Vehicle braking control method, device, equipment and storage medium

Abstract

The embodiment of the present invention discloses a vehicle braking control method, device, equipment and storage medium, including: obtaining the current vehicle driving information of the target vehicle, the current vehicle driving information including: at least one of the vehicle driving speed, wheel speed and deflection angular velocity; inputting the current vehicle driving information into a pre-trained target braking decision model to obtain the target braking control information corresponding to the current vehicle driving information; adjusting the current vehicle driving information of the target vehicle according to the target braking control information; the target braking decision model is obtained by reinforcement learning based on the target reward function; the target reward function is constructed based on the wheel slip rate and body deflection angle of the target vehicle. The technical solution of the embodiment of the present invention can train the braking control model based on the reward function constructed based on the wheel slip rate and body deflection angle, improve the robustness of the model, improve the accuracy of the vehicle braking control information output by the model, and improve the safety performance of the vehicle.

Description

Technical Field

[0001] Embodiments of the present invention relate to the field of vehicle braking technology, and in particular, to a vehicle braking control method, device, equipment and storage medium. Background Art

[0002] With the development of automobiles, users pay more and more attention to the safety of vehicle driving. The braking ability of a car is one of the important factors affecting the safety of vehicle driving. The slip rate can be used to characterize the braking ability of a vehicle.

[0003] At present, the existing vehicle braking control technology can use the pre-trained vehicle braking control model to obtain vehicle control information. The existing technology can train the vehicle braking control model based on the slip rate as a reward function, and then obtain the vehicle braking control information by inputting the current vehicle speed information and the current wheel speed information into the model. However, using only the slip rate as a reward function is too simplistic. Using only the slip rate as a reward function is prone to the problem of insufficient robustness of the vehicle braking control model, resulting in insufficient accuracy of the vehicle braking control information output by the model. Summary of the invention

[0004] The embodiments of the present invention provide a vehicle braking control method, device, equipment and storage medium, which can train a braking control model based on a reward function constructed based on wheel slip rate and vehicle body deflection angle, improve the robustness of the model, improve the accuracy of vehicle braking control information output by the model, and enhance vehicle safety performance.

[0005] In a first aspect, an embodiment of the present invention provides a vehicle braking control method, the method comprising:

[0006] Acquire current vehicle driving information of the target vehicle, wherein the current vehicle driving information includes: at least one of the vehicle driving speed, wheel speed, and deflection angular velocity;

[0007] Inputting the current vehicle driving information into a pre-trained target braking decision model to obtain target braking control information corresponding to the current vehicle driving information;

[0008] adjusting current vehicle driving information of the target vehicle according to the target braking control information;

[0009] Wherein, the target braking decision model is obtained by reinforcement learning based on a target reward function; the target reward function is constructed based on the wheel slip rate and body deflection angle of the target vehicle.

[0010] In a second aspect, an embodiment of the present invention provides a vehicle braking control device, the device comprising:

[0011] A vehicle driving information acquisition module, used to acquire current vehicle driving information of a target vehicle, wherein the current vehicle driving information includes: at least one of vehicle driving speed, wheel speed, and deflection angular velocity;

[0012] A braking control information determination module, used for inputting the current vehicle driving information into a pre-trained target braking decision model to obtain target braking control information corresponding to the current vehicle driving information;

[0013] A vehicle driving information adjustment module, used for adjusting the current vehicle driving information of the target vehicle according to the target braking control information;

[0014] Wherein, the target braking decision model is obtained by reinforcement learning based on a target reward function; the target reward function is constructed based on the wheel slip rate and body deflection angle of the target vehicle.

[0015] In a third aspect, an embodiment of the present invention provides a computer device, the computer device comprising:

[0016] one or more processors;

[0017] A memory for storing one or more programs;

[0018] When the one or more programs are executed by the one or more processors, the one or more processors implement the vehicle braking control method described in any embodiment.

[0019] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the vehicle braking control method described in any one of the embodiments.

[0020] The technical solution provided by the embodiment of the present invention obtains the current vehicle driving information of the target vehicle, the current vehicle driving information includes: at least one of the vehicle driving speed, wheel speed, and deflection angular velocity; inputs the current vehicle driving information into a pre-trained target braking decision model to obtain the target braking control information corresponding to the current vehicle driving information; adjusts the current vehicle driving information of the target vehicle according to the target braking control information; the target braking decision model is obtained by reinforcement learning based on the target reward function; the target reward function is constructed based on the wheel slip rate and body deflection angle of the target vehicle. The technical solution of the embodiment of the present invention solves the problem of insufficient robustness of the vehicle braking control model in the prior art, which leads to the problem of insufficient accuracy of the vehicle braking control information output by the model. The wheel slip rate and body deflection angle can be used as reward functions to train the vehicle braking control model, improve the robustness of the model, improve the accuracy of the vehicle braking control information output by the model, and improve the vehicle safety performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 is a flow chart of a vehicle braking control method provided by an embodiment of the present invention;

[0022] Figure 2 is a flow chart of another vehicle braking control method provided by an embodiment of the present invention;

[0023] Figure 3 This is a workflow diagram of the underlying logic of an ESP provided by an embodiment of the present invention;

[0024] Figure 4 This is a working flow chart for performing vehicle braking control provided by an embodiment of the present invention;

[0025] Figure 5 is a structural schematic diagram of a vehicle braking control device provided by an embodiment of the present invention;

[0026] Figure 6 It is a structural schematic diagram of a computer device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0027] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0028] Figure 1 It is a flow chart of a vehicle braking control method provided by an embodiment of the present invention. The embodiment of the present invention can be applied to the scenario of controlling a mobile vehicle to automatically adjust driving information. The method can be executed by a vehicle braking control device, which can be implemented by software and / or hardware.

[0029] like Figure 1 As shown, the vehicle braking control method includes the following steps:

[0030] S110: Acquire current vehicle driving information of the target vehicle.

[0031] The target vehicle may be a vehicle that needs to be braked. The current vehicle driving information may be driving information of the target vehicle in its current state. Exemplarily, the vehicle driving information includes information such as vehicle driving speed, wheel speed, and yaw angular velocity. Specifically, the real-time driving information of the target vehicle may be collected based on a preset driving information collection sensor to obtain the current vehicle driving information.

[0032] S120: Input the current vehicle driving information into a pre-trained target braking decision model to obtain target braking control information corresponding to the current vehicle driving information.

[0033] The target braking decision model may be a model ultimately used to determine the braking information of the target vehicle. The braking information may be information for controlling the braking components of the vehicle. By controlling the braking components of the vehicle, the driving information of the target vehicle can be adjusted to improve the driving safety of the vehicle.

[0034] Specifically, during the training process of the target braking decision model, a target reward function can be constructed based on the wheel slip rate and body deflection angle of the target vehicle, and then the sample data is input into the untrained preliminary braking decision model to obtain the corresponding sample prediction result, and then the reward value corresponding to the sample prediction result is determined based on the target reward function, and then the model parameters of the preliminary braking decision model are adjusted based on the reward value to obtain the target braking decision model. The technical solution of the embodiment of the present invention can be based on the reward function constructed based on the wheel slip rate and body deflection angle, and can integrate the reward function of the wheel slip rate and body deflection angle to determine the model, enrich the reference factors of the model reward function, improve the robustness of the model trained based on the reward function, improve the accuracy of the vehicle braking control information output by the model, and improve the vehicle safety performance.

[0035] Furthermore, the target braking control information may be information for controlling the braking components of the target vehicle. Specifically, the current vehicle driving information may be input into a pre-trained target braking decision model, so that the target braking decision model analyzes the current vehicle driving information and outputs the corresponding target braking control information.

[0036] S130: Adjust current vehicle driving information of the target vehicle according to the target braking control information.

[0037] After the target brake control information is determined, the control parameters of the corresponding control components can be generated according to the target brake control information, and the control parameters of the control components can be adjusted to adjust the current vehicle driving information. The brake components of the target vehicle include at least one of a return oil pump, an oil outlet valve, an oil inlet valve, a reversing valve, and a high-pressure reversing valve.

[0038] The technical solution provided by the embodiment of the present invention obtains the current vehicle driving information of the target vehicle, the current vehicle driving information includes: at least one of the vehicle driving speed, wheel speed, and deflection angular velocity; inputs the current vehicle driving information into a pre-trained target braking decision model to obtain the target braking control information corresponding to the current vehicle driving information; adjusts the current vehicle driving information of the target vehicle according to the target braking control information; the target braking decision model is obtained by reinforcement learning based on the target reward function; the target reward function is constructed based on the wheel slip rate and body deflection angle of the target vehicle. The technical solution of the embodiment of the present invention solves the problem of insufficient robustness of the vehicle braking control model in the prior art, which leads to the problem of insufficient accuracy of the vehicle braking control information output by the model. The wheel slip rate and body deflection angle can be used as reward functions to train the vehicle braking control model, improve the robustness of the model, improve the accuracy of the vehicle braking control information output by the model, and improve the vehicle safety performance.

[0039] Figure 2 This is another vehicle braking control method flow chart provided by an embodiment of the present invention. The embodiment of the present invention can be applied to the scene of controlling the automatic adjustment of driving information of a mobile vehicle. Based on the above embodiment, this embodiment further explains how to construct the reward function of the model; how to perform reinforcement learning based on the target reward function to obtain the target braking decision model; and how to adjust the current vehicle driving information of the target vehicle according to the target braking control information. The device can be implemented by software and / or hardware and integrated into a computer device with application development function.

[0040] like Figure 2 As shown, the vehicle braking control method includes the following steps:

[0041] S210, obtaining a preliminary braking decision model and preset vehicle driving sample information, inputting the preset vehicle driving sample information into the preliminary braking decision model, and obtaining sample braking control information.

[0042] Among them, the preliminary braking decision model can be an untrained model for determining the braking information of the target vehicle. Optionally, a convolutional neural network or a recurrent neural network can be used as the preliminary braking decision model. The preset vehicle driving sample information can be preset driving information about the target vehicle. Specifically, the driving information includes information such as vehicle driving speed, wheel speed, and vehicle body deflection angle. Furthermore, the driving information of the target vehicle in different simulation experiment scenarios can be collected to obtain the preset vehicle driving sample information. The preset vehicle driving sample information can be used as sample data to train the preliminary braking decision model.

[0043] The sample braking control information may be braking information corresponding to the preset vehicle driving sample information. The braking information may be information for controlling the vehicle braking component. By controlling the vehicle braking component, the driving information of the target vehicle is adjusted to improve the driving safety of the vehicle. Specifically, the preset vehicle driving sample information may be input into the preliminary braking decision model, and then the sample braking control information is output.

[0044] S220: Determine a target reward function according to the vehicle body deflection angle and wheel slip rate corresponding to the sample braking control information.

[0045] The body deflection angle may be the whole vehicle deflection angle of the target vehicle after the sample braking control information is executed. The wheel slip rate may be the slip rate of the wheel after the sample braking control information is executed. Specifically, the body deflection angle and the wheel slip rate may be calculated by the sample braking control information. Further, the target reward function may be a function for calculating the reward value corresponding to the sample braking control information. Specifically, the target reward function may be determined by the body deflection angle and the wheel slip rate.

[0046] Optionally, the target reward function is determined based on the vehicle body deflection angle and wheel slip rate corresponding to the sample braking control information, including: determining the slip rate reward function corresponding to the wheels of the target vehicle based on the wheel slip rate corresponding to the sample braking control information; and determining the target reward function based on the slip rate reward function and a preset deflection angle reward function.

[0047] Among them, the slip rate reward function can be a reward function determined from the perspective of wheel slip rate. Specifically, the wheel slip rate angles at different stages can have corresponding slip rate reward functions. Further, the preset deflection angle reward function can be a preset reward function determined from the perspective of the vehicle body deflection angle. Further, the slip rate reward function and the preset deflection angle reward function can be multiplied by corresponding weight coefficients and then added together to obtain the target reward function.

[0048] Optionally, a slip rate reward function corresponding to the wheel of the target vehicle is determined based on the wheel slip rate corresponding to the sample braking control information, including: when the wheel slip rate is less than a preset first slip rate threshold, determining a first weight value based on the vehicle body deflection angle, and weighting the wheel slip rate based on the first weight value to obtain a slip rate reward function; when the wheel slip rate is greater than a preset second slip rate threshold, weighting the wheel slip rate based on the second weight value to obtain a slip rate reward function; when the wheel slip rate is greater than the preset first slip rate threshold and less than the second slip rate threshold, using a preset constant value as the slip rate reward function.

[0049] The preset first slip rate threshold may be a preset smaller threshold value for wheel slip rate. The first weight value may be a weight value for wheel slip rate corresponding to the preset first slip rate threshold. Specifically, when the wheel slip rate is less than the preset first slip rate threshold, the first weight value may be determined according to the vehicle body deflection angle, and the wheel slip rate may be weighted according to the first weight value to obtain a slip rate reward function.

[0050] Correspondingly, the preset second slip rate threshold may be a preset larger threshold value regarding the wheel slip rate. The preset second slip rate threshold is greater than the first slip rate threshold. The second weight value may be a weight value regarding the wheel slip rate corresponding to the preset second slip rate threshold. Specifically, when the wheel slip rate is greater than the preset second slip rate threshold, the second weight value may be determined according to the vehicle body deflection angle, and the wheel slip rate may be weighted according to the second weight value to obtain a slip rate reward function. In addition, when the wheel slip rate is greater than the preset first slip rate threshold and less than the second slip rate threshold, the preset constant value may be directly used as the slip rate reward function.

[0051] S230. Determine a sample reward parameter corresponding to the sample braking control information based on the sample braking control information and the target reward function.

[0052] The sample reward parameter may be a reward value corresponding to the sample braking control information. Specifically, the braking parameter corresponding to the sample braking control information may be substituted into the target reward function to obtain the sample reward parameter.

[0053] Optionally, based on the sample braking control information and the target reward function, sample reward parameters of the sample braking control information are determined, including: determining the slip rate reward parameter based on the wheel slip rate and the slip rate reward function corresponding to the sample braking control information; determining the deflection angle reward parameter based on the wheel slip rate and the preset deflection angle reward function corresponding to the sample braking control information; determining the sample reward parameter based on the slip rate reward parameter and the deflection angle reward parameter.

[0054] Among them, the target reward function is determined by the slip rate reward function and the preset deflection angle reward function, so when determining the reward value, it can also be determined from the angles of the wheel slip rate and the body deflection angle respectively. Among them, the slip rate reward parameter can be a reward value determined from the wheel slip rate angle. Specifically, the wheel slip rate corresponding to the sample braking control information can be substituted into the slip rate reward function to obtain the slip rate reward parameter. Further, the deflection angle reward parameter can be a reward value determined from the body deflection angle angle. Specifically, the body deflection angle corresponding to the sample braking control information can be substituted into the preset deflection angle reward function to determine the deflection angle reward parameter. Finally, the slip rate reward parameter and the deflection angle reward parameter can be multiplied by the corresponding weight coefficients and then added together to obtain the sample reward parameter.

[0055] Optionally, a deflection angle reward parameter is determined based on the vehicle body deflection angle corresponding to the sample braking control information and the preset deflection angle reward function, including: determining a target deflection angle based on the vehicle speed and steering wheel angle of the target vehicle corresponding to the sample braking control information; determining a deflection angle difference based on the target deflection angle and the vehicle body deflection angle, and determining the deflection angle reward parameter based on the deflection angle difference.

[0056] Among them, the target deflection angle can be a reference standard deflection angle value corresponding to the sample braking control information. It can be understood that it is more optimal when the deflection angle of the target vehicle is the target deflection angle. Specifically, the vehicle speed and steering wheel angle of the target vehicle can be determined based on the sample braking control information, and then the target deflection angle can be further determined based on the vehicle speed and the steering wheel angle. Furthermore, the deflection angle difference can be the difference between the target deflection angle and the wheel deflection angle. After determining the deflection angle difference, the deflection angle reward parameter can be determined based on the deflection angle difference.

[0057] Exemplarily, the slip rate reward function is set to reward_slip. When the slip rate is greater than 0.2, reward_slip=-1*slip rate; when the slip rate is less than 0.05, reward_slip=-1*slip rate*(0.1-slip rate); when the slip rate is in the rest of the range, reward_slip=0.

[0058] Correspondingly, target yaw velocity = vehicle speed * sin function (steering wheel angle * π / 180 / 20.12) / (vehicle wheelbase * 3.6 * pi) * 180, yaw angle reward function reward_yaw = the absolute value of the actual yaw angle velocity and the target yaw velocity multiplied by -0.02. Total reward parameter reward = reward_slip + reward_yaw.

[0059] S240: Adjust the preliminary braking decision model according to the sample reward parameter to obtain a target braking decision model.

[0060] The target braking decision model may be a model ultimately used to determine the braking information of the target vehicle. Specifically, after obtaining the sample reward parameter, the model parameters of the preliminary braking decision model may be adjusted according to the sample reward parameter to obtain the target braking decision model.

[0061] Exemplarily, the process of training a model based on reinforcement learning includes the following steps:

[0062] 1. Application of reinforcement learning algorithm:

[0063] Discrete SAC algorithm: The discrete SAC algorithm is applied in a virtual environment, and the action space is defined as the valve action of ESP (Electronic Stability Program), including four discrete action values: pressure increase, pressure maintenance, pressure relief, and non-operation.

[0064] State and action definition: Take vehicle speed, four-wheel speed, and yaw angular velocity as states, and ESP valve action as discrete action values ​​to establish a state-action mapping relationship.

[0065] 2. Control the four-wheel braking force curve:

[0066] ESP valve action is associated with braking force: The ESP valve action changes the four-wheel braking force distribution. For example, boosting will increase the braking force of the corresponding wheel, maintaining pressure will maintain the current braking force, releasing pressure will reduce the braking force, and not working will not affect the braking force.

[0067] The influence of braking force on wheel speed, vehicle speed and yaw rate: By controlling the braking force of the four wheels, the wheel speed, vehicle speed and yaw rate are affected. For example, increasing the braking force of the front wheels can slow down the vehicle speed, reduce the wheel speed and change the yaw rate.

[0068] 3. Reward function design:

[0069] Optimization goal: Design a reward function to keep the four-wheel slip rate as low as possible below 0.2 while keeping the angular velocity as small as possible. This is used as a goal during training to guide the reinforcement learning model to learn the optimized braking control strategy.

[0070] 4. Strengthen the model structure diagram and ESP underlying logic. For example, Figure 3 FIG. 1 is a flowchart of the underlying logic of an ESP provided by an embodiment of the present invention. Figure 3As shown, the figure includes: control unit (ESP bottom layer) 110; oil return pump 120; oil outlet valve (normally closed) 130; oil inlet valve (normally open) 140; reversing valve (normally open) 150; high-pressure reversing valve (normally closed) 160; wheel end caliper pressure sensor 170; wheel speed sensor 180; vehicle speed 190; deflection angular velocity 200; reinforcement learning model 210 (strategy model, output action boost, pressure maintenance, pressure relief, not working). Here we only use one wheel control as an example, the valve body default state instruction is 0, and the working state is 1.

[0071] ESP controls the valve body and pump to increase, decrease and maintain the wheel pressure: Wheel pressure boost control: return oil pump works 80%, outflow valve 0, inflow valve 0, reversing valve 1, high-pressure reversing valve 0; Wheel pressure relief control: return oil pump works 50%, outflow valve 1, inflow valve 1, reversing valve 1, high-pressure reversing valve 0; Wheel pressure maintaining control: return oil pump works 50%, outflow valve 1, inflow valve 0, reversing valve 1, high-pressure reversing valve 0; Wheel pressure not working: all parameters are 0.

[0072] S250, obtaining current vehicle driving information of the target vehicle, and inputting the current vehicle driving information into a pre-trained target braking decision model to obtain target braking control information corresponding to the current vehicle driving information.

[0073] Among them, the current vehicle driving information may be the driving information of the target vehicle in the current state. Exemplarily, the vehicle driving information includes information such as the vehicle driving speed, wheel speed, and deflection angular velocity. Specifically, the real-time driving information of the target vehicle may be collected based on a preset driving information collection sensor to obtain the current vehicle driving information. Exemplarily, data such as vehicle speed, wheel speed, acceleration, deflection angular velocity, steering wheel angle, and ESP valve body movement may be collected through vehicle sensors. The collected data may also be cleaned and processed to ensure data quality and accuracy.

[0074] Furthermore, the target braking control information may be information for controlling the braking components of the target vehicle. Specifically, the current vehicle driving information may be input into a pre-trained target braking decision model, so that the target braking decision model analyzes the current vehicle driving information and outputs the corresponding target braking control information.

[0075] S260: Determine a target braking control parameter according to the target braking control information, and control a vehicle braking component of the target vehicle according to the target braking control parameter to adjust the current vehicle driving information.

[0076] The target braking control parameter may be a parameter for controlling a vehicle braking component. The vehicle braking component includes at least one of an oil return pump, an oil outlet valve, an oil inlet valve, a reversing valve, and a high-pressure reversing valve. Specifically, the control parameters of each vehicle braking component may be determined according to the target braking control information, thereby obtaining the target braking control parameter. Furthermore, the vehicle braking component of the target vehicle may be controlled based on the target braking control parameter, thereby adjusting the current vehicle driving information of the target vehicle and improving the vehicle driving safety.

[0077] For example, Figure 4 FIG. 1 is a flowchart of a vehicle braking control process provided by an embodiment of the present invention. Figure 4 As shown, the workflow of vehicle braking control includes the following steps:

[0078] S110, data collection: The first step is to install sensors on the car to collect data related to the braking system, such as vehicle speed, wheel speed, brake pedal position, vehicle acceleration, etc. These data will be transmitted to the data processing unit in real time.

[0079] S120, data processing: the data processing unit pre-processes the collected data, including filtering, denoising, calibration and other operations, to ensure the accuracy and consistency of the data.

[0080] S130, Machine Learning: Use machine learning algorithms to analyze the processed data to learn braking behavior in different driving scenarios. This can include supervised learning, reinforcement learning, or unsupervised learning, depending on specific needs.

[0081] S140, Environment Simulation: Using the learned model, create a virtual driving environment. This can include simulating different road types, traffic conditions, weather conditions, etc. The simulator needs to be able to generate a variety of driving scenarios for testing.

[0082] S150, Learning Strategy: The machine learning model will develop strategies to simulate different braking operations, taking into account environmental conditions and safety requirements. This includes the distribution of braking force, the operation time of the brakes, etc.

[0083] S160, Model Generation: After data collection and data cleaning, appropriate features are extracted or created from the data for model training. This can include operations such as feature selection, dimensionality reduction, and transformation. According to the nature of the task and the characteristics of the data, select an appropriate model architecture, such as a neural network, decision tree, support vector machine, etc. Initialize the weights and parameters of the model, usually using random initial values. Use the data of the training set and the current parameters of the model for forward propagation to calculate the predicted output of the model. Then design a penalty function (or cost function) to measure the gap between the model's prediction and the actual value. Common penalty functions include mean square error (MSE), cross entropy, etc. Use optimization algorithms (such as stochastic gradient descent, Adam, etc.) to update the weights and parameters of the model based on the gradient information. After repeated iterative training and simulation and parameter optimization, the machine learning model will generate a model that describes the desired behavior of the automotive braking system. The model training process may take hours, days, or even weeks, depending on the complexity of the model, the amount of data, and computing resources.

[0084] S170, Migrate to Embedded Hardware: Through mixed development in C and C++, migrate the generated model to an embedded control system that can respond and adjust the operation of the braking system in real time. This may require optimization to suit the performance requirements of the embedded hardware.

[0085] S180, Evaluation of iterative development: Through the transplanted AI-embedded electronic hardware control system, real-time data feedback and control interaction are carried out with the previously obtained environmental simulation model. The embedded AI model can predict the effect of the next control sequence in real time and give the current control strategy. The control response effect of the current control system is obtained through the simulation environment. According to the response effect, iterative development is carried out to obtain the optimal AI strategy model.

[0086] The technical solution provided by the embodiment of the present invention has the following beneficial effects:

[0087] 1. Improve braking system performance: Through machine learning and intelligent control, embodiments of the present invention can automatically optimize the performance of the braking system, making it perform better in different driving scenarios, including emergency braking, low-adhesion roads and complex driving conditions.

[0088] 2. Increase driving safety: The environmental simulation of the embodiments of the present invention allows comprehensive testing of the performance of the braking system in various scenarios to reduce the risk of accidents and improve the safety of drivers and passengers.

[0089] 3. Adaptive control: The real-time responsive control of the embedded control system ensures that the braking system can adaptively adjust according to the context of the virtual environment to ensure optimal performance.

[0090] In general, the embodiments of the present invention combine traditional braking systems with modern embedded and artificial intelligence technologies to bring higher performance, safety and adaptability to automotive braking systems. This not only helps improve road safety, but also provides a more comfortable driving experience for drivers and passengers.

[0091] The technical solution provided by the embodiment of the present invention obtains the preliminary braking decision model and the preset vehicle driving sample information, inputs the preset vehicle driving sample information into the preliminary braking decision model, and obtains the sample braking control information; determines the target reward function according to the vehicle body deflection angle and wheel slip rate corresponding to the sample braking control information; determines the sample reward parameter corresponding to the sample braking control information based on the sample braking control information and the target reward function; adjusts the preliminary braking decision model according to the sample reward parameter to obtain the target braking decision model; inputs the current vehicle driving information into the pre-trained target braking decision model to obtain the target braking control information corresponding to the current vehicle driving information; determines the target braking control parameter according to the target braking control information; controls the vehicle braking component of the target vehicle according to the target braking control parameter to adjust the current vehicle driving information. The technical solution of the embodiment of the present invention solves the problem of insufficient robustness of the vehicle braking control model in the prior art, which leads to the problem of insufficient accuracy of the vehicle braking control information output by the model. The wheel slip rate and the vehicle body deflection angle can be used as reward functions to train the vehicle braking control model, improve the model robustness, improve the accuracy of the vehicle braking control information output by the model, and improve the vehicle safety performance.

[0092] Figure 5 It is a structural schematic diagram of a vehicle braking control device provided by an embodiment of the present invention. The embodiment of the present invention can be applied to the scenario of controlling a mobile vehicle to automatically adjust driving information. The device can be implemented by software and / or hardware and integrated into a computer device with application development function.

[0093] like Figure 5 As shown, the vehicle braking control device includes: a vehicle driving information acquisition module 310, a braking control information determination module 320 and a vehicle driving information adjustment module 330.

[0094] Among them, the vehicle driving information acquisition module 310 is used to obtain the current vehicle driving information of the target vehicle, wherein the current vehicle driving information includes: at least one of the vehicle driving speed, wheel speed, and deflection angular velocity; the braking control information determination module 320 is used to input the current vehicle driving information into a pre-trained target braking decision model to obtain the target braking control information corresponding to the current vehicle driving information; the vehicle driving information adjustment module 330 is used to adjust the current vehicle driving information of the target vehicle according to the target braking control information; wherein the target braking decision model is obtained by reinforcement learning based on the target reward function; the target reward function is constructed based on the wheel slip rate and body deflection angle of the target vehicle.

[0095] The technical solution provided by the embodiment of the present invention obtains the current vehicle driving information of the target vehicle, the current vehicle driving information includes: at least one of the vehicle driving speed, wheel speed, and deflection angular velocity; inputs the current vehicle driving information into a pre-trained target braking decision model to obtain the target braking control information corresponding to the current vehicle driving information; adjusts the current vehicle driving information of the target vehicle according to the target braking control information; the target braking decision model is obtained by reinforcement learning based on the target reward function; the target reward function is constructed based on the wheel slip rate and body deflection angle of the target vehicle. The technical solution of the embodiment of the present invention solves the problem of insufficient robustness of the vehicle braking control model in the prior art, which leads to the problem of insufficient accuracy of the vehicle braking control information output by the model. The wheel slip rate and body deflection angle can be used as reward functions to train the vehicle braking control model, improve the robustness of the model, improve the accuracy of the vehicle braking control information output by the model, and improve the vehicle safety performance.

[0096] In an optional embodiment, the vehicle braking control device also includes: a model training module, which is used to: obtain a preliminary braking decision model and preset vehicle driving sample information; input the preset vehicle driving sample information into the preliminary braking decision model to obtain sample braking control information; determine the target reward function according to the vehicle body deflection angle and wheel slip rate corresponding to the sample braking control information; determine the sample reward parameters corresponding to the sample braking control information based on the sample braking control information and the target reward function; adjust the preliminary braking decision model according to the sample reward parameters to obtain the target braking decision model.

[0097] In an optional embodiment, the model training module includes: a target reward function determination submodule, used to: determine the slip rate reward function corresponding to the wheel of the target vehicle according to the wheel slip rate corresponding to the sample braking control information; determine the target reward function according to the slip rate reward function and a preset deflection angle reward function.

[0098] In an optional embodiment, the target reward function determination submodule includes: a slip rate reward function determination unit, which is used to: when the wheel slip rate is less than a preset first slip rate threshold, determine a first weight value according to the vehicle body deflection angle, and weight the wheel slip rate according to the first weight value to obtain the slip rate reward function; when the wheel slip rate is greater than a preset second slip rate threshold, weight the wheel slip rate based on the second weight value to obtain the slip rate reward function; when the wheel slip rate is greater than the preset first slip rate threshold and less than the second slip rate threshold, use a preset constant value as the slip rate reward function.

[0099] In an optional embodiment, the model training module includes: a sample reward parameter determination submodule, which is used to: determine the slip rate reward parameter based on the wheel slip rate corresponding to the sample braking control information and the slip rate reward function; determine the deflection angle reward parameter based on the vehicle body deflection angle corresponding to the sample braking control information and the preset deflection angle reward function; determine the sample reward parameter based on the slip rate reward parameter and the deflection angle reward parameter.

[0100] In an optional embodiment, the sample reward parameter determination submodule includes: a deflection angle reward parameter determination unit, used to: determine the target deflection angle according to the vehicle speed and steering wheel angle of the target vehicle corresponding to the sample braking control information; determine the deflection angle difference according to the target deflection angle and the vehicle body deflection angle, and determine the deflection angle reward parameter according to the deflection angle difference.

[0101] In an optional embodiment, the vehicle driving information adjustment module 330 is specifically used to: determine the target braking control parameters according to the target braking control information; control the vehicle braking components of the target vehicle according to the target braking control parameters to adjust the current vehicle driving information; wherein the vehicle braking components include: at least one of: a return oil pump, an oil outlet valve, an oil inlet valve, a reversing valve, and a high-pressure reversing valve.

[0102] The vehicle braking control device provided in the embodiment of the present invention can execute the vehicle braking control method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0103] Figure 6 A schematic diagram of the structure of a computer device provided in an embodiment of the present invention. Figure 6 A block diagram of an exemplary computer device 12 suitable for use in implementing embodiments of the present invention is shown. Figure 6The computer device 12 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention. The computer device 12 can be any terminal device with computing capability and can be configured in the vehicle brake control device.

[0104] like Figure 6 As shown, the computer device 12 is in the form of a general-purpose computing device. The components of the computer device 12 may include, but are not limited to: one or more processors or processing units 16, a system memory 28, and a bus 18 that connects various system components (including the system memory 28 and the processing unit 16).

[0105] The bus 18 may be one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor or a local bus using any of a variety of bus architectures. For example, these architectures include, but are not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MAC) bus, an Enhanced ISA bus, a Video Electronics Standards Association (VESA) local bus, and a Peripheral Component Interconnect (PCI) bus.

[0106] The computer device 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the computer device 12, including volatile and non-volatile media, removable and non-removable media.

[0107] The system memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache 32. The computer device 12 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system 34 may be used to read and write non-removable, non-volatile magnetic media ( Figure 6 not shown, usually called a "hard drive"). Although Figure 6 Not shown in the figure, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, a DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to the bus 18 via one or more data medium interfaces. The system memory 28 may include at least one program product having a set (e.g., at least one) of program modules that are configured to perform the functions of the various embodiments of the present invention.

[0108] A program / utility 40 having a set (at least one) of program modules 42 may be stored, for example, in system memory 28, such program modules 42 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment. Program modules 42 generally perform the functions and / or methods of the embodiments described herein.

[0109] The computer device 12 may also communicate with one or more external devices 14 (e.g., keyboards, pointing devices, displays 24, etc.), one or more devices that enable a user to interact with the computer device 12, and / or any device that enables the computer device 12 to communicate with one or more other computing devices (e.g., network cards, modems, etc.). Such communication may be performed through an input / output (I / O) interface 22. Furthermore, the computer device 12 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) through a network adapter 20. Figure 6 As shown, the network adapter 20 communicates with other modules of the computer device 12 via the bus 18. It should be understood that although Figure 6 Not shown, other hardware and / or software modules may be used in conjunction with computer device 12, including but not limited to microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0110] The processing unit 16 executes various functional applications and data processing by running the program stored in the system memory 28, for example, implementing the vehicle braking control method provided by the embodiment of the present invention, the method comprising:

[0111] Acquire current vehicle driving information of the target vehicle, wherein the current vehicle driving information includes: at least one of the vehicle driving speed, wheel speed, and deflection angular velocity;

[0112] Inputting the current vehicle driving information into a pre-trained target braking decision model to obtain target braking control information corresponding to the current vehicle driving information;

[0113] adjusting current vehicle driving information of the target vehicle according to the target braking control information;

[0114] Wherein, the target braking decision model is obtained by reinforcement learning based on a target reward function; the target reward function is constructed based on the wheel slip rate and body deflection angle of the target vehicle.

[0115] This embodiment provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, the vehicle braking control method provided by any embodiment of the present invention is implemented, including:

[0116] Acquire current vehicle driving information of the target vehicle, wherein the current vehicle driving information includes: at least one of the vehicle driving speed, wheel speed, and deflection angular velocity;

[0117] Inputting the current vehicle driving information into a pre-trained target braking decision model to obtain target braking control information corresponding to the current vehicle driving information;

[0118] adjusting current vehicle driving information of the target vehicle according to the target braking control information;

[0119] Wherein, the target braking decision model is obtained by reinforcement learning based on a target reward function; the target reward function is constructed based on the wheel slip rate and body deflection angle of the target vehicle.

[0120] The computer storage medium of the embodiment of the present invention can adopt any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to: an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer 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. In this document, a computer-readable storage medium can be any tangible medium containing or storing a program, which can be used by an instruction execution system, device or device or used in combination with it.

[0121] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0122] The program code embodied on the computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0123] Computer program code for performing the operations of the present invention may be written in one or more programming languages ​​or combinations thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0124] It should be understood by those skilled in the art that the modules or steps of the present invention described above can be implemented by a general-purpose computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, optionally, they can be implemented by a program code executable by a computer device, so that they can be stored in a storage device and executed by the computing device, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0125] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A vehicle braking control method, characterized in that: include: Acquire current vehicle driving information of the target vehicle, wherein the current vehicle driving information includes: at least one of the vehicle driving speed, wheel speed, and deflection angular velocity; Inputting the current vehicle driving information into a pre-trained target braking decision model to obtain target braking control information corresponding to the current vehicle driving information; adjusting current vehicle driving information of the target vehicle according to the target braking control information; The training process of the target braking decision model includes: Obtaining a preliminary braking decision model and preset vehicle driving sample information; Inputting the preset vehicle driving sample information into the preliminary braking decision model to obtain sample braking control information; determining a target reward function according to a vehicle body deflection angle and a wheel slip rate corresponding to the sample braking control information; Based on the sample braking control information and the target reward function, determining a sample reward parameter corresponding to the sample braking control information; Adjusting the preliminary braking decision model according to the sample reward parameter to obtain the target braking decision model; Wherein, determining the target reward function according to the vehicle body deflection angle and wheel slip rate corresponding to the sample braking control information includes: Determining a slip rate reward function corresponding to the wheels of the target vehicle according to the wheel slip rate corresponding to the sample braking control information; Determining the target reward function according to the slip rate reward function and a preset deflection angle reward function; wherein the preset deflection angle reward function is a preset reward function determined from the perspective of the vehicle body deflection angle; Wherein, determining the sample reward parameter of the sample braking control information based on the sample braking control information and the target reward function includes: determining a slip rate reward parameter according to the wheel slip rate corresponding to the sample braking control information and the slip rate reward function; Determining a deflection angle reward parameter according to the vehicle body deflection angle corresponding to the sample braking control information and the preset deflection angle reward function; The sample reward parameter is determined according to the slip rate reward parameter and the deflection angle reward parameter.

2. The method according to claim 1, characterized in that The step of determining the slip rate reward function corresponding to the wheel of the target vehicle according to the wheel slip rate corresponding to the sample braking control information includes: When the wheel slip rate is less than a preset first slip rate threshold, determining a first weight value according to the vehicle body deflection angle, and weighting the wheel slip rate according to the first weight value to obtain the slip rate reward function; When the wheel slip rate is greater than a preset second slip rate threshold, weighting the wheel slip rate based on a second weight value to obtain the slip rate reward function; When the wheel slip rate is greater than the preset first slip rate threshold and less than the preset second slip rate threshold, a preset constant value is used as the slip rate reward function.

3. The method according to claim 1, characterized in that The step of determining a deflection angle reward parameter according to the vehicle body deflection angle corresponding to the sample braking control information and the preset deflection angle reward function comprises: determining a target deflection angle according to a vehicle speed and a steering wheel angle of a target vehicle corresponding to the sample brake control information; A deflection angle difference is determined according to the target deflection angle and the vehicle body deflection angle, and the deflection angle reward parameter is determined according to the deflection angle difference.

4. The method according to claim 1, characterized in that: The adjusting the current vehicle driving information of the target vehicle according to the target braking control information includes: determining a target braking control parameter according to the target braking control information; Controlling the vehicle brake assembly of the target vehicle according to the target brake control parameter to adjust the current vehicle driving information; Wherein, the vehicle brake assembly includes: at least one of an oil return pump, an oil outlet valve, an oil inlet valve, a reversing valve, and a high-pressure reversing valve.

5. A vehicle brake control device, characterized in that: The device comprises: A vehicle driving information acquisition module, used to acquire current vehicle driving information of a target vehicle, wherein the current vehicle driving information includes: at least one of vehicle driving speed, wheel speed, and deflection angular velocity; A braking control information determination module, used for inputting the current vehicle driving information into a pre-trained target braking decision model to obtain target braking control information corresponding to the current vehicle driving information; A vehicle driving information adjustment module, used for adjusting the current vehicle driving information of the target vehicle according to the target braking control information; The vehicle braking control device further includes: a model training module, which is used to: obtain a preliminary braking decision model and preset vehicle driving sample information; input the preset vehicle driving sample information into the preliminary braking decision model to obtain sample braking control information; determine a target reward function according to a vehicle body deflection angle and a wheel slip rate corresponding to the sample braking control information; determine a sample reward parameter corresponding to the sample braking control information based on the sample braking control information and the target reward function; adjust the preliminary braking decision model according to the sample reward parameter to obtain the target braking decision model; The model training module includes: a target reward function determination submodule, which is used to: determine the slip rate reward function corresponding to the wheel of the target vehicle according to the wheel slip rate corresponding to the sample braking control information; determine the target reward function according to the slip rate reward function and a preset deflection angle reward function; The model training module includes: a sample reward parameter determination submodule, which is used to: determine the slip rate reward parameter according to the wheel slip rate corresponding to the sample braking control information and the slip rate reward function; determine the deflection angle reward parameter according to the vehicle body deflection angle corresponding to the sample braking control information and the preset deflection angle reward function; wherein the preset deflection angle reward function is a preset reward function determined from the perspective of the vehicle body deflection angle; determine the sample reward parameter according to the slip rate reward parameter and the deflection angle reward parameter.

6. A computer device, characterized in that: The computer device comprises: one or more processors; A memory for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the vehicle braking control method as described in any one of claims 1-4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the vehicle braking control method as described in any one of claims 1 to 4 is implemented.

Images