Road feeling simulation data determination method and device, controller, vehicle and medium

Abstract

The embodiment of the invention discloses a road feeling simulation data determination method and device, a controller, a vehicle and a medium. The method comprises the steps of predicting a first steering torque of a target vehicle through a road feeling simulation model based on vehicle state information of the target vehicle; calculating a second steering torque of the target vehicle based on the vehicle dynamics model; and according to the first steering torque and the second steering torque, determining a target steering wheel torque so as to perform road feeling simulation control on the target vehicle based on the target steering wheel torque. Therefore, the first steering torque of the target vehicle is predicted based on the vehicle state information through the road feeling simulation model, and the second steering torque of the target vehicle is calculated based on the vehicle dynamics model, so that the target steering wheel torque for performing road feeling simulation on the vehicle is determined by combining the first steering torque and the second steering torque; the road feeling simulation can be stably and accurately carried out in a complex driving scene, so that the reliability of the road feeling simulation is improved.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a method, apparatus, controller, vehicle, and medium for determining road sense simulation data. Background Technology

[0002] With the development of electric and intelligent vehicles, X-by-Wire technology has received increasing attention in recent years, accompanying the advancement of automotive electronics. As a novel steering system, X-by-Wire greatly promotes vehicle intelligence and ease of handling. However, because it eliminates the mechanical connection between the steering wheel and the steering assembly, road resistance cannot be directly fed back to the steering wheel. Therefore, it is necessary to simulate the driver's "road feel" to ensure normal and safe driving.

[0003] However, during the research and practice of existing technologies, it was found that existing methods for determining road feel simulation data cannot stably and accurately simulate road feel in complex driving scenarios, resulting in low reliability of road feel simulation. Summary of the Invention

[0004] This application provides a method, apparatus, controller, vehicle, storage medium, and product for determining road sense simulation data, which can stably and accurately perform road sense simulation and improve the reliability of road sense simulation.

[0005] To achieve the above objectives, according to a first aspect of this application, a method for determining road feel simulation data is provided, the method comprising:

[0006] Based on the vehicle state information of the target vehicle, the first steering torque of the target vehicle is predicted through a road feel simulation model;

[0007] Calculate the second steering torque of the target vehicle based on the vehicle dynamics model;

[0008] Based on the first steering torque and the second steering torque, a target steering wheel torque is determined to perform road feel simulation control on the target vehicle based on the target steering wheel torque.

[0009] According to a second aspect of this application, a method for determining road feel simulation data is provided, the method comprising:

[0010] Obtain the vehicle status information of the target vehicle;

[0011] Based on the vehicle state information, the steering torque is predicted using a road feel simulation model to obtain the first steering torque of the target vehicle.

[0012] According to a third aspect of this application, a road feel simulation data determination device is provided, the device comprising:

[0013] The prediction module is used to predict the first steering torque of the target vehicle based on the vehicle state information of the target vehicle through a road feel simulation model;

[0014] The calculation module is used to calculate the second steering torque of the target vehicle based on the vehicle dynamics model;

[0015] The determination module is used to determine the target steering wheel torque based on the first steering torque and the second steering torque, so as to perform road feel simulation control on the target vehicle based on the target steering wheel torque.

[0016] According to a fourth aspect of this application, a road feel simulation data determination device is provided, the device comprising:

[0017] The information acquisition module is used to acquire the vehicle status information of the target vehicle.

[0018] The torque prediction module is used to predict the steering torque based on the vehicle state information using a road feel simulation model, and to obtain the first steering torque of the target vehicle.

[0019] According to a fifth aspect of this application, a controller is provided, including a processor and a memory, the memory storing an application program, and the processor being configured to run the application program in the memory to implement the road sense simulation data determination method provided in the embodiments of this application.

[0020] According to a sixth aspect of this application, a vehicle is provided, the vehicle including the controller provided in the third aspect of this application.

[0021] According to a seventh aspect of this application, a computer-readable storage medium is provided, the computer-readable storage medium storing a computer program adapted for loading by a processor to perform the steps in any of the road sense simulation data determination methods provided in the embodiments of this application.

[0022] According to an eighth aspect of this application, a computer program product is provided, the computer program product including a computer program stored in a computer-readable storage medium; when a processor of a controller reads the computer program from the computer-readable storage medium, the processor executes the computer program, causing the controller to perform the steps in the road sense simulation data determination method provided in the embodiments of this application.

[0023] In the road feel simulation data determination method, apparatus, controller, vehicle, storage medium, and product of this application, a first steering torque of the target vehicle is predicted based on the vehicle state information of the target vehicle using a road feel simulation model; a second steering torque of the target vehicle is calculated based on a vehicle dynamics model; and a target steering wheel torque is determined based on the first and second steering torques, so as to perform road feel simulation control on the target vehicle based on the target steering wheel torque. Thus, by predicting the first steering torque of the target vehicle based on vehicle state information using a road feel simulation model, and calculating the second steering torque of the target vehicle based on a vehicle dynamics model, and combining the first and second steering torques, the target steering wheel torque used for road feel simulation of the vehicle can be determined. This enables stable and accurate road feel simulation in complex driving scenarios, thereby improving the reliability of road feel simulation. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying 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.

[0025] Figure 1 This is a schematic diagram illustrating an implementation scenario of a method for determining road sense simulation data provided in this application.

[0026] Figure 2 This is a flowchart illustrating a method for determining road sense simulation data provided in an embodiment of this application;

[0027] Figure 3 This is a schematic diagram of the specific architecture of a method for determining road sense simulation data provided in an embodiment of this application;

[0028] Figure 4 This is another flowchart illustrating a method for determining road sense simulation data provided in an embodiment of this application;

[0029] Figure 5 This is a schematic diagram of the model structure of a method for determining road sense simulation data provided in an embodiment of this application;

[0030] Figure 6 This is a schematic diagram of another model structure of a method for determining road sense simulation data provided in an embodiment of this application;

[0031] Figure 7 This is a schematic flowchart illustrating a method for determining road sense simulation data provided in an embodiment of this application.

[0032] Figure 8 This is a schematic diagram of the road sense simulation data determination device provided in the embodiments of this application;

[0033] Figure 9 This is another structural schematic diagram of the road sense simulation data determination device provided in the embodiments of this application;

[0034] Figure 10 This is a schematic diagram of the controller provided in the embodiments of this application. Detailed Implementation

[0035] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the protection scope of this application.

[0036] In the description of this application, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more features. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0037] This application provides a method, apparatus, controller, vehicle, storage medium, and product for determining road sense simulation data. The road sense simulation data determination apparatus can be integrated into a controller, which can be applied to a server or a terminal device.

[0038] The server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery network (CDN) acceleration services, and big data and artificial intelligence platforms. The terminal can include, but is not limited to, mobile phones, computers, smart voice interaction devices, smart home appliances, vehicle terminals, and aircraft. The terminal and server can be directly or indirectly connected via wired or wireless communication, which is not limited herein.

[0039] Optionally, the controller can be integrated into a vehicle, which can be a gasoline-powered vehicle, a plug-in hybrid electric vehicle, or a new energy vehicle, etc., and this application does not specifically limit it.

[0040] Please see Figure 1 Taking the road sense simulation data determination device integrated into the controller as an example, Figure 1This is a schematic diagram of an implementation scenario for the road feel simulation data determination method provided in this application. The controller can be integrated into a vehicle. The controller can predict the first steering torque of the target vehicle based on the vehicle state information of the target vehicle through a road feel simulation model; calculate the second steering torque of the target vehicle based on a vehicle dynamics model; and determine the target steering wheel torque based on the first steering torque and the second steering torque, so as to perform road feel simulation control on the target vehicle based on the target steering wheel torque.

[0041] It should be noted that, Figure 1 The schematic diagram illustrating the implementation environment of the road sense simulation data determination method is merely an example. The implementation environment of the road sense simulation data determination method described in this application is for the purpose of more clearly illustrating the technical solutions of this application and does not constitute a limitation on the technical solutions provided in this application. Those skilled in the art will understand that, with the evolution of road sense simulation data determination and the emergence of new business scenarios, the technical solutions provided in this application are also applicable to similar technical problems.

[0042] The solutions provided in this application are specifically illustrated through the following embodiments. It should be noted that the order of description of the following embodiments is not intended to limit the preferred order of the embodiments.

[0043] This embodiment will be described from the perspective of a road sense simulation data determination device, which can be integrated into a controller.

[0044] Please see Figure 2 , Figure 2 This is a flowchart illustrating the method for determining road sense simulation data provided in an embodiment of this application. The method for determining road sense simulation data includes:

[0045] Step S101: Based on the vehicle state information of the target vehicle, predict the first steering torque of the target vehicle through a road feel simulation model.

[0046] The target vehicle can be a vehicle used for road feel simulation, and the vehicle state information can be information describing the driving state of the target vehicle. The road feel simulation model can be a deep learning model used to predict the vehicle's road feel torque, which can be the road feedback torque simulated in the steer-by-wire system, used to simulate the steering torque of the vehicle's road feel during driving. The first steering torque can be the road feel torque predicted by the road feel simulation model.

[0047] Optionally, the road feel simulation model can be a deep learning model, for example, a fully connected neural network model based on a multilayer perceptron (MLP).

[0048] Road feel simulation refers to the use of a motor in a vehicle's steer-by-wire (SBW) system to simulate steering resistance corresponding to actual road conditions, allowing the driver to experience "force feedback" similar to that of traditional mechanical steering when operating the steering wheel. Since SBW eliminates the mechanical connection between the steering wheel and the steering assembly, road resistance cannot be directly fed back to the steering wheel. Therefore, to enable the driver to perceive the vehicle's driving status and ensure normal and safe driving, road resistance during vehicle movement can be fed back to the steering wheel in the form of steering torque, thus achieving road feel simulation.

[0049] Optionally, the vehicle status information may include at least one of the following: steering wheel angle, steering wheel angular velocity, steering wheel torque, vehicle speed, yaw rate, lateral acceleration, vertical load, and front wheel angle.

[0050] In one specific embodiment, the data acquisition for the road feel simulation model requires careful selection of input variables. On the one hand, it must ensure that the input variables are meaningful for the model's predictions; that is, irrelevant variables should not be selected. On the other hand, they must fully represent the impact of the vehicle's state on the steering torque output; that is, too few variables should be selected, as insufficient representation of the vehicle's state can indeed limit the model's predictive capabilities and lack universality. Therefore, based on experience with traditional dynamics models, the input variables include steering wheel angle, steering wheel angular velocity, vehicle speed, vehicle yaw rate, steering wheel torque, vehicle lateral acceleration, vehicle vertical load, and front wheel steering angle. In other words, the vehicle state information can include the target vehicle's steering wheel angle, steering wheel angular velocity, steering wheel torque, vehicle speed, yaw rate, lateral acceleration, vertical load, and front wheel angle. The road feel torque is related to the interaction between the vehicle's front wheels and the ground. Therefore, based on the rich vehicle state information, the road feel simulation model can obtain vehicle state parameters related to the vehicle's front wheels, thereby more accurately obtaining the road feel information generated during the simulation of the target vehicle's driving process, achieving more accurate prediction of the target vehicle's road feel torque, and improving the reliability of road feel simulation.

[0051] Step S102: Calculate the second steering torque of the target vehicle based on the vehicle dynamics model.

[0052] The vehicle dynamics model can be a mathematical model describing the dynamic behavior of a vehicle. It is based on the principles of mechanics, kinematics, and dynamics, treating the vehicle as a system and describing it mathematically. The second steering torque can be a steering torque calculated based on the vehicle dynamics model, used to simulate the road feel generated when driving the target vehicle.

[0053] There are several ways to calculate the second steering torque of the target vehicle based on the vehicle dynamics model. For example, the steering-related torque associated with the steering torque of the target vehicle can be calculated based on the vehicle state information through the vehicle dynamics model; and the second steering torque of the target vehicle can be determined based on the steering-related torque.

[0054] The steering-related torque can be the steering torque associated with road feel simulation.

[0055] Optionally, the steering-related torque may include at least one of the following: the torque generated by the stiffness of the steering system of the target vehicle, the torque generated by the system damping, the torque generated by the system equivalent damping, the system friction torque, and the tire return torque.

[0056] There are several ways to determine the second steering torque of the target vehicle based on the steering-related torque. For example, the steering-related torques can be summed directly to obtain the second steering torque of the target vehicle. Alternatively, a corresponding weight can be set for each steering-related torque according to the actual situation, so that the steering-related torques can be weighted and summed to obtain the second steering torque of the target vehicle.

[0057] In one specific embodiment, the steering-related torque may include the torque generated by the stiffness of the target vehicle's steering system, the torque generated by the system damping, the torque generated by the system's equivalent damping, the system friction torque, and the tire self-centering torque. A steering torque calculation model can be used to determine the second steering torque of the target vehicle based on the steering-related torque. The calculation formula for this steering torque calculation model is as follows:

[0058]

[0059] Among them, M feedback The calculated second steering torque, M, can be represented. stiffness The torque M generated by the stiffness of the steering system damping M is the torque generated by the system damping. inertia The torque generated by the system's equivalent damping can be obtained from bogie bench tests. Its proportion in the second steering torque is small, thus its impact on the system is relatively minor. M friction The system friction torque describes the effect of friction in the steering system on the second steering torque. It can be calculated using the Coulomb friction torque model and smoothed using the hyperbolic tangent function (tanh). In the item, W is the tire return torque. a This is the steering factor. The magnitude of the second steering torque can be calculated in real time using a semi-empirical formula, bench test parameters, and real-time vehicle status information.

[0060] Step S103: Determine the target steering wheel torque based on the first steering torque and the second steering torque, so as to perform road feel simulation control on the target vehicle based on the target steering wheel torque.

[0061] The target steering wheel torque can be the steering torque fed back to the steering wheel based on the road feel simulation results, used to simulate the road feel of the target vehicle being driven.

[0062] There are several ways to determine the target steering wheel torque based on the first steering torque and the second steering torque. For example, the target steering torque can be determined based on the first steering torque and the second steering torque, and the target steering wheel torque can be calculated based on the target steering torque.

[0063] The target steering torque can be the road feel steering torque determined based on the first steering torque and the second steering torque, and the target steering wheel torque is the steering torque that needs to be applied to the steering wheel of the target vehicle to simulate the road feel of the target vehicle currently being driven.

[0064] There are several ways to determine the target steering torque based on the first steering torque and the second steering torque. For example, the torque difference between the first steering torque and the second steering torque can be calculated; the target steering torque can be determined based on the torque difference and a preset torque difference threshold.

[0065] The torque difference can be the difference between the first steering torque and the second steering torque. The preset torque difference threshold can be a pre-set torque difference threshold, which can be used to measure the reliability of the first steering torque.

[0066] There are several ways to determine the target steering torque based on the torque difference and the preset torque difference threshold. For example, if the torque difference is not greater than the preset torque difference threshold, the first steering torque can be determined as the target steering torque; if the torque difference is greater than the preset torque difference threshold, the second steering torque can be determined as the target steering torque.

[0067] While the second steering torque calculated based on the vehicle dynamics model is not precise enough, it is unlikely to produce significant errors. The first steering torque predicted by the road feel simulation model is more accurate. However, predictions based on deep learning models suffer from instability and poor interpretability, potentially leading to anomalies. Therefore, combining the vehicle dynamics model and the road feel simulation model can ensure accurate and stable road feel simulation. For example, the second steering torque calculated based on the vehicle dynamics model can be used to assist in verifying the first steering torque predicted by the road feel simulation model. This allows for the assessment of the rationality and safety of the first steering torque output by the road feel simulation model, thereby addressing the difference between the first and second steering torques. When the distance is too large, it can be determined that the first steering torque predicted by the road feel simulation model is unreasonable. Therefore, the second steering torque calculated based on the vehicle dynamics model can be determined as the final target steering torque. When the difference between the first steering torque and the second steering torque is not large, it can be determined that the first steering torque predicted by the road feel simulation model is reasonable. Therefore, the first steering torque calculated based on the road feel simulation model can be determined as the final target steering torque. Thus, the embodiments of this application implement a road feel feedback control strategy based on a deep learning model and verify it through a vehicle dynamics model to form a dual-redundant road feel feedback control strategy. This strategy can achieve stable and accurate road feel simulation in complex driving scenarios, thereby improving the reliability of road feel simulation.

[0068] Optionally, there are several ways to calculate the target steering wheel torque based on the target steering torque. For example, the driver can apply a certain operating torque to the steering wheel, which is transmitted to the torsion bar through the steering shaft. The torque sensor measures the operating torque and direction applied by the driver to the steering wheel. Alternatively, the real-time load can be estimated through a road feel simulation model. At the same time, the vehicle speed sensor can measure the current vehicle speed, and then transmit it along with the power signal, vehicle status signal, and motor position signal to the corresponding electronic control unit (ECU) in the vehicle. The ECU calculates the ideal target steering wheel torque according to its built-in control strategy. In this way, a corresponding current command can be output to the motor. The target steering wheel torque generated by the motor is amplified by the worm gear reduction mechanism and superimposed on the driver's operating torque to overcome the steering resistance torque and achieve vehicle steering.

[0069] Optionally, the model parameters of the road feel simulation model can be updated in real time during the road feel simulation control process of the vehicle to ensure the robustness and stability of the model under different vehicle states and driving conditions.

[0070] Specifically, it can obtain the expected steering torque corresponding to the target vehicle; based on the expected steering torque and the first steering torque, the model parameters of the road feel simulation model are updated.

[0071] The desired steering torque can be the steering torque expected for road feel simulation under the current driving conditions. The model parameters can be the parameters of the road feel simulation model, and may include parameters such as weights and biases.

[0072] There are several ways to update the model parameters of the road feel simulation model based on the expected steering torque and the first steering torque. For example, the expected steering torque includes the expected steering torque corresponding to multiple driving conditions. The target driving condition of the target vehicle can be determined based on the vehicle status information. The model parameters of the road feel simulation model are updated based on the target driving condition, the expected steering torque, and the first steering torque.

[0073] The driving condition can be information describing the driving state of the target vehicle, including conditions such as driving straight, turning, rainy weather, snowy weather, and uphill driving. Each driving condition can be set with a corresponding desired steering torque. The target driving condition can be the driving condition in which the target vehicle is currently operating.

[0074] There are several ways to update the model parameters of the road feel simulation model based on the target driving condition, the desired steering torque, and the first steering torque. For example, the target desired steering torque corresponding to the target driving condition can be determined from the desired steering torque; the model parameters of the road feel simulation model can be updated based on the difference between the target desired steering torque and the first steering torque.

[0075] The target desired steering torque can be the desired steering torque corresponding to the target driving condition. Therefore, the model parameters of the road feel simulation model can be adjusted based on the difference between the target desired steering torque and the first steering torque, thereby updating the road feel simulation model.

[0076] Optionally, the model parameters such as weights and biases of the road feel simulation model can be updated periodically to maintain the robustness and performance stability of the road feel simulation model.

[0077] In the calculation of steering torque for steer-by-wire road feel feedback, traditional methods based on dynamic models often require simplification or approximation due to the nonlinearity of the vehicle model, resulting in errors in the final result. To address this, this application proposes a steering torque prediction system for steer-by-wire road feel feedback using a neural network model. This system collects state variables related to the vehicle's state, builds a deep learning network, employs a multilayer perceptron neural network for model construction, and incorporates feature engineering of input variables to better map the relationship between input and output. Simultaneously, it performs redundancy verification of the steering torque using a conventional dynamic model to ensure that the output steering torque does not exhibit unconventional state values. After verification with the second steering torque obtained from a traditional vehicle dynamics model algorithm, it is applied to the road feel torque request of the road feel simulation motor, fulfilling the steer-by-wire requirement for road feel simulation. Therefore, this application uses a more accurate and generalized road feel simulation model based on vehicle dynamics for road feel torque calculation, effectively improving the reliability of road feel simulation.

[0078] In one embodiment, the steer-by-wire road feel simulation system based on a multilayer perceptron neural network provided in this application may include: a road feel simulation model for predicting the simulated steering torque; a data verification and preprocessing module for preprocessing the acquired road feel sensor data (i.e., vehicle state information) and optimizing the state data during vehicle operation in model training, including data cleaning and feature extraction; a neural network model training module for training a recurrent neural network model using the preprocessed road feel sensor data to achieve regression prediction of the steering torque; and a steering torque verification module for verifying the prediction results of the road feel simulation model based on traditional vehicle dynamics model-based steering torque calculations and controlling the vehicle road feel simulation motor to output the target steering torque. Therefore, when this steer-by-wire road feel simulation system is used for simulation testing, it works normally without any abnormal road feel feedback values. The system is stable and reliable, ensuring normal and safe driving for the driver.

[0079] Optionally, the steer-by-wire road feel simulation system may also include a data storage unit for storing historical vehicle state parameters, environmental information, and driver control commands to facilitate subsequent model training and performance optimization.

[0080] Optionally, the steer-by-wire road feel simulation system may also include an anomaly detection module, which is used to detect whether the first steering torque output by the road feel simulation model is abnormal, such as whether it exceeds the safe range or is a system malfunction. Based on the detection results, corresponding safety measures can be taken, such as alarms or reducing the control amplitude.

[0081] Therefore, the steering torque prediction system based on a road feel simulation model provided in this application can monitor vehicle state information (such as vehicle state parameters and environmental information) during steering and update the input data of the road feel simulation model in real time. This ensures the real-time performance, robustness, and stability of the road feel simulation model under different vehicle states and driving conditions. Simultaneously, the system can evaluate and monitor the rationality and safety of the steering torque control behavior output by the road feel simulation model by setting typical thresholds perceptible to the driver. The system verifies and monitors whether the control behavior is abnormal and takes corresponding actions. It periodically assists in updating the weights and biases of the road feel simulation model to maintain the model's robustness and performance stability. Combined with a vehicle dynamics model, it ensures dual redundancy in vehicle steering control, improving the reliability of road feel simulation and driving safety. In one embodiment, please refer to... Figure 3 , Figure 3 This is a schematic diagram of the specific architecture of a road feel simulation data determination method provided in an embodiment of this application. In the steering torque prediction system, it can include a road feel simulation model, a vehicle dynamics model, and a steering wheel module. The steering wheel module can include a road feel controller and a road feel motor. This allows it to acquire input data such as the target vehicle's steering wheel angle, steering wheel angular velocity, steering wheel torque, vehicle speed, yaw rate, lateral acceleration, vertical load, front wheel angle, and desired steering torque. These data can then be input into the road feel simulation model to predict the first steering torque. Data such as steering wheel angular velocity, steering wheel torque, vehicle speed, and front wheel steering angle are input into the vehicle dynamics model to calculate the second steering torque. Based on the torque difference between the first and second steering torques and a preset torque difference threshold, the target steering torque can be determined. Then, based on the road feel controller and road feel motor in the steering wheel module, the target steering torque and corresponding steering wheel angle can be determined. The corresponding torque is then applied to the steering wheel based on the target steering torque to achieve road feel simulation. At the same time, the parameters of the road feel simulation model can be updated according to the difference between the desired steering torque and the first steering torque to achieve robustness and stability of the model under different vehicle states and driving conditions, further improving the reliability of road feel simulation.

[0082] As described above, the embodiments of this application predict the first steering torque of the target vehicle based on the vehicle state information using a road feel simulation model; calculate the second steering torque of the target vehicle based on a vehicle dynamics model; and determine the target steering wheel torque based on the first and second steering torques, thereby performing road feel simulation control of the target vehicle based on the target steering wheel torque. In this way, by predicting the first steering torque of the target vehicle based on vehicle state information using a road feel simulation model, and calculating the second steering torque of the target vehicle based on a vehicle dynamics model, and combining the first and second steering torques to determine the target steering wheel torque used for road feel simulation, stable and accurate road feel simulation can be achieved in complex driving scenarios, thereby improving the reliability of road feel simulation.

[0083] To facilitate better implementation of the road sense simulation data determination method provided in the embodiments of this application, this application also provides another road sense simulation data determination method. The meanings of the terms used are the same as in the road sense simulation data determination method described above, and specific implementation details can be found in the descriptions of the above method embodiments. For example, please refer to... Figure 4 , Figure 4 This is another schematic flowchart of a method for determining road sense simulation data according to an embodiment of this application. The method for determining road sense simulation data may include the following steps:

[0084] Step S201: Obtain the vehicle status information of the target vehicle.

[0085] The vehicle status information may include at least one of the following: steering wheel angle, steering wheel angular velocity, steering wheel torque, vehicle speed, yaw rate, lateral acceleration, vertical load, and front wheel angle.

[0086] Step S202: Based on the vehicle state information, the steering torque is predicted using a road feel simulation model to obtain the first steering torque of the target vehicle.

[0087] The road feel simulation model can include multiple network layers, and the multiple network layers adopt a fully connected structure.

[0088] Optionally, each network layer may include at least one neuron, and the input information of each neuron may include a combination of the input from each neuron in the previous layer and the corresponding weight factor. The input information is activated using the hyperbolic tangent function to obtain the output of each neuron.

[0089] The hyperbolic tangent function tanh can be used as the activation function in the road feel simulation model, and a bias b can be added. The expression for the tanh function is as follows:

[0090] Optionally, based on vehicle state information, the steering torque can be predicted using a road feel simulation model to obtain the first steering torque of the target vehicle in various ways. For example, the network layer can include an input layer, a hidden layer, and an output layer. The input layer can receive vehicle state information; the hidden layer can extract and transform features from the vehicle state information received by the input layer to obtain feature processing results; and the output layer can process the feature processing results to obtain the first steering torque of the target vehicle.

[0091] The road perception simulation model can be a neural network model employing a multilayer perceptron structure. The input layer receives input data such as vehicle state information; each neuron corresponds to a feature of the input data. The hidden layer, located between the input and output layers, can contain multiple layers, each composed of multiple neurons. This hidden layer can perform nonlinear transformations on the input data using a weight matrix and extract higher-order features. Each hidden layer can contain up to 20 neurons. The output layer generates the final prediction result. All layers are fully connected, meaning each neuron in the current layer is connected to all neurons in the next layer.

[0092] In one specific embodiment, please refer to Figure 5 , Figure 5 This is a schematic diagram of the model structure of a road sense simulation data determination method provided in an embodiment of this application. The road sense simulation model can be a fully connected neural network including one input layer, five hidden layers, and 16 neurons in each layer.

[0093] Optional, please refer to Figure 6 , Figure 6 This is a schematic diagram of another model structure for a road sense simulation data determination method provided in this application embodiment. It shows the neural network structure in the l-th layer of the road sense simulation model network structure, where the input of a neuron is a combination of the output of the i-th neuron in the previous layer and the corresponding weight factor (w). The activation state n of each neuron describes the sum of all weighted inputs and bias values ​​b, and can be expressed as:

[0094]

[0095] The output of the neuron can then be obtained using the transfer function (also known as the activation function) f:

[0096]

[0097] Here, W can be represented as a matrix of weighting factors (w).

[0098] Optionally, cross-validation can be used to train the road feel simulation model. Specifically, the training data corresponding to the road feel simulation model can be divided into a training set and a validation set; the road feel simulation model is trained using the training set and cross-validated using the validation set to obtain the trained road feel simulation model.

[0099] Optionally, when training the road feel simulation model, a numerical optimization algorithm based on the Jacobian matrix can be used to optimize the model parameters.

[0100] The numerical optimization algorithm can be the Levenberg-Marquardt algorithm (LMBP algorithm for short).

[0101] In one specific embodiment, this application employs the squared loss function, most commonly used in regression problems. Specifically, during model training, the squared error between the actual steering torque of the vehicle and the output of the road feel simulation model is used as the loss function to optimize and update the weights w and biases b of the road feel simulation model, thereby improving its predictive performance. The loss function can be defined as follows:

[0102]

[0103] One approach is to use the LMBP algorithm for model optimization. LMBP replaces the Hessian matrix calculation with the Jacobian matrix, which improves model training efficiency. It is a numerical optimization algorithm that accelerates convergence backpropagation (BP) and is suitable for optimizing models with squared loss functions. Since LMBP cannot guarantee convergence to the global minimum of the error function, the training dataset of the road feel simulation model can be divided into 20 groups to train the initial network of the road feel simulation model simultaneously. This means that for the model training of the road feel simulation model, 20 networks with randomly initialized weights were analyzed, and the optimal model of the network was trained from 20 different directions to approximate the model. The accuracy of the network model under different initial conditions was evaluated in the subsequent cross-validation. To determine a network model that is very suitable for the current problem, it is essential to collect a large amount of effective data and perform data preprocessing.

[0104] Optionally, the learning rate can be set to 0.01 and the number of iterations (epochs) to 100. To avoid overfitting and affecting robustness, the existing samples can be divided into training data, test data, and validation data. The training data is presented to the road sense simulation model in consecutive epochs to determine the optimal setting of weighted connections. This allows for cross-validation to improve the generalization performance of the network structure and determine the optimal network structure. During training, approximately 10-15% of the samples can be randomly selected from the total sample set as validation and test samples. After each epoch, the connection weights are adjusted accordingly, and the training error of the training data used to adjust the weights is first determined. Then, the network is tested with unknown data, and the validation and test errors are determined, observing the trend of the validation error.

[0105] Before training the road feel simulation model, a large amount of vehicle state information needs to be collected as training data. Simultaneously, the corresponding vehicle steering torque also needs to be collected. Regarding data processing conditions, routine data collection is conducted for users, including scenarios they encounter such as urban roads and highways. At the same time, data collection is also conducted for special conditions such as parking and turning. In terms of road surface types, data collection must cover various road surfaces, including dry, wet, and icy surfaces, to obtain training data under various conditions and improve the robustness of the trained road feel simulation model.

[0106] Optionally, the collected training data may also need to undergo data preprocessing, which may include:

[0107] Input data evaluation: The impact of input data can be verified by gradually changing the input data. This involves gradually changing the input data to verify the nonlinear impact of each data point on the network output quality, transforming the original input data into more informative features, thereby improving the model's predictive performance.

[0108] Data augmentation: Data augmentation techniques are used to expand input data and generate more training samples. For example, in vehicle driving data, input data such as speed, acceleration, and angular velocity can be randomly perturbed to generate more training samples, thereby increasing data diversity and improving the model's generalization performance.

[0109] Normalization: Normalizing the input data scales it to the same range to prevent scale differences between different input data from negatively affecting model training. In this application embodiment, a normalization method can be used to normalize the number of inputs.

[0110] Feature selection: Correlation analysis of selected inputs can be achieved through methods such as statistical analysis, feature engineering, and principal component analysis (PCA) to improve the training efficiency and generalization performance of the road sense simulation model.

[0111] In one specific embodiment, please refer to Figure 7 , Figure 7 This is a schematic diagram illustrating the specific process of a method for determining road feel simulation data provided in an embodiment of this application. It allows for the identification of the required neural network model, selection of network input variables, collection of raw training data, preprocessing of the raw data, and subsequent model training and validation. The accuracy of the trained road feel simulation model is then determined. If accurate, the model can be deployed. If inaccurate, if the training volume is low, model training can continue. If the problem lies with the model itself, the required neural network model can be re-identified. If the problem is with the input data type, the input variables of the selected network can be readjusted. If the problem is with insufficient input samples, raw training data can continue to be collected. If the problem is with feature engineering, feature engineering can be used to preprocess the raw data. In this way, a road feel simulation model is trained.

[0112] Therefore, the road feel simulation model provided in this application is a network model specifically designed for road feel simulation. Taking into account the network depth and the number of neurons, a fully connected neural network with 5 hidden layers and 16 neurons per layer is proposed. Furthermore, the activation function used is the tanh function, which has better performance in handling regression problems compared to the sigmoid function, thus resulting in a road feel simulation model with good performance and accuracy. In addition, the training optimizer settings of the road feel simulation model in this application embodiment are different. Compared to the traditional BP optimization method, this application embodiment uses the LMBP algorithm for model optimization, which is suitable for optimizing models with squared loss functions, thereby improving training efficiency. Simultaneously, this application embodiment performs cross-validation of the model after each training epoch and observes the validation residuals in real time. When the validation residuals continuously increase, the training is adjusted to achieve optimization, resulting in a model with strong predictive ability. In addition, in terms of data preprocessing, this application embodiment uses methods such as input value evaluation, data augmentation, normalization, and feature engineering to improve the model training effect. Based on the gradual perturbation of the input values ​​and the correlation analysis of the selected inputs, the input data is dynamically adjusted to ensure the robustness and stability of the road feel simulation model under different vehicle states and driving conditions, and further training to obtain a road feel simulation model with better road feel simulation effect.

[0113] As can be seen from the above, the embodiments of this application obtain the vehicle state information of the target vehicle; based on the vehicle state information, the steering torque is predicted using a road feel simulation model to obtain the first steering torque of the target vehicle. Thus, by accurately predicting the first steering torque of the vehicle using a road feel simulation model based on the vehicle state information of the target vehicle, the reliability of road feel simulation is improved.

[0114] To facilitate better implementation of the road feeling simulation data determination method provided in the embodiments of the present application, the embodiments of the present application also provide a device based on the above road feeling simulation data determination method. The meanings of the nouns are the same as those in the above road feeling simulation data determination method, and the specific implementation details can be referred to the descriptions in the method embodiments.

[0115] For example, as Figure 8 shown, it is a schematic structural diagram of the road feeling simulation data determination device provided in the embodiments of the present application. The road feeling simulation data determination device may include a prediction module 301, a calculation module 302, and a determination module 303, specifically as follows:

[0116] The prediction module 301 is configured to predict the first steering torque of the target vehicle through a road feeling simulation model based on the vehicle state information of the target vehicle; [[ID=⑨]] [[ID=⑩]]

[0117] [[ID=⑪]]The calculation module 302 is configured to calculate the second steering torque of the target vehicle based on a vehicle dynamics model; [[ID=⑫]] [[ID=⑬]]

[0118] [[ID=⑭]]The determination module 303 is configured to determine the target steering wheel torque according to the first steering torque and the second steering torque, so as to perform road feeling simulation control on the target vehicle based on the target steering wheel torque. [[ID=⑮]] [[ID=⑯]]

[0119] [[ID=⑰]]In one embodiment, the vehicle state information includes at least one of the steering wheel angle, steering wheel angular velocity, steering wheel torque, vehicle speed, yaw angular velocity, lateral acceleration, vertical load, and front wheel angle of the target vehicle. [[ID=⑱]] [[ID=⑲]]

[0120] [[ID=⑳]]In one embodiment, the determination module 303 is specifically configured to: [[ID=㉑]] [[ID=㉒]]

[0121] [[ID=㉓]]Determine the target steering torque according to the first steering torque and the second steering torque; [[ID=㉔]] [[ID=㉕]]

[0122] [[ID=㉖]]Calculate the target steering wheel torque based on the target steering torque, and the target steering wheel torque is the torque that needs to be applied to the steering wheel of the target vehicle. [[ID=㉗]] [[ID=㉘]]

[0123] [[ID=㉙]]In one embodiment, the above determination of the target steering torque according to the first steering torque and the second steering torque is specifically configured to: [[ID=㉚]] [[ID=㉛]]

[0124] [[ID=㉜]]Calculate the torque difference between the first steering torque and the second steering torque; [[ID=㉝]] [[ID=㉞]]

[0125] [[ID=㉟]]Determine the target steering torque based on the torque difference and a preset torque difference threshold. [[ID=㊱]] [[ID=㊲]]

[0126] [[ID=㊳]]In one embodiment, the above determination of the target steering torque based on the torque difference and the preset torque difference threshold includes: [[ID=㊴]] [[ID=㊵]]

[0127] [[ID=㊶]]If the torque difference is not greater than the preset torque difference threshold, determine the first steering torque as the target steering torque;

[0128] If the torque difference is greater than the preset torque difference threshold, the second steering torque is determined as the target steering torque.

[0129] In one embodiment, the road feel simulation data determination device further includes a model update module, used for:

[0130] Obtain the desired steering torque for the target vehicle;

[0131] The model parameters of the road feel simulation model are updated based on the expected steering torque and the first steering torque.

[0132] In one embodiment, the desired steering torque includes desired steering torques corresponding to multiple driving conditions. The above-mentioned updating of the model parameters of the road feel simulation model based on the desired steering torque and the first steering torque includes:

[0133] Based on the vehicle status information, determine the target driving condition of the target vehicle.

[0134] The model parameters of the road feel simulation model are updated based on the target driving conditions, the desired steering torque, and the first steering torque.

[0135] In one embodiment, updating the model parameters of the road feel simulation model based on the target driving condition, the desired steering torque, and the first steering torque includes:

[0136] Determine the target desired steering torque corresponding to the target driving condition from the desired steering torque;

[0137] The model parameters of the road feel simulation model are updated based on the difference between the target desired steering torque and the first steering torque.

[0138] In one embodiment, the calculation module 302 is used for:

[0139] Based on vehicle state information, the steering torque associated with the steering torque of the target vehicle is calculated using a vehicle dynamics model.

[0140] Determine the second steering torque of the target vehicle based on the steering-related torque.

[0141] In one embodiment, the steering-related torque includes at least one of the following: the torque generated by the stiffness of the steering system of the target vehicle, the torque generated by the system damping, the torque generated by the system equivalent damping, the system friction torque, and the tire return torque.

[0142] As can be seen from the above, in this embodiment, the prediction module 301 predicts the first steering torque of the target vehicle based on the vehicle state information and the road feel simulation model; the calculation module 302 calculates the second steering torque of the target vehicle based on the vehicle dynamics model; and the determination module 303 determines the target steering wheel torque based on the first and second steering torques, so as to perform road feel simulation control on the target vehicle based on the target steering wheel torque. Thus, by predicting the first steering torque of the target vehicle based on the vehicle state information using the road feel simulation model, and calculating the second steering torque of the target vehicle based on the vehicle dynamics model, and combining the first and second steering torques, the target steering wheel torque used for road feel simulation of the vehicle can be determined. This enables stable and accurate road feel simulation in complex driving scenarios, thereby improving the reliability of road feel simulation.

[0143] To facilitate better implementation of the road sense simulation data determination method provided in the embodiments of this application, this application also provides an apparatus based on the above-described road sense simulation data determination method. The meanings of the terms used are the same as in the road sense simulation data determination method described above, and specific implementation details can be found in the descriptions in the method embodiments.

[0144] For example, such as Figure 9 The diagram shown is another structural schematic of the road feel simulation data determination device provided in this application embodiment. This road feel simulation data determination device may include an information acquisition module 401 and a torque prediction module 402, as detailed below:

[0145] The information acquisition module 401 is used to acquire the vehicle status information of the target vehicle;

[0146] The torque prediction module 402 is used to predict the steering torque based on vehicle state information and a road feel simulation model to obtain the first steering torque of the target vehicle.

[0147] In one embodiment, the road feel simulation model includes multiple network layers, which are fully connected.

[0148] In one embodiment, each network layer includes at least one neuron, and the input information of each neuron includes a combination of the input from each neuron in the previous layer and the corresponding weight factor. The input information is activated using a hyperbolic tangent function to obtain the output of each neuron.

[0149] In one embodiment, the network layer includes an input layer, a hidden layer, and an output layer. The torque prediction module 402 is used for:

[0150] Vehicle status information is received through the input layer;

[0151] The hidden layer extracts and transforms features from the vehicle state information received from the input layer to obtain the feature processing result.

[0152] The first steering torque of the target vehicle is obtained by processing the feature processing results through the output layer.

[0153] In one embodiment, the road feel simulation data determination device further includes a model training module, used for:

[0154] The training data corresponding to the road feel simulation model is divided into a training set and a validation set;

[0155] The road feel simulation model is trained using a training set and cross-validated using a validation set to obtain the trained road feel simulation model.

[0156] In one embodiment, when training the road feel simulation model, a numerical optimization algorithm based on the Jacobian matrix is ​​used to optimize the model parameters.

[0157] As can be seen from the above, in this embodiment of the application, the vehicle state information of the target vehicle is obtained by the information acquisition module 401; the torque prediction module 402 predicts the steering torque based on the vehicle state information using a road feel simulation model, thereby obtaining the first steering torque of the target vehicle. Thus, by accurately predicting the first steering torque of the vehicle using a road feel simulation model based on the vehicle state information of the target vehicle, the reliability of road feel simulation is improved.

[0158] Accordingly, embodiments of this application also provide a controller, such as Figure 10 As shown, Figure 10 This is a schematic diagram of the controller structure provided in an embodiment of this application. The controller 500 includes a processor 501 with one or more processing cores, a memory 502 with one or more computer-readable storage media, and a computer program stored in the memory 502 and executable on the processor. The processor 501 and the memory 502 are electrically connected. Those skilled in the art will understand that the controller structure shown in the figure does not constitute a limitation on the controller, and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0159] The processor 501 is the control center of the controller 500. It connects various parts of the controller 500 through various interfaces and lines. By running or loading software programs and / or units stored in the memory 502, and calling data stored in the memory 502, it executes various functions of the controller 500 and processes data. The processor 501 can be a CPU, GPU, network processor (NP), etc., and can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application.

[0160] In this embodiment, the processor 501 in the controller 500 loads the instructions corresponding to the processes of one or more applications into the memory 502 according to the following steps, and the processor 501 runs the applications stored in the memory 502 to achieve various functions, such as:

[0161] Based on the vehicle state information of the target vehicle, the first steering torque of the target vehicle is predicted through a road feel simulation model; based on the vehicle dynamics model, the second steering torque of the target vehicle is calculated; based on the first steering torque and the second steering torque, the target steering wheel torque is determined, and road feel simulation control of the target vehicle is performed based on the target steering wheel torque.

[0162] Alternatively, obtain the vehicle status information of the target vehicle; based on the vehicle status information, use a road feel simulation model to predict the steering torque and obtain the first steering torque of the target vehicle.

[0163] Furthermore, the various functions implemented by running the application stored in memory 502 can also be found in the descriptions of the foregoing embodiments, and will not be repeated here.

[0164] For details on the implementation of each of the above operations, please refer to the previous examples, which will not be repeated here.

[0165] Optional, such as Figure 10 As shown, the controller 500 also includes: a touch display screen 503, an radio frequency circuit 504, an audio circuit 505, an input unit 506, and a power supply 507. The processor 501 is electrically connected to the touch display screen 503, the radio frequency circuit 504, the audio circuit 505, the input unit 506, and the power supply 507. Those skilled in the art will understand that... Figure 10 The controller structure shown does not constitute a limitation on the controller and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0166] The touch display screen 503 can be used to display a graphical user interface (GUI) and receive operation commands generated by the user interacting with the GUI. The touch display screen 503 may include a display panel and a touch panel. The display panel can be used to display information input by the user or information provided to the user, as well as various graphical user interfaces of the controller. These graphical user interfaces can be composed of graphics, text, icons, video, and any combination thereof. Optionally, the display panel can be configured using a liquid crystal display (LCD), organic light-emitting diode (OLED), or other similar technologies. The touch panel can be used to collect touch operations performed by the user on or near it (such as operations performed by the user using a finger, stylus, or any suitable object or accessory on or near the touch panel), generate corresponding operation commands, and execute the corresponding program according to the operation commands. Optionally, the touch panel may include two parts: a touch detection device and a touch controller. The touch detection device detects the user's touch location and the signal generated by the touch operation, transmitting the signal to the touch controller. The touch controller receives touch information from the touch detection device, converts it into touch point coordinates, and sends it to the processor 501. It can also receive and execute commands from the processor 501. The touch panel can cover the display panel. When the touch panel detects a touch operation on or near it, it transmits the information to the processor 501 to determine the type of touch event. Subsequently, the processor 501 provides corresponding visual output on the display panel based on the type of touch event. In this embodiment, the touch panel and the display panel can be integrated into the touch display screen 503 to achieve input and output functions. However, in some embodiments, the touch panel and the touch display screen 503 can be implemented as two independent components to achieve input and output functions. That is, the touch display screen 503 can also be used as part of the input unit 506 to achieve input functions.

[0167] The radio frequency circuit 504 can be used to transmit and receive radio frequency signals to establish wireless communication with network devices or other controllers, and to transmit and receive signals with network devices or other controllers.

[0168] Audio circuitry 505 can be used to provide an audio interface between the user and the controller via a speaker and a microphone. Audio circuitry 505 converts received audio data into electrical signals, transmits them to the speaker, and the speaker converts them into sound signals for output. Conversely, the microphone converts collected sound signals into electrical signals, which are then received by audio circuitry 505, converted back into audio data, and output to processor 501 for processing. The audio data is then transmitted via radio frequency circuitry 504 to, for example, another controller, or output to memory 502 for further processing. Audio circuitry 505 may also include an earphone jack to provide communication between peripheral headphones and the controller.

[0169] The input unit 506 can be used to receive input target video and generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function control.

[0170] Power supply 507 is used to supply power to the various components of controller 500. Optionally, power supply 507 can be logically connected to processor 501 through a power management system, thereby enabling functions such as charging, discharging, and power consumption management through the power management system. Power supply 507 may also include one or more DC or AC power supplies, recharging systems, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components.

[0171] although Figure 10 As not shown in the diagram, the controller 500 may also include a camera, sensor, wireless fidelity module, Bluetooth module, etc., which will not be described in detail here.

[0172] In the above embodiments, the descriptions of each embodiment have different focuses. Parts not described in detail in a particular embodiment can be found in the relevant descriptions of other embodiments. It should be noted that the controller provided in this application embodiment and the method for determining road sense simulation data in the above embodiments belong to the same concept. The specific implementation process is detailed in the above method embodiments and will not be repeated here.

[0173] As can be seen from the above, the controller provided in this application embodiment can predict the first steering torque of the target vehicle based on the vehicle state information of the target vehicle through a road feel simulation model; calculate the second steering torque of the target vehicle based on a vehicle dynamics model; and determine the target steering wheel torque based on the first and second steering torques, thereby performing road feel simulation control of the target vehicle based on the target steering wheel torque. In this way, by predicting the first steering torque of the target vehicle based on vehicle state information through a road feel simulation model, and calculating the second steering torque of the target vehicle based on a vehicle dynamics model, and combining the first and second steering torques to determine the target steering wheel torque used for road feel simulation, stable and accurate road feel simulation can be achieved in complex driving scenarios, thereby improving the reliability of road feel simulation.

[0174] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be performed by instructions, or by instructions controlling related hardware. These instructions can be stored in a computer-readable storage medium and loaded and executed by a processor.

[0175] Therefore, embodiments of this application provide a computer-readable storage medium, including a computer program, which, when run on a controller, causes the controller to execute any of the road sense simulation data determination methods provided in embodiments of this application. For example, the computer program can execute the steps of the following road sense simulation data determination method:

[0176] Based on the vehicle state information of the target vehicle, the first steering torque of the target vehicle is predicted through a road feel simulation model; based on the vehicle dynamics model, the second steering torque of the target vehicle is calculated; based on the first steering torque and the second steering torque, the target steering wheel torque is determined, and road feel simulation control of the target vehicle is performed based on the target steering wheel torque.

[0177] Alternatively, obtain the vehicle status information of the target vehicle; based on the vehicle status information, use a road feel simulation model to predict the steering torque and obtain the first steering torque of the target vehicle.

[0178] Furthermore, the detailed steps of the above method can be found in the description of the foregoing embodiments, and will not be repeated here.

[0179] For details on the implementation of each of the above operations, please refer to the previous examples, which will not be repeated here.

[0180] The computer-readable storage medium may include: read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0181] Since the computer program stored in the computer-readable storage medium can execute any of the road sense simulation data determination methods provided in the embodiments of this application, the beneficial effects that any of the road sense simulation data determination methods provided in the embodiments of this application can achieve can be realized, as detailed in the preceding embodiments, and will not be repeated here.

[0182] According to one aspect of this application, a computer program product is also provided, comprising a computer program stored in a computer-readable storage medium; when a processor of a controller reads the computer program from the computer-readable storage medium, the processor executes the computer program, causing the controller to perform the methods provided in various optional implementations of the above embodiments.

[0183] In the above embodiments of the road feel simulation data determination device, computer-readable storage medium, controller, and computer program product, the descriptions of each embodiment have different focuses. Parts not described in detail in a particular embodiment can be referred to in the relevant descriptions of other embodiments. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process and beneficial effects of the road feel simulation data determination device, computer-readable storage medium, computer program product, controller, and their corresponding units described above can be referred to the description of the road feel simulation data determination method in the above embodiments, and will not be repeated here.

[0184] The foregoing has provided a detailed description of a method, apparatus, controller, vehicle, computer-readable storage medium, and computer program product for determining road sense simulation data according to embodiments of this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A road-feeling simulation data determination method characterized by comprising: The method comprises: predicting a first steering torque of the target vehicle by a road feel simulation model based on vehicle state information of the target vehicle; calculating a second steering torque of the target vehicle based on a vehicle dynamics model; determining a target steering wheel torque based on the first steering torque and the second steering torque, so as to perform road feel simulation control on the target vehicle based on the target steering wheel torque.

2. The road feel simulation data determination method according to claim 1, characterized by, The vehicle state information comprises at least one of a steering wheel angle, a steering wheel angular velocity, a steering wheel torque, a vehicle speed, a yaw angular velocity, a lateral acceleration, a vertical load and a front wheel angle of the target vehicle.

3. The road feel simulation data determination method according to claim 1, characterized by, The determining of the target steering wheel torque based on the first steering torque and the second steering torque comprises: determining a target steering torque based on the first steering torque and the second steering torque; calculating a target steering wheel torque based on the target steering torque, the target steering wheel torque being a torque required to be applied to the steering wheel of the target vehicle.

4. The road feel simulation data determination method according to claim 3, characterized by, The determining of the target steering torque based on the first steering torque and the second steering torque comprises: calculating a torque difference between the first steering torque and the second steering torque; determining a target steering torque based on the torque difference and a preset torque difference threshold.

5. The road feel simulation data determination method according to claim 4, characterized by, The determining of the target steering torque based on the torque difference and the preset torque difference threshold comprises: if the torque difference is not greater than the preset torque difference threshold, determining the first steering torque as the target steering torque; if the torque difference is greater than the preset torque difference threshold, determining the second steering torque as the target steering torque.

6. The road feel simulation data determination method according to claim 1, characterized by, The method further comprises: obtaining an expected steering torque corresponding to the target vehicle; updating a model parameter of the road feel simulation model based on the expected steering torque and the first steering torque.

7. The road feel simulation data determination method according to claim 6, characterized by, The expected steering torque comprises expected steering torques corresponding to a plurality of driving conditions, and the updating of the model parameter of the road feel simulation model based on the expected steering torque and the first steering torque comprises: determining a target driving condition in which the target vehicle is located according to the vehicle state information; updating the model parameter of the road feel simulation model based on the target driving condition, the expected steering torque and the first steering torque.

8. The road feel simulation data determination method according to claim 7, characterized by, The updating of the model parameter of the road feel simulation model based on the target driving condition, the expected steering torque and the first steering torque comprises: determining a target expected steering torque corresponding to the target driving condition in the expected steering torque; updating the model parameter of the road feel simulation model based on a difference between the target expected steering torque and the first steering torque.

9. The road feel simulation data determination method according to any one of claims 1 to 8, characterized by, The calculating of the second steering torque of the target vehicle based on the vehicle dynamics model comprises: calculating a steering-related torque associated with the steering torque of the target vehicle by a vehicle dynamics model based on the vehicle state information; determining the second steering torque of the target vehicle according to the steering-related torque.

10. The road feel simulation data determination method according to claim 9, characterized by, The turning associated moment includes at least one of a moment generated by stiffness of a turning system of the target vehicle, a moment generated by system damping, a moment generated by system equivalent damping, a system friction moment, and a tire aligning moment.

11. A road feel simulation data determination method characterized by comprising: The method comprises: obtaining vehicle state information of a target vehicle; based on the vehicle state information, predicting a turning moment by using a road feel simulation model to obtain a first turning moment of the target vehicle.

12. The road feel simulation data determination method according to claim 11, characterized by, The road feel simulation model comprises a plurality of network layers, and full connection structures are adopted between the network layers.

13. The road feel simulation data determination method according to claim 12, characterized by, Each network layer comprises at least one neuron, input information of each neuron comprises combined information from input of each neuron of a previous layer and a corresponding weight factor, and a hyperbolic tangent function is used to activate the input information to obtain output of each neuron.

14. The road feel simulation data determination method according to claim 12, characterized by, The network layer comprises an input layer, a hidden layer, and an output layer. The method further comprises: dividing training data corresponding to the road feel simulation model into a training set and a validation set; training the road feel simulation model by using the training set, and cross-validating the road feel simulation model by using the validation set to obtain a trained road feel simulation model. When training the road feel simulation model, a numerical optimization algorithm based on a Jacobian matrix is used to optimize model parameters.

15. The road feel simulation data determination method according to claim 11, characterized by, The method comprises: a prediction module configured to predict a first turning moment of a target vehicle by using a road feel simulation model based on vehicle state information of the target vehicle; a calculation module configured to calculate a second turning moment of the target vehicle based on a vehicle dynamics model; 16. The road feel simulation data determination method according to claim 15, characterized by, a determination module configured to determine a target steering wheel torque based on the first turning moment and the second turning moment, and to perform road feel simulation control on the target vehicle based on the target steering wheel torque.

17. A road feeling simulation data determination device characterized by comprising: The controller comprises a processor and a memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor is caused to execute steps of the method in any one of claims 1-16. The vehicle comprises the controller in claim 18. The computer program is used to cause the controller to execute steps of the method in any one of claims 1-16 when the computer program is executed on the controller. The computer program or instructions are executed by the processor to implement steps of the method in any one of claims 1-16.

18. A controller characterized by comprising: ​ 19. A vehicle characterized by comprising: ​ 20. A computer-readable storage medium, characterized in that, ​ 21. A computer program product, characterised in that, ​

Images