Automobile braking method and device, automobile and equipment

Abstract

The embodiment of the invention discloses an automobile braking method and device, an automobile and equipment, and belongs to the technical field of automobiles. In the automobile driving process of a driver, the driver is recognized in a biological feature recognition mode so that a speed prediction model matched with the driving habit of the driver can be obtained, and the driving habit of the driver can be predicted. In this way, in the automobile driving process of a driver, if braking occurs, the deceleration conforming to the driving habit of the driver is predicted through the speed prediction model on the basis of the braking information, and the automobile is controlled to decelerate according to the deceleration, so that personalized configuration of the braking mode is achieved, the automobile can be braked according to the driving habit of the driver, and the driving experience of the driver is improved. Seamless joint of the braking response characteristics and inherent habits of a driver is achieved, the driving comfort and controllability are greatly improved, and then the braking experience of the driver is improved.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of automobile technology, and in particular to an automobile braking method, device, automobile, and equipment. Background Art

[0002] Cars are equipped with a brake pedal. While driving, drivers apply the brakes to decelerate the car. In current cars, the brake pedal and brake actuator are connected through a simple mechanical connection and hydraulic pressure, resulting in a relatively fixed relationship between the vehicle's deceleration and the brake pedal's travel. However, this fixed relationship is difficult to adapt to drivers' driving habits, resulting in a poor braking experience. Summary of the Invention

[0003] The embodiments of the present application provide a vehicle braking method, device, vehicle, and equipment that can enhance the driver's braking experience. The technical solution is as follows:

[0004] In one aspect, a method for braking an automobile is provided, the method comprising:

[0005] Performing biometric identification on the driver of the vehicle to obtain the driver's identity information;

[0006] During the driving of the vehicle, in response to a braking operation, obtaining braking information, the braking information including a travel of a brake pedal in the vehicle;

[0007] predicting a first deceleration of the vehicle based on the braking information using a speed prediction model corresponding to the identity information, the speed prediction model being trained based on historical driving data corresponding to the identity information, the historical driving data including historical braking information of the driver during historical driving of the vehicle and deceleration corresponding to the historical braking information;

[0008] Based on the first deceleration, the vehicle is controlled to decelerate.

[0009] In another aspect, a vehicle braking device is provided, comprising:

[0010] An identification module, configured to perform biometric identification on the driver of the vehicle to obtain the driver's identity information;

[0011] an acquisition module, configured to acquire braking information in response to a braking operation during the driving of the vehicle, wherein the braking information includes a travel of a brake pedal in the vehicle;

[0012] a prediction module, configured to predict a first deceleration of the vehicle based on the braking information using a speed prediction model corresponding to the identity information, the speed prediction model being trained based on historical driving data corresponding to the identity information, the historical driving data including historical braking information of the driver during historical driving of the vehicle and decelerations corresponding to the historical braking information;

[0013] A control module is configured to control the vehicle to decelerate based on the first deceleration.

[0014] On the other hand, a car is provided, comprising a processor and a memory, wherein the memory stores at least one computer program, and the at least one computer program is loaded and executed by the processor to implement the operations performed by the car braking method described in the above aspect.

[0015] On the other hand, a computer device is provided, which includes a processor and a memory, wherein the memory stores at least one computer program, and the at least one computer program is loaded and executed by the processor to implement the operations performed by the vehicle braking method described in the above aspect.

[0016] In the solution provided in the embodiment of the present application, while the driver is driving the car, biometric recognition is used to identify the driver so that a speed prediction model that matches the driver's driving habits can be obtained. In this way, if braking occurs during the driver's driving of the car, the speed prediction model is used to predict the deceleration that matches the driver's driving habits based on the braking information, and the car is controlled to decelerate according to the deceleration. In this way, personalized configuration of the braking method is achieved, and the car can be braked according to the driver's driving habits, achieving seamless connection between the braking response characteristics and the driver's inherent habits, greatly improving driving comfort and controllability, and thus improving the driver's braking experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0018] Figure 1 This is a structural diagram of an automobile braking system provided by an embodiment of the present application;

[0019] Figure 2 This is a flow chart of a vehicle braking method provided by an embodiment of the present application;

[0020] Figure 3 This is a flow chart of another automobile braking method provided by an embodiment of the present application;

[0021] Figure 4 This is a flow chart of another automobile braking method provided by an embodiment of the present application;

[0022] Figure 5 It is a structural schematic diagram of an automobile braking device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0023] In order to make the objectives, technical solutions and advantages of the embodiments of the present application clearer, the implementation methods of the present application will be further described in detail below with reference to the accompanying drawings.

[0024] As used herein, the terms "first," "second," and the like may be used to describe various concepts, but unless otherwise specified, these concepts are not limited by these terms. These terms are used solely to distinguish one concept from another. For example, a first deceleration may be referred to as a second deceleration, and similarly, a second deceleration may be referred to as a first deceleration without departing from the scope of this application.

[0025] It should be noted that the information (including but not limited to user device information, user personal information, etc.), data (including but not limited to data used for analysis, storage, display, etc.), and signals involved in this application are all authorized by the user or fully authorized by all parties, and the collection, use, and processing of relevant data must comply with the relevant laws, regulations, and standards of the relevant countries and regions. For example, the identity information, braking information, driving data, etc. involved in this application are all obtained with full authorization.

[0026] The vehicle braking method provided in the embodiments of the present application can be performed by a vehicle or a computer device. Optionally, the computer device is provided as a terminal or a server. Optionally, the terminal is a computer, a mobile phone, a tablet computer, or other terminal. Optionally, the server is a single server, multiple servers, or a cloud computing platform.

[0027] In some embodiments, the computer device is taken as a server as an example. Figure 1 This is a structural diagram of an automobile braking system provided in an embodiment of the present application. Figure 1 As shown, the automobile braking system includes an automobile 101 and a server 102 . A communication connection is established between the automobile 101 and the server 102 , and the automobile 101 and the server 102 interact through the established communication connection.

[0028] The server 102 is used to provide services for the car 101. The car 101 is used to obtain the driver's driving data when the driver authorizes the driver.

[0029] In one possible implementation, the car 101 is also used to send driving data to the server 102, so that the server 102 trains a speed prediction model corresponding to the driver based on the driver's driving data, and provides the car 101 with a speed prediction model corresponding to the driver, so that the driving speed of the car 101 can be controlled based on the speed prediction model corresponding to the driver during the subsequent process of the driver driving the car 101.

[0030] In another possible implementation, the server 102 is configured to provide an initial speed prediction model for the vehicle 101. The vehicle 101 receives the initial speed prediction model and trains the initial speed prediction model based on the driver's driving data to obtain a speed prediction model corresponding to the driver. Subsequently, when the driver is driving the vehicle 101, the vehicle 101 controls its driving speed based on the speed prediction model corresponding to the driver.

[0031] Figure 2 This is a flow chart of a vehicle braking method provided by an embodiment of the present application. Taking the method executed by a vehicle as an example, Figure 2 As shown, the method includes:

[0032] 201. Perform biometric identification on the driver of the car to obtain the driver's identity information.

[0033] In the embodiment of the present application, the speed prediction model for a driver is trained based on the driver's historical braking information and the deceleration corresponding to that historical braking information. The speed prediction model for the driver can simulate the driver's braking habits while driving a car, and the speed prediction models for different drivers may be different. Therefore, when a driver is driving a car, biometric identification is required to identify the driver. Based on the identified identity information, a corresponding speed prediction model can be obtained, and braking assistance can be provided to the driver based on the speed prediction model.

[0034] The biometric feature can be any type of feature, for example, biometric identification can be fingerprint identification, facial recognition, etc. The identity information is used to represent the driver, and the identity information can be represented in any form, for example, the identity information can be represented by the driver's name, or the identity information can be represented by biometric information obtained by biometric identification, such as fingerprint information or facial information.

[0035] 202. During the driving of the vehicle, in response to a braking operation, braking information is obtained, where the braking information includes a travel of a brake pedal in the vehicle.

[0036] In the embodiment of the present application, while driving a car, the driver may step on the brake pedal to trigger a braking operation on the car in the hope of reducing the car's speed. Furthermore, the deceleration of the car is positively correlated with the travel of the brake pedal. Therefore, in response to the braking operation, braking information is obtained so that the car can be subsequently decelerated according to the braking information.

[0037] The braking information can be represented in any form, for example, the braking information is represented in text form. The brake pedal of a car is also called the brake pedal, which is used to brake the car while it is driving to reduce the speed of the car. The stroke of the brake pedal can be represented in any form, for example, the stroke of the brake pedal is represented in the form of distance, or in the form of a percentage. For example, the stroke of the brake pedal is represented in the form of distance, and the stroke of the brake pedal is 1 cm, 2 cm, etc. For example, the stroke of the brake pedal is represented in the form of a percentage, and a stroke of 0% indicates that the driver has not stepped on the brake pedal, and the brake pedal is in the default state at this time. 100% indicates that the driver has fully stepped on the pedal, and the brake pedal has the maximum stroke at this time.

[0038] 203. Predicting a first deceleration of the vehicle based on braking information using a speed prediction model corresponding to the identity information, wherein the speed prediction model is trained based on historical driving data corresponding to the identity information, wherein the historical driving data includes historical braking information of the driver during historical driving of the vehicle and the deceleration corresponding to the historical braking information.

[0039] In this embodiment of the present application, the speed prediction model corresponding to the identity information is trained based on the driver's historical driving data. Therefore, the speed prediction model can simulate the driver's driving habits, i.e., the speed prediction model is a speed prediction model that matches the driver's driving habits. The speed prediction model is used to predict deceleration in conjunction with braking information. The speed prediction model corresponding to the identity information can predict the vehicle's first deceleration based on the current braking information, ensuring that the first deceleration matches the current braking information.

[0040] 204. Based on the first deceleration, control the vehicle to decelerate.

[0041] In the embodiment of the present application, since the first deceleration is equivalent to the deceleration determined according to the driver's driving habits combined with the current braking information, controlling the car's deceleration based on the first deceleration can ensure that the car's deceleration matches the driver's driving habits, thereby ensuring the driver's braking experience.

[0042] In the solution provided in the embodiment of the present application, while the driver is driving the car, biometric recognition is used to identify the driver so that a speed prediction model that matches the driver's driving habits can be obtained. In this way, if braking occurs during the driver's driving of the car, the speed prediction model is used to predict the deceleration that matches the driver's driving habits based on the braking information, and the car is controlled to decelerate according to the deceleration. In this way, personalized configuration of the braking method is achieved, and the car can be braked according to the driver's driving habits, achieving seamless connection between the braking response characteristics and the driver's inherent habits, greatly improving driving comfort and controllability, and thus improving the driver's braking experience.

[0043] above Figure 2 What is shown is only the basic process of this application. The solution provided by this application will be further elaborated below based on a specific implementation method.

[0044] Figure 3 This is a flow chart of a vehicle braking method provided by an embodiment of the present application. Taking the method executed by a vehicle as an example, Figure 3 As shown, the method includes:

[0045] 301. Perform biometric identification on the driver of the car to obtain the driver's identity information.

[0046] In a possible implementation, the car is provided with a fingerprint recognition module, and step 301 includes: collecting the driver's fingerprint information through a fingerprint recognition model, identifying the fingerprint information, and obtaining the driver's identity information.

[0047] In an embodiment of the present application, a fingerprint recognition module provided on a car is used to take a user's fingerprint to identify the user's identity information. The fingerprint recognition module can be set at any position of the car. For example, the fingerprint recognition module is set on the steering wheel or door of the car. With the driver's authorization, the fingerprint recognition module can collect the driver's fingerprint information, and then perform identity recognition according to the fingerprint information to identify the driver's identity information.

[0048] The fingerprint recognition module is any type of electronic component capable of collecting fingerprints, for example, a fingerprint recognition sensor. The fingerprint recognition module can use any fingerprint recognition technology to perform fingerprint recognition, for example, ultrasonic fingerprint recognition technology, to ensure the accuracy of identity recognition.

[0049] In another possible implementation, a facial recognition device is installed in front of the driver's seat, and step 301 includes: collecting the driver's facial information through the facial recognition device, identifying the facial information, and obtaining the driver's identity information.

[0050] In the embodiments of the present application, the facial recognition device can be any type of device, for example, a camera. With the driver's authorization, the facial recognition device can capture a facial image of the driver in the driver's seat and then perform identity recognition based on the facial image to identify the driver's identity information.

[0051] For example, cars use AI technology, dynamic recognition, infrared and other facial recognition technologies for facial recognition, which can identify whether the driver is three-dimensional, stereoscopic, and alive, and photos cannot be scanned successfully to ensure that the real driver is identified, thereby ensuring the accuracy and safety of facial recognition.

[0052] 302. During the driving process of the vehicle, in response to the stepping operation of the brake pedal, the travel of the brake pedal is detected and brake-related information is obtained; the brake-related information and the detected travel are determined as braking information.

[0053] In the embodiment of the present application, the braking information includes not only the travel of the brake pedal in the car, but also brake-related information. Braking-related information refers to information related to when the car enters a deceleration state, and the braking-related information will have an impact on the deceleration of the car. Since when braking occurs during the driving of the car, the deceleration of the car is not only related to the travel of the brake pedal, but also to the current driving speed of the car, the road conditions on which the car is located, the weather conditions at the location of the car, etc. Therefore, when the driver steps on the brake pedal while driving the car, not only the travel of the brake pedal is detected, but also the braking-related information is obtained. The braking-related information and the travel are used as braking information to enrich the content included in the braking information, so that the subsequent braking conditions of the car match the braking information, thereby ensuring the braking effect.

[0054] In a possible implementation, the braking-related information includes at least one of the vehicle's driving speed, the type of the road on which the vehicle is traveling, or weather information at the vehicle's location.

[0055] The driving speed refers to the vehicle's speed when braking. Road type can be classified in any manner. For example, road type can be classified based on road function, such as highway type, main road type, branch road type, etc. For example, road type can be classified based on road surface conditions, such as dry type, wet type, snowy type, muddy type, etc. For example, road type can be classified based on road topography and geographical features, such as mountainous type, plain type, etc. Weather information indicates the weather at the vehicle's location, such as rain, snow, icy road, etc.

[0056] In an embodiment of the present application, in addition to the brake pedal stroke being related to the car's deceleration, the car's driving speed, the road type, or the weather information at the car's location are all related to the car's deceleration. Therefore, the car's driving speed, the road type, or the weather information at the car's location are also included as part of the braking information to enrich the content of the braking information and ensure the subsequent braking effect.

[0057] Optionally, in response to a braking operation, the vehicle's driving speed is detected to obtain the vehicle's driving speed; based on the vehicle's location, the car's road can be found by querying the map information, and then the road type of the road can be obtained from the map information; based on the vehicle's location, the car interacts with the server to query the weather at that location from the server.

[0058] It should be noted that the embodiment of the present application is described by taking the example that the braking information also includes braking-related information. In another embodiment, there is no need to execute the above step 302, but other methods are adopted to obtain braking information in response to the braking operation, and the braking information includes the travel of the brake pedal in the car.

[0059] 303. The braking-related information and the travel are encoded respectively through a speed prediction model corresponding to the identity information to obtain the characteristics of the braking-related information and the travel characteristics. The speed prediction model is trained based on the historical driving data corresponding to the identity information. The historical driving data includes the historical braking information of the driver during the historical driving process of the car and the deceleration corresponding to the historical braking information.

[0060] In the embodiment of the present application, once the driver's identity information is identified, a speed prediction model corresponding to the identity information is obtained. The braking-related information and travel distance in the braking information are then encoded to obtain features of the braking-related information and travel distance features, which can then be used to predict deceleration. The features of the braking-related information and travel distance features can be represented in any manner, for example, in the form of feature vectors.

[0061] In one possible implementation, the braking-related information includes at least one of the vehicle's driving speed, the road type, or weather information at the vehicle's location. The process of obtaining the characteristics of the braking-related information includes: encoding each item in the braking-related information through a speed prediction model to obtain the characteristics of each item, and combining the characteristics of each item in the braking-related information into the characteristics of the braking-related information.

[0062] The characteristics of the braking-related information include the characteristics of each piece of information in the braking-related information. For example, the braking-related information includes the vehicle's speed, the road type, and the weather information at the vehicle's location. That is, the braking-related information includes a total of four pieces of information. Therefore, the characteristics of the braking-related information include the characteristics of each of the four pieces of information.

[0063] In the embodiment of the present application, when braking occurs during driving, the deceleration needs to be obtained using the speed prediction model corresponding to the driver. In other words, obtaining the speed prediction model corresponding to the identity information includes the following two methods.

[0064] The first method is to send a model acquisition request to the server based on the identity information. The model acquisition request carries the identity information, and the server receives the speed prediction model returned in response to the model acquisition request.

[0065] In an embodiment of the present application, the speed prediction model corresponding to the identity information may not be stored in the car, but the speed prediction model of each driver is stored in the server providing services for the car. The server stores the correspondence between the identity information and the speed prediction model. The car interacts with the server, and the server responds to the model acquisition request and queries the speed prediction model stored corresponding to the identity information based on the carried identity information. The server only provides the car with the speed prediction model corresponding to the identity information so that the deceleration can be predicted based on the speed prediction model later.

[0066] Optionally, the car performs a local query based on the identity information. If no speed prediction model corresponding to the identity information is found locally, the car sends a model acquisition request to the server based on the identity information. The model acquisition request carries the identity information, and the car receives the speed prediction model returned by the server in response to the model acquisition request.

[0067] In an embodiment of the present application, a speed prediction model may be stored in the car. For the currently recognized identity information, the car may query from the speed prediction model corresponding to the identity information stored in the car, or it may query from the speed prediction model corresponding to the identity information stored in the car. Therefore, the car prioritizes querying from the local to avoid repeated acquisition of the speed prediction model.

[0068] The second method is to query the speed prediction model corresponding to the identity information from the first corresponding relationship based on the identity information, where the first corresponding relationship includes at least one identity information and the speed prediction model corresponding to each identity information.

[0069] In the embodiment of the present application, considering that the same car may be driven by different drivers, a speed prediction model can be trained for each driver based on their historical driving data. The speed prediction models for different drivers are then stored in the same car. This allows any driver to retrieve the speed model corresponding to their identified identity from the stored speed prediction model, ensuring the accuracy and efficiency of the retrieved speed model. This also allows the same car to achieve personalized braking effects for different drivers, ensuring a consistent braking experience for each driver.

[0070] 304. Through the speed prediction model, based on the weight of the braking-related information and the weight of the stroke, the features of the braking-related information and the stroke features are weightedly fused to obtain a fused feature.

[0071] In the embodiment of the present application, different information in the braking information has different effects on vehicle braking. Some information in the braking information has a greater impact on vehicle braking, while other information has a smaller impact. Therefore, weights are set according to the degree of influence of the brake pedal travel and the braking-related information on vehicle braking. The features of the braking-related information and the travel features are then weighted and fused according to the weights to ensure that the fused features accurately reflect the vehicle's braking conditions, thereby ensuring that the deceleration can be accurately predicted.

[0072] The weight is used to indicate the degree of influence on vehicle braking. The greater the weight, the greater the influence on vehicle braking; the smaller the weight, the smaller the influence on vehicle braking.

[0073] In one possible implementation, the braking-related information includes at least one of the vehicle's driving speed, the road type, or weather information at the vehicle's location, and the characteristics of the braking-related information include the characteristics of each item of information in the braking-related information. Then, according to the weight of each item of information in the braking-related information and the weight of the trip, the characteristics of each item of information in the braking-related information and the trip characteristics are weightedly fused to obtain the fused characteristics.

[0074] In the embodiment of the present application, considering that different information contained in the braking-related information may have different effects on vehicle braking, weights are set for different types of information contained in the braking-related information to ensure the accuracy of the obtained fusion features.

[0075] 305. Decode the fused features through the speed prediction model to obtain the first deceleration.

[0076] In the embodiment of the present application, since the fusion feature can accurately reflect the relationship between the braking information and the vehicle braking, and can accurately describe the vehicle braking situation, the fusion feature can be decoded through the speed prediction model to obtain the first deceleration, and ensure that the first deceleration matches the braking information, and is also in line with the driver's driving habits, thereby ensuring the subsequent braking effect and the driver's braking experience.

[0077] It should be noted that the embodiment of the present application is described by taking the example of predicting the first deceleration by encoding and decoding the speed prediction model. In another embodiment, there is no need to perform the above steps 302-305, but other methods are adopted to predict the first deceleration of the car based on the braking information through the speed prediction model corresponding to the identity information.

[0078] 306. Based on the first deceleration, query a control parameter corresponding to the first deceleration from a second corresponding relationship, where the second corresponding relationship indicates control parameters corresponding to different decelerations.

[0079] In an embodiment of the present application, when the first deceleration is obtained through the speed prediction model, it is necessary to control the vehicle to decelerate according to the first deceleration, and the actual braking condition of the vehicle is related to the control parameters of the vehicle. The second corresponding relationship is equivalent to the mapping relationship between the deceleration and the control parameters of the vehicle, which can reflect what the control parameters of the vehicle are when the vehicle decelerates according to different decelerations. Therefore, based on the first deceleration, the control parameters corresponding to the first deceleration are queried from the second corresponding relationship, so as to control the deceleration of the vehicle according to the queried control parameters, so that the actual deceleration of the vehicle is consistent with the first deceleration, avoiding the deviation between the actual deceleration of the vehicle and the first deceleration, and ensuring the braking effect of the vehicle.

[0080] Among them, the control parameters are parameters that affect the deceleration of the car. For example, the control parameters are the car's braking force, pressure buildup time, brake line pressure, etc. Among them, the braking force refers to the friction force applied by the braking system on the wheel, which is used to hinder the rotation of the wheel and slow down or stop the vehicle; the pressure buildup time refers to the time from the driver stepping on the brake pedal to the time the braking system reaches the target pressure; the brake line pressure refers to the pressure of the brake fluid in the brake system pipeline.

[0081] 307. Control the vehicle to decelerate based on the control parameter corresponding to the first deceleration.

[0082] In an embodiment of the present application, when the control parameters corresponding to the first deceleration are queried, the car is controlled to decelerate according to the queried control parameters, so that the car meets the control parameters corresponding to the first deceleration, and thus the actual deceleration of the car is equal to the first deceleration.

[0083] For example, the control parameters corresponding to the first deceleration indicate the braking force and the pressure building time. Then, according to the braking force and the pressure building time corresponding to the first deceleration, the car is controlled to decelerate so that the actual deceleration of the car is equal to the first deceleration.

[0084] It should be noted that the embodiment of the present application is described by taking the example of controlling the deceleration of the vehicle through the control parameters corresponding to the first deceleration. In another embodiment, there is no need to execute the above steps 306-307, but other methods are adopted to control the deceleration of the vehicle based on the first deceleration.

[0085] In the solution provided in the embodiment of the present application, while the driver is driving the car, biometric recognition is used to identify the driver so that a speed prediction model that matches the driver's driving habits can be obtained. In this way, if braking occurs during the driver's driving of the car, the speed prediction model is used to predict the deceleration that matches the driver's driving habits based on the braking information, and the car is controlled to decelerate according to the deceleration. In this way, personalized configuration of the braking method is achieved, and the car can be braked according to the driver's driving habits, achieving seamless connection between the braking response characteristics and the driver's inherent habits, greatly improving driving comfort and controllability, and thus improving the driver's braking experience.

[0086] It should be noted that the above Figure 3 The embodiment shown is explained by taking the first deceleration output by the speed prediction model to control the deceleration of the vehicle as an example. In another embodiment, after the above step 305, the deceleration of the vehicle can also be controlled in combination with the second deceleration corresponding to the brake pedal stroke in the vehicle. That is, the process of controlling the deceleration of the vehicle includes the following steps 1 to 3.

[0087] Step 1: Obtain a second deceleration, where the second deceleration is the deceleration of the vehicle caused by the brake pedal being depressed.

[0088] In the embodiments of the present application, when the brake pedal in a vehicle is stepped on, it activates the brake actuator to brake the vehicle, causing it to decelerate. That is, when the brake pedal is stepped on for a certain distance, in addition to the deceleration predicted by the speed prediction model, changes in the brake pedal's own travel also produce a deceleration. Furthermore, this deceleration is positively correlated with the brake pedal's travel; the greater the brake pedal's travel, the greater the deceleration. The second deceleration is the deceleration generated by the vehicle itself when the brake pedal is stepped on for that distance, in addition to the deceleration predicted by the speed prediction model.

[0089] In one possible implementation, the process of obtaining the second deceleration includes querying a first mapping relationship based on the brake pedal travel to obtain the second deceleration, wherein the first mapping relationship includes multiple preset brake pedal travels and a preset deceleration corresponding to each preset travel.

[0090] Step 2: Perform weighted fusion on the first deceleration and the second deceleration to obtain a first fused deceleration.

[0091] In an embodiment of the present application, taking into account that the first deceleration predicted by the speed prediction model may be inconsistent with the second deceleration corresponding to the travel of the brake pedal, the first deceleration and the second deceleration are weightedly fused to obtain a first fused deceleration to ensure that the first fused deceleration matches the braking information and ensure the subsequent braking effect.

[0092] In one possible implementation, step 307 includes: performing weighted fusion on the first deceleration and the second deceleration based on the first weight and the second weight to obtain a first fused deceleration.

[0093] The first weight is the weight corresponding to the deceleration predicted by the speed prediction model, and the second weight is the weight corresponding to the deceleration caused by the brake pedal being depressed. Optionally, the sum of the first weight and the second weight is 1.

[0094] Optionally, the first weight and the second weight may change, and the process of obtaining the first weight and the second weight includes: querying the first weight and the second weight from a second mapping relationship based on the braking information. The second mapping relationship includes multiple preset braking information, the first weight and the second weight corresponding to each preset braking information.

[0095] In the embodiment of the present application, each preset braking information represents a braking scenario. Different preset braking information represents different braking scenarios. For each braking scenario, a first weight and a second weight are set. This allows the deceleration predicted by the speed prediction model to be combined with the deceleration resulting from the brake pedal's travel, according to the set first and second weights, for each braking scenario. Therefore, by querying the second mapping relationship based on the braking information, the first and second weights corresponding to the braking information can be obtained, ensuring that the obtained first and second weights match the braking scenario and the subsequent braking effect is guaranteed.

[0096] Optionally, the process of determining the first weight and the second weight includes: determining the similarity between the braking information and each preset braking information in the second mapping relationship, and determining the first weight and the second weight corresponding to the preset braking information with the greatest similarity as the first weight and the second weight corresponding to the braking information.

[0097] In the embodiment of the present application, the similarity can be determined in any manner. For example, the braking information includes multiple pieces of information, and the similarity is determined by the number of identical items between the braking information and the preset braking information, or a pre-similarity method is used to determine the similarity.

[0098] Step 3: Control the vehicle to decelerate according to the first fusion deceleration.

[0099] In the embodiment of the present application, the first fusion deceleration not only conforms to the driver's driving habits, but also matches the stroke of the brake pedal being depressed. Decelerating according to the first fusion deceleration can improve the braking effect of the car and enhance the driver's braking experience.

[0100] On the basis of the embodiment shown above, the driver can also manually adjust the deceleration, that is, the method also includes the following three methods.

[0101] The first method is: displaying a first deceleration on a screen of the car; adjusting the first deceleration in response to an adjustment operation on the first deceleration on the screen; and controlling the car to decelerate based on the adjusted first deceleration.

[0102] In an embodiment of the present application, the first deceleration is the deceleration predicted by the speed prediction model based on the braking information. The first deceleration is displayed on the screen. If the driver is not satisfied with the first deceleration, the driver can adjust the deceleration according to the deceleration displayed on the screen to adjust the deceleration according to the braking information. At this time, the car deceleration is no longer controlled based on the first deceleration, but is controlled according to the adjusted first deceleration, thereby realizing a solution for the driver to manually control the deceleration, which can ensure that the car deceleration meets the user's expectations and ensures the driver's experience.

[0103] In one possible implementation, after the driver manually adjusts the first deceleration, the braking information and the adjusted first deceleration can also be used as training data for the subsequent optimization of the speed prediction model, that is, the training process includes: training the speed prediction model based on the adjusted first deceleration and the first deceleration.

[0104] In an embodiment of the present application, the adjusted first deceleration is based on the deceleration obtained after manual adjustment by the driver, which can reflect that the driver is dissatisfied with the first deceleration. Therefore, the speed prediction model is further optimized based on the first deceleration and the adjusted first deceleration to ensure that the deceleration output by the subsequent speed prediction model is more in line with the user's wishes, thereby improving the accuracy of the speed prediction model.

[0105] Optionally, the process of training the speed prediction model includes: determining a loss value based on the adjusted first deceleration and the first deceleration, and training the speed prediction model based on the loss value, wherein the loss value is used to represent the difference between the adjusted first deceleration and the first deceleration.

[0106] The second method is: displaying a second deceleration on the car's screen; adjusting the second deceleration in response to a deceleration adjustment operation for the travel on the screen; performing weighted fusion on the first deceleration and the adjusted second deceleration to obtain a second fused deceleration; and controlling the car to decelerate according to the second fused deceleration.

[0107] In an embodiment of the present application, the second deceleration is the deceleration generated when the brake pedal is stepped on the above-mentioned stroke. By displaying the second deceleration on the screen, it is equivalent to displaying the deceleration generated when the brake pedal is stepped on the above-mentioned stroke on the screen. At this time, the driver can adjust the deceleration according to the deceleration displayed on the screen to adjust the deceleration generated when the brake pedal is stepped on the above-mentioned stroke, thereby ensuring that the final fused deceleration will also be updated, and the car is controlled to decelerate according to the updated second fused deceleration, realizing a solution for the driver to manually control the deceleration, which can ensure that the car deceleration meets the user's expectations and ensure the driver's experience.

[0108] In one possible implementation, after the deceleration is adjusted according to the first method, the speed prediction model can also be trained, that is, the training process includes: training the speed prediction model based on the first fused deceleration and the second fused deceleration.

[0109] In an embodiment of the present application, the second fused deceleration is based on the deceleration obtained after manual adjustment by the driver, which can reflect that the driver is dissatisfied with the first fused deceleration. Therefore, the speed prediction model is trained based on the first fused deceleration and the second fused deceleration to ensure that the deceleration output by the subsequent speed prediction model is more in line with the user's wishes, thereby improving the accuracy of the speed prediction model.

[0110] Optionally, the speed prediction model is trained based on the difference between the first fused deceleration and the second fused deceleration.

[0111] It should be noted that the aforementioned adjustment of the second deceleration can be effective once or multiple times. If the adjustment of the second deceleration is effective once, and the brake pedal stroke during the next braking is the same as the aforementioned stroke, the deceleration corresponding to the stroke will continue to be the second deceleration. If the adjustment of the second deceleration is effective multiple times, and the brake pedal stroke during the next braking is the same as the aforementioned stroke, the deceleration corresponding to the stroke will be the adjusted second deceleration. If the adjustment of the second deceleration is effective multiple times, the first mapping relationship will be updated.

[0112] In a possible implementation, after the second deceleration is adjusted, the first mapping relationship is updated based on the adjusted second deceleration and the corresponding stroke.

[0113] In an embodiment of the present application, the first mapping relationship includes multiple preset strokes of the brake pedal and a preset deceleration corresponding to each preset stroke. The second deceleration is a preset deceleration included in the first mapping relationship. According to the above method, the second deceleration is adjusted, which is equivalent to adjusting the deceleration corresponding to the stroke. The first mapping relationship is updated to increase or decrease the deceleration at the same stroke, thereby realizing personalized setting of the first mapping relationship and improving user experience.

[0114] Optionally, the process of updating the first mapping relationship includes: determining a change rate of the deceleration based on the second deceleration and the adjusted second deceleration, and adjusting each preset deceleration in the first mapping relationship based on the change rate.

[0115] In an embodiment of the present application, after the driver manually adjusts the deceleration corresponding to any stroke, it is equivalent to triggering the adjustment operation of the preset deceleration corresponding to each preset stroke in the first mapping relationship. Then, each preset deceleration is adjusted according to the rate of change to ensure the convenience of deceleration adjustment.

[0116] Optionally, the process of determining the change rate includes: determining a difference between the adjusted second deceleration and the second deceleration, and determining a ratio of the difference to the second deceleration as the change rate.

[0117] It should be noted that the above is an example of adjusting the first mapping relationship when braking occurs any time during driving. In another embodiment, when the car is not driving, the driver can operate in the car to display the first mapping relationship or the preset deceleration corresponding to any preset stroke in the first mapping relationship on the car screen. After that, the preset deceleration is adjusted in the above manner, thereby adjusting the first mapping relationship.

[0118] The third method: displaying the first fusion deceleration on the screen; adjusting the first fusion deceleration in response to the deceleration adjustment operation for the braking information on the screen to obtain a third fusion deceleration; and controlling the vehicle to decelerate according to the third fusion deceleration.

[0119] In an embodiment of the present application, the first fused deceleration is equivalent to the deceleration corresponding to the braking information, and the first fused deceleration is displayed on the screen. If the driver is not satisfied with the first fused deceleration, the driver can adjust the deceleration according to the deceleration displayed on the screen to adjust the deceleration according to the braking information, and control the car to decelerate according to the updated third fused deceleration, thereby realizing a solution for the driver to manually control the deceleration, which can ensure that the car's deceleration meets the user's expectations and ensures the driver's experience.

[0120] In one possible implementation, after adjusting the deceleration according to the second method, the speed prediction model can be trained. That is, the training process includes: training the speed prediction model based on the first fused deceleration and the third fused deceleration;

[0121] In an embodiment of the present application, the third fused deceleration is based on the deceleration obtained after manual adjustment by the driver, which can reflect that the driver is dissatisfied with the first fused deceleration. Therefore, the speed prediction model is trained based on the first fused deceleration and the third fused deceleration to ensure that the deceleration output by the subsequent speed prediction model is more in line with the user's wishes, so as to improve the accuracy of the speed prediction model.

[0122] On the basis of the embodiment shown above, before using the speed preset model to predict deceleration based on braking information, the speed prediction model needs to be trained. The process of training the speed prediction model includes: with the driver's authorization, obtaining the driver's driving data while the driver is driving the car; predicting the deceleration based on the braking information in the driving data through the speed prediction model; training the speed prediction model based on the predicted deceleration and the deceleration in the driving data; and determining the trained speed prediction model as the speed prediction model corresponding to the identity information.

[0123] In an embodiment of the present application, with the driver's authorization, the car can obtain the driver's driving data during the process of driving the car, which includes braking information and corresponding deceleration each time braking occurs. The speed prediction model is then trained based on the driving data, and the trained speed prediction model is determined as the speed prediction model corresponding to the identity information to ensure that the trained speed prediction model matches the driver's driving habits, thereby realizing a solution of training a dedicated speed prediction model for the driver and ensuring the subsequent driving experience and braking experience of the driver.

[0124] In a possible implementation, the speed prediction model before training based on the driver's driving data is an arbitrary network model, or is a model that has been preliminarily trained based on the driving data of multiple drivers.

[0125] In an embodiment of the present application, the speed prediction model before training is a model that has been preliminarily trained based on the driving data of multiple drivers. This allows the speed model before training to have a certain prediction ability. Subsequently, training is performed based on the driving data of a specific driver to obtain a speed prediction model exclusive to the specific driver. This can avoid the situation where the speed prediction model training is inaccurate due to insufficient driving data of a single driver, and can improve the accuracy of the speed prediction model.

[0126] In an embodiment of the present application, before training the speed prediction model, the driver's historical driving data needs to be collected so that the speed prediction model can be subsequently trained using this historical driving data. The driving data collection process involves installing a high-precision pressure sensor and position sensor on the vehicle's brake pedal. These sensors can then collect the brake pedal travel. Furthermore, a speed detector inside the vehicle can detect the deceleration rate when the driver steps on the brake pedal, and the detected deceleration rate is displayed on the vehicle's display screen to provide deceleration feedback to the driver. This allows the driver to trigger a deceleration adjustment operation using buttons on the vehicle, voice input, or a button light on the display screen, thereby adjusting the vehicle's deceleration rate, such as increasing or decreasing it, while maintaining the same brake pedal travel. If the driver adjusts the displayed deceleration rate, the collected pedal travel and adjusted deceleration rate are used as historical driving data. If the driver does not adjust the displayed deceleration rate, the collected pedal travel and displayed deceleration rate are used as historical driving data. Furthermore, for any stroke of the brake pedal, the deceleration corresponding to that stroke has an upper and lower limit. These upper and lower limits enable the vehicle to brake within a safe range. When the driver triggers a deceleration adjustment operation, the deceleration can only be adjusted between the lower and upper limits. If the deceleration reaches the upper limit corresponding to the current stroke, the deceleration cannot be increased any further. If the deceleration reaches the lower limit corresponding to the current stroke, the deceleration cannot be decreased any further. Furthermore, in the process of collecting historical driving data, the vehicle's braking-related information is also collected. For example, when the driver steps on the brake pedal, the vehicle not only collects the brake pedal stroke and the vehicle's deceleration, but also collects the vehicle's speed, the type of road props on the vehicle's current road, the weather at the vehicle's location, and other information. The collected brake pedal stroke, deceleration, speed, the type of road props on the vehicle's current road, the weather at the vehicle's location, and other information are used as historical driving data.

[0127] In an embodiment of the present application, before using historical driving data to train a speed prediction model, the historical driving data needs to be preprocessed to filter out missing values or outliers in the historical driving data, and the historical driving data needs to be standardized so that the same type of data is within the same range to ensure the accuracy of the processed historical driving data. The processed historical driving data is divided into a training set and a test set. The speed prediction model is trained using the historical driving data in the training set. After the speed prediction model is trained, the speed prediction model is tested using the historical driving data in the test set to verify the accuracy of the speed prediction model.

[0128] In an embodiment of the present application, the speed prediction model can be any type of neural network model, for example, the speed prediction model is a random forest model or a support vector regression (SVR) model using SVM (Support Vector Machines).

[0129] In the case of a random forest model, the speed prediction model is implemented by importing a random forest regressor from the ensemble module of sklearn (a machine learning library) in Python (a high-level programming language) to facilitate subsequent model construction. The train_test_split function (a dataset splitting function) is introduced to split the given dataset into training and test sets according to a specified ratio. The mean_squared_error function is introduced to evaluate the performance of the random forest regression model during training. Assuming that historical driving data has been collected in a DataFrame (a data structure), including deceleration-related features and deceleration, deceleration-related features include pedal travel, speed, and road type, the historical driving data is randomly divided into training and test sets. For example, the training set accounts for 20% of the total historical driving data, and the test set accounts for 80%. A random forest model is then established, specifying the number of decision trees and the random seed. The input of the random forest model is the braking information from the training set, and the output is the predicted deceleration. Afterwards, the mean square error loss function is used to calculate the loss value based on the predicted deceleration and the actual deceleration corresponding to the braking information in the training set, and the random forest model is trained based on the loss value.

[0130] After the random forest model is trained, the trained random forest model can be used for calibration and adjustment to obtain the second corresponding relationship in the above embodiment. During the calibration and adjustment process, the trained random forest model outputs the deceleration corresponding to the braking information based on the given braking information. In the process of controlling the deceleration of the vehicle based on the deceleration, the actual deceleration of the vehicle is detected, and the difference between the actual deceleration and the deceleration corresponding to the braking information is determined. Then, based on this difference, the vehicle system parameters are adjusted to reduce the error; through repeated iterations, the calibration parameters (i.e., control parameters) are gradually adjusted to obtain the second corresponding relationship, so that the second corresponding relationship indicates the control parameters corresponding to different decelerations, so that the deceleration of the vehicle can be controlled by combining the deceleration output by the random forest model and the second corresponding relationship, so that the actual deceleration of the vehicle is consistent with the deceleration output by the random forest model.

[0131] In the case of a support vector regression (SVR) model for speed prediction, the train_test_split function is imported from Python to split the given dataset into training and test sets according to a specified ratio. Since the SVR model is sensitive to data scale, a function is introduced to normalize the historical driving data. Normalization ensures that all data are on the same scale, preventing bias in the speed prediction model caused by large numerical ranges in some data. A SVR model is established and trained using the scikit-learn library. The SVR model parameters include the kernel function, the regularization parameter C, and the kernel parameter Gamma. The kernel function maps the input data from the original space to a higher-dimensional feature space, making the data linearly separable in the new space. The regularization parameter C controls the model's tolerance for error, balancing model complexity and training error. The kernel parameter Gamma controls the kernel width, affecting the model's locality and generalization capabilities. When creating the SVR model, a radial basis function (RBF) is specified as the kernel function, the regularization parameter C is set to 1, and the kernel parameter Gamma is set to scale. Scale is used to indicate that the value of the kernel function parameter Gamma is automatically calculated based on the variance and number of data features, which enables the support vector regression model to perform relatively stably on data of varying scales. The input of the support vector regression model is the braking information in the training set, and the output of the support vector regression model is the predicted deceleration. A loss function is then used to calculate the loss value based on the predicted deceleration and the actual deceleration corresponding to the braking information in the training set. The support vector regression model is then trained based on this loss value. Furthermore, to improve the prediction accuracy of the support vector regression model, its hyperparameters can be optimized, for example, the kernel function and the kernel function parameter Gamma.

[0132] After the support vector regression model is trained, the trained support vector regression model can be used for calibration and adjustment to obtain the second corresponding relationship in the above embodiment. During the calibration and adjustment process, the trained support vector regression model outputs the deceleration corresponding to the braking information based on the given braking information. In the process of controlling the deceleration of the car based on the deceleration, the actual deceleration of the car is detected, and the difference between the actual deceleration and the deceleration corresponding to the braking information is determined. Then, based on this difference, the car system parameters are adjusted to reduce the error; through repeated iterations, the calibration parameters (i.e., control parameters) are gradually adjusted to obtain the second corresponding relationship, so that the second corresponding relationship indicates the control parameters corresponding to different decelerations, so that the deceleration of the car can be controlled by combining the deceleration output by the support vector regression model and the second corresponding relationship, so that the actual deceleration of the car is consistent with the deceleration output by the support vector regression model.

[0133] For the aforementioned random forest model or support vector regression model, cross-validation is used to select the optimal hyperparameters for more accurate calibration. This ensures that the speed prediction model performs reliably in different driving scenarios (urban, highway, and icy roads), avoids the risk of overfitting due to a single data partition, improves the generalization capability of the speed prediction model, analyzes which vehicle parameters have a significant impact on deceleration, and optimizes these system parameters. Furthermore, the speed prediction model can be embedded in the actual system for real-time adjustment.

[0134] After deploying the speed prediction model to the vehicle system, its performance needs to be continuously monitored and retrained using new driving data to account for potential environmental changes and system aging. This approach allows the speed prediction model to be used to calibrate pedal travel and deceleration, achieving precise vehicle control. Based on the trained speed prediction model, the brake pedal's response characteristics can be adjusted in real time to provide a personalized braking experience.

[0135] Building on the above-described embodiments, in this embodiment, a car can use facial recognition cameras and eye tracking technology to monitor the driver's fatigue, and monitor the driver's attention through head posture detection and cameras. When driver fatigue or inattention is detected, the car can enhance braking feedback and alert the driver through sound, vibration, or visual signals. The car allows the driver to set and adjust the brake pedal response characteristics by providing a user interface via an on-board touch screen or mobile phone application. The car stores the driver's personalized settings in the vehicle's control unit or cloud database so that they are automatically loaded next time for user convenience.

[0136] In this embodiment, to implement this solution, hardware is installed in the vehicle, including fingerprint recognition sensors and facial recognition cameras, pressure and position sensors on the brake pedal, and a high-performance electronic control unit (ECU). An identity recognition module is developed to implement fingerprint and facial recognition capabilities; a behavior learning and adaptation module is developed to utilize machine learning algorithms for driver behavior analysis and model generation; a status monitoring and safety reminder module is developed to implement fatigue monitoring and attention detection; and a personalized settings module is developed, with a user interface designed and parameter storage and loading capabilities. The hardware and software systems are integrated into the vehicle, ensuring communication and coordination between modules. Data transmission and control signal exchange are implemented via the vehicle network. Hardware-in-the-loop testing is conducted in a laboratory environment to verify the system's functionality and performance. Testing is conducted under actual road conditions to assess the system's stability and reliability. Optimization and adjustments are made based on the test results to improve system performance. System design documentation and technical specifications are completed, production line layout and manufacturing process design are carried out, and pilot production is conducted to verify the production process and quality control. Through the above technical solutions, the intelligent brake pedal system can provide a personalized braking experience, improve driving safety and comfort, and meet the needs of modern intelligent and personalized vehicles.

[0137] On the basis of the above-mentioned embodiment, the embodiment of the present application further provides a flow chart of a vehicle braking method, as shown in FIG. Figure 4 As shown, the method includes: the vehicle uses fingerprint / face sensors for identity recognition; a high-level intelligent algorithm model performs complex calculations and predictions, transmitting data or instructions to the IPB (Integrated Braking System) to assist the IPB in more intelligent braking control; and interacting with the HMI (Human Machine Interface) to provide the driver with braking-related predictions or recommendations, enhancing the driving experience and safety. The IPB receives information from the fingerprint / face sensor and the high-level intelligent algorithm model, performs comprehensive processing, and then sends instructions to the brake to control the braking process. It also interacts with the HMI to provide feedback on information such as the braking system status. The HMI (Human Machine Interface) enables human-system interaction, receiving information from the high-level intelligent algorithm model to display braking status and prompts to the driver; it also receives operational instructions input by the driver through the interface. The brake receives IPB instructions and executes the braking action, thus realizing the vehicle's braking function.

[0138] The solutions provided in the embodiments of this application utilize biometric technology and artificial intelligence to enable a brake pedal system that accurately identifies the driver and intelligently analyzes their driving behavior, automatically adjusting the brake pedal's response and force based on the driver's habits and preferences. This personalized brake pedal control system not only improves driving comfort and safety, but also enhances the driver's sense of control and trust in the vehicle, providing a more intelligent and personalized driving experience.

[0139] Figure 5 This is a schematic structural diagram of an automobile braking device provided in an embodiment of the present application. Figure 5 As shown, the method includes:

[0140] Identification module 501, used to perform biometric identification on the driver of the car to obtain the driver's identity information;

[0141] An acquisition module 502 is configured to acquire braking information in response to a braking operation during driving of the vehicle, the braking information including a travel of a brake pedal in the vehicle;

[0142] A prediction module 503 is configured to predict a first deceleration of the vehicle based on the braking information using a speed prediction model corresponding to the identity information, where the speed prediction model is trained based on historical driving data corresponding to the identity information, where the historical driving data includes historical braking information of the driver during historical driving of the vehicle and decelerations corresponding to the historical braking information;

[0143] The control module 504 is configured to control the vehicle to decelerate based on the first deceleration.

[0144] In one possible implementation, the braking information also includes braking-related information. The acquisition module 502 is used to detect the travel of the brake pedal in response to the stepping operation of the brake pedal and obtain the braking-related information; the braking-related information and the detected travel are determined as braking information; wherein the braking-related information includes at least one of the vehicle's driving speed, the road type of the road, or the weather information of the vehicle's location.

[0145] In another possible implementation, the apparatus further includes:

[0146] A sending module is used to send a model acquisition request to the server based on the identity information, the model acquisition request carries the identity information, and receive the speed prediction model returned by the server in response to the model acquisition request; or

[0147] The query module is used to query the speed prediction model corresponding to the identity information from the first corresponding relationship based on the identity information. The first corresponding relationship includes at least one identity information and the speed prediction model corresponding to each identity information.

[0148] In another possible implementation, the braking information also includes braking-related information; the prediction module 503 is used to encode the braking-related information and the stroke respectively through the speed prediction model to obtain the characteristics of the braking-related information and the stroke characteristics; through the speed prediction model, based on the weight of the braking-related information and the weight of the stroke, the characteristics of the braking-related information and the stroke characteristics are weightedly fused to obtain a fused feature; through the speed prediction model, the fused feature is decoded to obtain the first deceleration.

[0149] In another possible implementation, the acquisition module 502 is further configured to acquire a second deceleration, where the second deceleration is a deceleration of the vehicle caused by a brake pedal being depressed.

[0150] The control module 504 is configured to perform weighted fusion of the first deceleration and the second deceleration to obtain a first fused deceleration; and control the vehicle to decelerate according to the first fused deceleration.

[0151] In another possible implementation, the control module 504 is also used to display a second deceleration on the screen of the car; adjust the second deceleration in response to a deceleration adjustment operation for the travel on the screen; perform weighted fusion of the first deceleration and the adjusted second deceleration to obtain a second fused deceleration; control the car to decelerate according to the second fused deceleration; or, display the first fused deceleration on the screen; adjust the first fused deceleration in response to a deceleration adjustment operation for braking information on the screen to obtain a third fused deceleration; and control the car to decelerate according to the third fused deceleration.

[0152] In another possible implementation, the apparatus further includes:

[0153] A training module is used to train the speed prediction model based on the first fusion deceleration and the second fusion deceleration; or to train the speed prediction model based on the first fusion deceleration and the third fusion deceleration.

[0154] In another possible implementation, the control module 504 is used to query the control parameters corresponding to the first deceleration from the second correspondence based on the first deceleration, where the second correspondence indicates the control parameters corresponding to different decelerations; and control the vehicle to decelerate based on the control parameters corresponding to the first deceleration.

[0155] In another possible implementation, the apparatus further includes:

[0156] a display module, configured to display the first deceleration on a screen of the vehicle;

[0157] an adjusting module, configured to adjust the first deceleration in response to an adjustment operation on the first deceleration on the screen;

[0158] The control module 504 is further configured to control the vehicle to decelerate based on the adjusted first deceleration.

[0159] In another possible implementation, the apparatus includes:

[0160] The training module is used to train the speed prediction model based on the adjusted first deceleration and the first deceleration.

[0161] In another possible implementation, the apparatus further includes:

[0162] The acquisition module 502 is further configured to acquire the driver's driving data while the driver is driving the car, with the driver's authorization;

[0163] The prediction module 503 is further configured to predict deceleration based on braking information in the driving data using a speed prediction model;

[0164] The training module is used to train the speed prediction model based on the predicted deceleration and the deceleration in the driving data; and the trained speed prediction model is determined as the speed prediction model corresponding to the identity information.

[0165] It should be noted that the automobile braking device provided in the above embodiment is merely illustrated by the division of the aforementioned functional modules. In actual applications, the aforementioned functions can be distributed among different functional modules as needed. This means that a computer device or the internal structure of an automobile can be divided into different functional modules to perform all or part of the functions described above. Furthermore, the automobile braking device provided in the above embodiment and the automobile braking method embodiment share the same concept. The specific implementation process is detailed in the method embodiment and will not be further described here.

[0166] An embodiment of the present application also provides a car, which includes a processor and a memory, wherein the memory stores at least one computer program, and the at least one computer program is loaded and executed by the processor to implement the operations performed by the car braking method of the above embodiment.

[0167] An embodiment of the present application also provides a computer device, which includes a processor and a memory, wherein the memory stores at least one computer program, and the at least one computer program is loaded and executed by the processor to implement the operations performed by the automobile braking method of the above embodiment.

[0168] Those skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware, or may be accomplished by a program instructing the relevant hardware, and the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a disk, or an optical disk, etc.

[0169] The above are only optional embodiments of the embodiments of the present application and are not intended to limit the embodiments of the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the embodiments of the present application should be included in the scope of protection of this application.

Claims

1. A vehicle braking method, characterized in that: The method comprises: Performing biometric identification on the driver of the vehicle to obtain the driver's identity information; During the driving of the vehicle, in response to a braking operation, obtaining braking information, the braking information including a travel of a brake pedal in the vehicle; predicting a first deceleration of the vehicle based on the braking information using a speed prediction model corresponding to the identity information, the speed prediction model being trained based on historical driving data corresponding to the identity information, the historical driving data including historical braking information of the driver during historical driving of the vehicle and deceleration corresponding to the historical braking information; Based on the first deceleration, the vehicle is controlled to decelerate.

2. The method according to claim 1, characterized in that The braking information further includes braking-related information. The obtaining of the braking information in response to the braking operation includes: In response to a stepping operation on the brake pedal, detecting a stroke of the brake pedal and acquiring the brake-related information; determining the braking-related information and the detected travel as the braking information; The braking-related information includes at least one of the driving speed of the vehicle, the type of the road on which the vehicle is located, or weather information at the location of the vehicle.

3. The method according to claim 1, characterized in that Before predicting the first deceleration of the vehicle based on the braking information using the speed prediction model corresponding to the identity information, the method further includes: Based on the identity information, sending a model acquisition request to a server, the model acquisition request carrying the identity information, and receiving the speed prediction model returned by the server in response to the model acquisition request; or Based on the identity information, the speed prediction model corresponding to the identity information is searched from a first corresponding relationship, where the first corresponding relationship includes at least one identity information and a speed prediction model corresponding to each identity information.

4. The method according to claim 1, wherein The braking information also includes braking association information; and the predicting of the first deceleration of the vehicle based on the braking information using a speed prediction model corresponding to the identity information includes: Encoding the braking-related information and the travel respectively through the speed prediction model to obtain features of the braking-related information and features of the travel; By using the speed prediction model, based on the weight of the braking-related information and the weight of the travel, the features of the braking-related information and the travel features are weightedly fused to obtain a fused feature; The fusion feature is decoded using the speed prediction model to obtain the first deceleration.

5. The method according to claim 1, characterized in that Before controlling the vehicle to decelerate based on the first deceleration, the method further includes: obtaining a second deceleration, where the second deceleration is a deceleration of the vehicle caused by the brake pedal being depressed for the distance; The controlling the vehicle to decelerate based on the first deceleration includes: Performing weighted fusion on the first deceleration and the second deceleration to obtain a first fused deceleration; The vehicle is controlled to decelerate according to the first fusion deceleration.

6. The method according to claim 1, characterized in that The controlling the vehicle to decelerate based on the first deceleration includes: Based on the first deceleration, querying a control parameter corresponding to the first deceleration from a second correspondence relationship, where the second correspondence relationship indicates control parameters corresponding to different decelerations; The vehicle is controlled to decelerate based on a control parameter corresponding to the first deceleration.

7. The method according to claim 1, characterized in that Before predicting the first deceleration of the vehicle based on the braking information using the speed prediction model corresponding to the identity information, the method further includes: With the driver's authorization, obtaining the driver's driving data while the driver is driving the car; predicting deceleration based on braking information in the driving data using the speed prediction model; training the speed prediction model based on the predicted deceleration and the deceleration in the driving data; The trained speed prediction model is determined as the speed prediction model corresponding to the identity information.

8. An automobile braking device, characterized in that: The device comprises: An identification module, configured to perform biometric identification on the driver of the vehicle to obtain the driver's identity information; an acquisition module, configured to acquire braking information in response to a braking operation during the driving of the vehicle, wherein the braking information includes a travel of a brake pedal in the vehicle; a prediction module, configured to predict a first deceleration of the vehicle based on the braking information using a speed prediction model corresponding to the identity information, the speed prediction model being trained based on historical driving data corresponding to the identity information, the historical driving data including historical braking information of the driver during historical driving of the vehicle and decelerations corresponding to the historical braking information; A control module is configured to control the vehicle to decelerate based on the first deceleration.

9. An automobile, characterized in that: The vehicle includes a processor and a memory, wherein the memory stores at least one computer program, and the at least one computer program is loaded and executed by the processor to implement the operations performed by the vehicle braking method according to any one of claims 1 to 7.

10. A computer device, characterized in that: The computer device includes a processor and a memory, wherein the memory stores at least one computer program, and the at least one computer program is loaded and executed by the processor to implement the operations performed by the vehicle braking method according to any one of claims 1 to 7.

Images