional point cloud data is determined based on the parameters of each point in the regional point cloud data, and the parameters of each point in the regional point cloud data are updated according to the parameter mean.

4. The method according to claim 2, characterized in that: Determining a predicted tire position of the vehicle based on the current vehicle state includes: Determining a current predicted driving trajectory of the vehicle based on a current speed, a current acceleration, and a current heading angle in the current vehicle state; A predicted tire position of each tire in the vehicle is determined based on the current predicted driving trajectory of the vehicle.

5. The method according to claim 1, characterized in that The training process of the suspension parameter adjustment model includes: A training data set is collected based on a pre-built vehicle test bench, wherein the vehicle test bench is composed of a test vehicle fixed in parallel slide rails, the suspension height of the test vehicle is adjusted by the oil volume of the oil cylinder in the suspension height adjuster, and the test vehicle is always in contact with the road surface; A neural network model is constructed, and the neural network model is trained based on the training data set to obtain the suspension parameter adjustment model.

6. The method according to claim 1, characterized in that After controlling the suspension system of the vehicle according to the target suspension parameters, the method further includes: Acquiring a body posture of the vehicle; constructing a posture change curve of the vehicle according to each vehicle body posture, and judging whether the vehicle meets the model optimization condition based on the change rate of the posture change curve; If so, the suspension parameter adjustment model is optimized and trained.

7. The method according to claim 1, characterized in that The method further comprises: Determining reference suspension parameters based on the current road surface information, the current suspension parameters, and a vehicle suspension model of the vehicle, wherein the vehicle suspension model is a mathematical model describing a relationship between a road surface and a suspension system; Accordingly, after obtaining the target suspension parameters of the vehicle, the method further includes: The target suspension parameters are updated based on the reference suspension parameters.

8. The method according to claim 1, characterized in that Before controlling the suspension system of the vehicle according to the target suspension parameters, the method further includes: determining whether the target suspension parameter is greater than a preset parameter threshold, and if so, updating the target suspension parameter according to the preset parameter threshold; and / or, A parameter difference between the target suspension parameter and the current suspension parameter is determined, and if the parameter difference is greater than a preset parameter difference threshold, the target suspension parameter is updated according to the preset parameter difference threshold.

9. An electronic device, characterized in that: The electronic device comprises: Processor and memory; The processor is used to execute the steps of the vehicle suspension control method according to any one of claims 1 to 8 by calling the program or instruction stored in the memory.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a program or instruction that causes a computer to execute the steps of the vehicle suspension control method according to any one of claims 1 to 8.

Images