Control method and device of vehicle, vehicle and storage medium

Abstract

The application discloses a control method and device of a vehicle, the vehicle and a storage medium. The method comprises the following steps: acquiring current driving data of the vehicle; determining recommended following parameters by using an individualized prediction model based on the current driving data; wherein the individualized prediction model is obtained by fine-tuning a trained prediction model based on historical driving data of a target driver; and controlling a rear lamp of the vehicle to work based on the recommended following parameters. The application can actively help the driver of the rear vehicle to adjust the following strategy in the early stage of risk, prevent the following risk from happening, and improve the following safety of the rear vehicle.

Description

Technical Field

[0001] This application relates to the field of vehicle control technology, and more specifically, to a vehicle control method, device, vehicle, and storage medium. Background Technology

[0002] With the increasing popularity and development of vehicles, in order to improve vehicle driving safety, the advanced driver-assistance systems (ADAS) inside the vehicle will usually estimate the distance between the vehicle (own vehicle) and the vehicle in front. When the distance between the two is less than the safe distance, it indicates that the vehicle needs to brake urgently, and then controls the taillights to display a warning symbol to remind the driver of the following vehicle to pay attention to driving safety.

[0003] However, the triggering mechanism of the aforementioned ADAS system is a passive response, meaning that it only issues a warning when danger is about to occur, giving the driver of the following vehicle a short reaction time, resulting in the continued existence of potential safety hazards when following another vehicle. Summary of the Invention

[0004] This application provides a vehicle control method, device, vehicle, and storage medium to improve following safety.

[0005] In a first aspect, some embodiments of this application provide a vehicle control method, the method comprising: acquiring current driving data of the vehicle; determining recommended following parameters based on the current driving data using a personalized prediction model; wherein the personalized prediction model is obtained by fine-tuning a trained prediction model based on the historical driving data of a target driver; and controlling the vehicle's taillights to operate based on the recommended following parameters.

[0006] Secondly, some embodiments of this application also provide a vehicle control device, which includes an acquisition module, a determination module, and a control module. The acquisition module acquires the vehicle's current driving data. The determination module determines recommended following parameters based on the current driving data using a personalized prediction model; wherein the personalized prediction model is obtained by fine-tuning a trained prediction model based on the target driver's historical driving data. The control module controls the vehicle's taillights to operate based on the recommended following parameters.

[0007] Thirdly, some embodiments of this application also provide a vehicle including one or more processors, a memory, and one or more applications, wherein the one or more applications are stored in the memory and configured to be executed by one or more processors and configured to perform the methods described above.

[0008] Fourthly, embodiments of this application also provide a computer-readable storage medium storing computer program instructions. These computer program instructions can be invoked by a processor to execute the methods described above.

[0009] Fifthly, embodiments of this application also provide a computer program product that, when executed, implements the above-described method.

[0010] This application provides a vehicle control method, device, vehicle, and storage medium. In this method, the vehicle acquires current driving data, determines recommended following parameters based on a personalized prediction model, and finally controls the vehicle's taillights based on the recommended following parameters.

[0011] In this application, the vehicle employs an "active guidance" following warning strategy. It acquires real-time driving data (e.g., vehicle speed, distance to the vehicle in front, steering wheel angle, etc.) and determines recommended following parameters based on a personalized predictive model (e.g., recommended following distance and speed for the following vehicle). Clear following guidance is provided to the following vehicle through the taillights. Therefore, this application can proactively assist the driver of the following vehicle in adjusting their following strategy at an early stage of risk, preventing potential following hazards and improving following safety.

[0012] Furthermore, the personalized prediction model in this application is obtained by fine-tuning the trained prediction model based on the target driver's historical driving data. In other words, based on the trained prediction model, the target driver's historical driving data is used to fine-tune and train the model again to obtain the personalized prediction model. Here, "target driver" refers to the driver driving the vehicle.

[0013] This application achieves personalized fine-tuning of the prediction model, thus binding the trained model to the target driver. On one hand, fine-tuning the model using the target driver's historical driving data reduces training costs and improves training efficiency. On the other hand, the trained personalized prediction model can provide targeted following parameters based on the target driver's driving habits and style, aligning with the target driver's subjective feelings and objective capabilities, making the determined following parameters more accurate and reasonable. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1 This is a schematic diagram of the vehicle structure provided in the embodiments of this application.

[0016] Figure 2 This is a flowchart illustrating a vehicle control method provided in the first embodiment of this application.

[0017] Figure 3 This is a flowchart illustrating a vehicle control method provided in the second embodiment of this application.

[0018] Figure 4 This is a flowchart illustrating a vehicle control method provided in the third embodiment of this application.

[0019] Figure 5 This is a flowchart illustrating a vehicle control method provided in the fourth embodiment of this application.

[0020] Figure 6 This is a block diagram of the vehicle control device provided in the embodiments of this application.

[0021] Figure 7 This is a block diagram of the vehicle provided in the embodiments of this application. Detailed Implementation

[0022] To make the technical problems, technical solutions, and beneficial effects solved by this application clearer, the following detailed description is provided in conjunction with embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0023] In related technologies, Advanced Driver Assistance Systems (ADAS) typically control the taillights to display a warning symbol when the distance between the vehicle (own vehicle) and the vehicle in front is less than a safe distance, in order to remind the driver of the following vehicle to pay attention to driving safety. Specifically, the "safe distance" is usually calculated based on the vehicle's speed and a fixed formula, such as a distance calculation formula based on the "2-second rule".

[0024] Therefore, for different drivers, the "safe distance" in related technologies is a fixed value at the same speed, ignoring the crucial factor of individual driver differences and failing to provide truly "personalized" following guidance. Specifically, the "safe distance" perceived by an experienced, quick-reacting driver differs from that of a novice or conservative driver. General warning strategies may be overly sensitive to the former, and frequent triggering of warnings to following vehicles may cause them to deactivate their ADAS systems. Conversely, they may provide insufficient warnings to the latter, failing to effectively prevent following accidents.

[0025] In addition, the triggering mechanism of ADAS systems is a passive response, that is, it only issues a warning when danger is about to occur, giving the driver of the following vehicle a short reaction time, which means that the hidden dangers of following vehicles still exist.

[0026] To address the aforementioned problems, the inventors of this application propose a vehicle control method, device, vehicle, and storage medium. The method includes: acquiring the vehicle's current driving data; determining recommended following parameters based on the current driving data using a personalized prediction model; wherein the personalized prediction model is obtained by fine-tuning a trained prediction model based on the target driver's historical driving data; and controlling the vehicle's taillights to operate based on the recommended following parameters.

[0027] In this application, the vehicle employs an "active guidance" following warning strategy. It acquires real-time driving data (e.g., vehicle speed, distance to the vehicle in front, steering wheel angle, etc.) and determines recommended following parameters based on a personalized predictive model (e.g., recommended following distance and speed for the following vehicle). Clear following guidance is provided to the following vehicle through the taillights. Therefore, this application can proactively assist the driver of the following vehicle in adjusting their following strategy at an early stage of risk, preventing potential following hazards and improving following safety.

[0028] Furthermore, the personalized prediction model in this application is obtained by fine-tuning the prediction model based on the target driver's historical driving data. In other words, based on the trained prediction model, the target driver's historical driving data is used to fine-tune the model to obtain the personalized prediction model.

[0029] This application achieves personalized fine-tuning of the prediction model, thus binding the trained model to the target driver. On one hand, fine-tuning the model using the target driver's historical driving data reduces training costs and improves training efficiency. On the other hand, the trained personalized prediction model can provide targeted following parameters based on the target driver's driving habits and style, aligning with the target driver's subjective feelings and objective capabilities, making the determined following parameters more accurate and reasonable.

[0030] To facilitate a detailed explanation of the present application, the application environment of the embodiments of the present application will be described below with reference to the accompanying drawings. Please refer to... Figure 1The diagram illustrates the application environment of the control method provided in this application embodiment. The method is applied to vehicle 100, which refers to a means of transportation driven or towed by a power device for the purpose of carrying people or transporting goods. Vehicle 100 includes, but is not limited to, cars, suburban utility vehicles (SUVs), multi-purpose vehicles (MPVs), driverless ride-hailing vehicles, new energy electric vehicles, etc.

[0031] In this embodiment, the vehicle 100 may include a body 210, a center console 230, and taillights (not shown in the figure). The body 210 serves to fix and house the center console 230, taillights, and other components. The body 210 may include a driver's cabin (not shown in the figure), which is used to seat the driver, allowing the driver to control the vehicle 100 through a control system (e.g., center console 230, drive system, braking system, etc.) located within the driver's cabin.

[0032] The center console 230 is used to process signals and data, and control the vehicle body 210 to operate according to the processed parameters (e.g., forward movement, braking, steering, etc.). Specifically, the center console 230 can integrate one or more controllers, such as a vehicle domain controller (VDC). The VDC is responsible for integrating and managing multiple electronic control units (ECUs) to improve the intelligence and automation level of the vehicle 100.

[0033] In some possible embodiments, the central control unit 230 is used to fine-tune the trained prediction model to obtain a personalized prediction model, and to store the personalized prediction model.

[0034] In some other possible embodiments, vehicle 100 is communicatively connected to server 300, which may store a trained prediction model. Vehicle 100 sends the target driver's historical driving data to server 300, which is also used to fine-tune the trained prediction model to obtain a personalized prediction model. The trained personalized prediction model is then sent to the center console 230 of vehicle 100, where it is stored.

[0035] Therefore, in this embodiment, the personalized prediction model is trained by server 300, which can save the computing resources of vehicle 100 and ensure that there are sufficient computing resources to respond to environmental data and driving data during the driving process of vehicle 100. Specifically, server 300 can be the back-end server corresponding to vehicle 100, and the number of servers 300 can be one or a server network composed of multiple servers. This embodiment does not limit this.

[0036] In this embodiment, the central control unit 230 is also used to generate corresponding headlight control commands based on the recommended following parameters determined by the personalized prediction model, thereby controlling the rear headlights to operate. Specifically, the vehicle's rear headlights can be one or more of brake lights, rear turn signals, and rear fog lights.

[0037] In some possible embodiments, the taillight may include an array of LEDs (not shown in the figure), wherein the LEDs in the array may be arranged in an M*N pattern or an irregular pattern. Specifically, the taillight may be a digital taillight, a pixel taillight, etc., which can display a specified pattern, number, or text by illuminating at least some of the LEDs.

[0038] In some possible embodiments, the taillights may have image projection capabilities, which can use projection technology (e.g., DMD-based digital light processing technology) to project specified patterns, numbers, or text onto the road surface behind. Specifically, the taillights may be intelligent interactive lights.

[0039] The control method applied to the vehicle 100 in the above embodiments will be described below.

[0040] Please see Figure 2 The first embodiment of this application illustrates a vehicle control method, which may include the following steps.

[0041] Step S210: Obtain the vehicle's current driving data.

[0042] In this embodiment, current driving data refers to the driving data corresponding to the data reading moment. The driving data may include vehicle driving parameters and driver operation parameters. Specifically, vehicle driving parameters may include at least one of the following: vehicle speed, acceleration, distance between the vehicle (own vehicle) and the vehicle in front, and relative speed between the vehicle (own vehicle) and the vehicle in front. Driver operation parameters may include at least one of the following: steering wheel angle, accelerator pedal opening, and brake pedal opening.

[0043] As an example, driving data may include speed, acceleration, distance between the vehicle and the vehicle in front, steering wheel angle, throttle opening, and brake opening. In other possible examples, driving data may also include multiple other parameters, which are not specifically limited in this embodiment. Specifically, the data dimensions of the driving data correspond one-to-one with the data dimensions of the input parameters of the personalized prediction model. It is easy to understand that there is a positive correlation between the data dimensions of the driving data and the accuracy of the prediction results of the personalized prediction model. That is, the higher the data dimension of the driving data, the more comprehensively it can reflect the driver's driving habits and style, making the prediction results given by the model more accurate and reliable.

[0044] In one implementation, the vehicle responds to a start command by acquiring its current driving data at specified intervals. The specified interval can be a default value, determined by researchers based on extensive test data. For example, the specified interval could be 1 second, 3 seconds, 5 seconds, etc.

[0045] As another implementation method, the vehicle can acquire the relative speed between itself and the vehicle behind it. When the relative speed is greater than or equal to a preset speed, the vehicle acquires its current driving data. The preset speed can be a pre-set speed value, which can be derived by researchers based on extensive test data. Specifically, when the relative speed is greater than or equal to the preset speed, there is a high probability of a rear-end collision. In this situation, the vehicle acquires its current driving data. On the one hand, this allows for timely warnings to the following vehicle to remind it to pay attention to following safety; on the other hand, it avoids the vehicle frequently acquiring current driving data, thus saving the vehicle's computing resources and avoiding frequent control of the taillights. For example, the preset speed could be 30 km / h, 50 km / h, etc.

[0046] Step S220: Based on the current driving data, determine the recommended following parameters through a personalized prediction model.

[0047] In this embodiment, the personalized prediction model is obtained by fine-tuning the trained prediction model based on the target driver's historical driving data. The current driving data serves as the input parameter of the personalized prediction model, and the recommended following parameter is the output parameter. Here, "target driver" refers to the driver of the vehicle.

[0048] Specifically, recommended following parameters may include at least one of recommended following distance and recommended following speed. The recommended following distance is the recommended value for the distance between the following vehicle and the vehicle behind, and the recommended following speed is the recommended value for the speed of the following vehicle.

[0049] In some possible embodiments, the recommended following parameters may include only one of the recommended following distance and the recommended following speed to reduce the difficulty of model training. In other possible embodiments, the recommended following parameters may include both the recommended following distance and the recommended following speed to provide more comprehensive following suggestions to following vehicles.

[0050] It should be noted that the "prediction model" in this embodiment is a pre-trained big data model, obtained by training on a large amount of driving data from drivers. Specifically, the model architecture of the prediction model can be any one of the following: a temporal neural network model based on an attention mechanism (Transformer), a temporal convolutional network model, a long short-term memory network model, or a basic recurrent neural network model. This embodiment does not limit the specific training method of the prediction model.

[0051] It's easy to understand that if current driving data is directly input into a trained prediction model, the resulting recommended following parameters can only reflect the driving habits and styles of most drivers. This still results in the problem of "ignoring individual driver differences" and cannot provide truly "personalized" following guidance.

[0052] Therefore, in this embodiment, based on the trained prediction model, the model is further fine-tuned using the target driver's historical driving data to obtain a personalized prediction model. In this case, the fine-tuned model is bound to the target driver. On the one hand, fine-tuning the model using the target driver's historical driving data can reduce the model training cost and improve the model training efficiency; on the other hand, the trained personalized prediction model can provide targeted recommended following parameters based on the target driver's driving habits and style, which conforms to the target driver's subjective feelings and objective capabilities, making the determined recommended following parameters more accurate and reasonable.

[0053] In one implementation, the personalized prediction model can be pre-stored in the center console. Inputting current driving data into the personalized prediction model determines the corresponding recommended following parameters. In another implementation, the personalized prediction model can be stored in a server corresponding to the vehicle, with the vehicle communicating with the server. The vehicle sends its current driving data to the server and receives the recommended following parameters determined by the server based on the personalized prediction model. Specifically, the training process for the personalized prediction model is described below.

[0054] Step S230: Based on the recommended following parameters, control the vehicle's taillights to operate.

[0055] As one implementation method, the vehicle can directly display the recommended following parameters through the vehicle's taillights.

[0056] As another implementation method, the vehicle can determine corresponding headlight control commands based on recommended following parameters, such as headlight flashing commands, headlight color control commands, etc., and control the vehicle's taillights to operate based on these headlight control commands.

[0057] Specifically, the control logic for the taillights is described below in the instruction manual.

[0058] This application provides a vehicle control method. The vehicle in this method employs an "active guidance" following warning strategy, which acquires real-time driving data (e.g., vehicle speed, distance to the vehicle in front, steering wheel angle, etc.) and determines recommended following parameters (e.g., recommended following distance and speed for the following vehicle) based on a personalized prediction model. Clear following guidance is provided to the following vehicle through the taillights. Therefore, this application can proactively assist the driver of the following vehicle in adjusting their following strategy at an early stage of potential risk, preventing following hazards before they occur and improving following safety.

[0059] Furthermore, this embodiment of the application personalizes and fine-tunes the prediction model, thus binding the fine-tuned model to the target driver. On the one hand, fine-tuning the training using the target driver's historical driving data can reduce model training costs and improve training efficiency. On the other hand, the trained personalized prediction model can provide targeted recommended following parameters based on the target driver's driving habits and style, conforming to the target driver's subjective feelings and objective capabilities, making the determined recommended following parameters more accurate and reasonable.

[0060] This section introduces the training process of personalized prediction models, which may include the following steps.

[0061] Step A10: Obtain the target driver's historical driving data.

[0062] In this embodiment, historical driving data includes vehicle driving parameters and driver operation parameters.

[0063] As one implementation method, historical driving data can be data recorded from previous driving history. After each driving task is completed, the vehicle can store the driving data corresponding to this driving task (e.g., stored locally or in the cloud). In subsequent processes, the vehicle responds to model training instructions and obtains the target driver's historical driving data.

[0064] As another implementation method, historical driving data can also be data recorded during the current driving task. For example, in response to a start command, the vehicle uses driving data within a preset time period (e.g., 10 minutes) as historical driving data. After the vehicle has been driving for a period of time, a personalized prediction model is trained based on the acquired historical driving data.

[0065] In other words, the personalized prediction model in this embodiment is trained on the vehicle's current driving task to ensure that it can more accurately and reliably reflect the driver's driving habits. Furthermore, it avoids situations where the personalized prediction model is not trained on the current driver's historical driving data when the driver changes, thus ensuring the accuracy of subsequent recommended following parameters.

[0066] Step A20: Filter the historical driving data to determine the target driving data.

[0067] In this embodiment, the target driving data refers to the driving data corresponding to the absence of a braking command within a preset time period. The preset time period can be a default value, which can be derived by researchers based on a large amount of test data. For example, the preset time period could be 5s, 8s, 10s, etc. The "braking command" here can be generated by the vehicle's internal control system or by the vehicle in response to the driver's braking operation.

[0068] For example, a vehicle's Automatic Emergency Braking System (AEBS) uses sensors (such as radar and cameras) at the front of the vehicle to monitor the distance and relative speed to the vehicle in front or obstacles. When AEBS determines that a collision is imminent and the driver has not taken sufficient braking measures, it will automatically generate a braking command and apply full or partial braking force to the vehicle to avoid a collision or minimize its impact.

[0069] For example, a vehicle's Forward Collision Warning System (FCWS) continuously monitors the distance and relative speed to other vehicles, pedestrians, or obstacles using sensors (such as radar and cameras) located in front of the vehicle. When the FCWS calculates a potential collision risk and the driver fails to take appropriate braking or steering action, it will issue a braking command.

[0070] Therefore, this embodiment filters historical driving data and identifies the "positive sample data" in the historical driving data as target driving data. This target driving data can reflect the following behavior pattern that the driver subjectively considers "safe and comfortable" during driving, reducing the interference of external environmental factors on driving data. This allows the personalized prediction model, which is subsequently fine-tuned and trained based on the target driving data, to more accurately predict the recommended following parameters, thereby improving the prediction accuracy of the personalized prediction model.

[0071] Step A30: Based on the target driving data, fine-tune the trained prediction model to determine the personalized prediction model.

[0072] In this embodiment, the prediction model is trained based on driving data from multiple drivers and is used to determine recommended following parameters.

[0073] As one implementation method, the vehicle can acquire the model training strategy corresponding to the prediction model, which can be stored locally or in the cloud. For example, the model training strategy may include the formula for the loss function, the selection of the optimizer (Adam, AdamW, etc.), the value of the learning rate, gradient pruning strategies, etc. Then, based on the same model training strategy, using target driving data as training samples, the trained prediction model is trained again to obtain a personalized prediction model.

[0074] This embodiment does not limit the specific training process of the personalized prediction model. In some possible embodiments, the vehicle can pre-set the number of training loops (e.g., 10, 20, 40, etc.). When the current number is the number of training loops, the model result obtained in this training is used as the personalized prediction model. In other possible embodiments, the vehicle can pre-set a loss threshold and use a validation set to calculate the loss of the model result obtained in each training iteration. If the loss is less than or equal to the loss threshold, the model result obtained in this training is used as the personalized prediction model.

[0075] It should be noted that, since the training samples used for fine-tuning the prediction model are the driving data of a single driver (i.e., the target driver), the amount of training data is relatively small. Therefore, the training task of the personalized prediction model can be completed within the vehicle's center console. Of course, to save vehicle computing resources, the training task of the personalized prediction model can also be completed on a server; this embodiment does not limit this approach.

[0076] Please see Figure 3 The second embodiment of this application illustrates a vehicle control method, which specifically describes a control strategy for the taillights. The method may include the following steps.

[0077] Step S310: Obtain the vehicle's current driving data.

[0078] Specifically, the implementation method of step S310 can be found in the relevant description of step S210, and will not be repeated here.

[0079] In some possible embodiments, the historical driving data used for fine-tuning the personalized prediction model and the vehicle's current driving data can both be data recorded in the same driving task, with the historical driving data acquired earlier than the current driving data. In this case, the vehicle first completes the fine-tuning training of the personalized prediction model based on the historical driving data, and then predicts the recommended following parameters based on the current driving data.

[0080] Therefore, the personalized prediction model can accurately reflect the current driver's driving habits, avoiding situations where the current driver and the target driver corresponding to the personalized prediction model are not the same person, thus ensuring the accuracy of subsequent recommended following parameters. Furthermore, due to the small amount of training sample data, the vehicle can complete the training task of the personalized prediction model in a timely manner during this driving task, ensuring that recommended following parameters are reliably determined in subsequent driving processes.

[0081] Step S320: Based on the current driving data, determine the recommended following parameters through a personalized prediction model.

[0082] Specifically, the implementation method of step S320 can be found in the relevant description of step S220, which will not be repeated here.

[0083] In some possible embodiments, the personalized prediction model can be trained on the vehicle's current driving task. In this case, the vehicle can directly input the current driving data into the personalized prediction model to obtain recommended following parameters.

[0084] In other possible embodiments, the personalized prediction model is pre-trained and stored in the vehicle. To avoid the situation where multiple drivers share a vehicle and the "current driver and the target driver corresponding to the personalized prediction model are not the same person," the vehicle in this embodiment can store the identity information of the target driver corresponding to the personalized prediction model along with the personalized prediction model.

[0085] In subsequent processes, the vehicle can obtain the current driver's identity information (e.g., facial recognition, voiceprint verification, etc.). If the current driver's identity information matches the target driver's identity information, the current driving data is input into a personalized prediction model to obtain recommended following parameters. If the two identity information do not match, the vehicle can use the current driving data as historical driving data to fine-tune and train the prediction model, obtaining a personalized prediction model corresponding to the current driver, and then determining the recommended following parameters based on this model. Of course, in subsequent processes, the vehicle can store the current driver's identity information and the corresponding personalized prediction model to establish a model library that stores the identity information of different drivers and different personalized prediction models.

[0086] Step S330: Based on the recommended following parameters, control the vehicle's taillights to operate.

[0087] In this embodiment, the recommended following parameters may include a recommended following distance. Specifically, step S330 may include steps S3310 to S3330.

[0088] Step S3310: Obtain the actual following distance between the following vehicle and the vehicle behind.

[0089] As one implementation method, a vehicle can obtain the actual following distance between vehicles by using radar (e.g., millimeter-wave radar, ultrasonic radar) installed at the rear of the vehicle.

[0090] As another implementation method, the vehicle can acquire rear images of the vehicle through a rear-facing camera installed at the rear of the vehicle, and use computer vision algorithms to detect, identify and track targets in the rear images, and calculate the actual following distance between the following vehicles.

[0091] Step S3320: Determine the headlight control command based on the recommended following distance and the actual following distance.

[0092] In this embodiment, the vehicle follows the recommended distance and the actual distance. Based on the actual following situation of the vehicle behind, the corresponding headlight control command can be determined to improve the following safety of the vehicle behind.

[0093] As one implementation method, the vehicle can determine the corresponding headlight control command based on the difference between the recommended following distance and the actual following distance.

[0094] For example, if the difference between the recommended following distance and the actual following distance is less than 0, it means that the actual following distance is greater than the recommended following distance. The headlight control command can be to display green text (e.g., "Following Safe").

[0095] For example, if the difference between the recommended following distance and the actual following distance is greater than 0 but less than the specified distance (e.g., 10m or 20m), it means that the actual following distance is less than the recommended following distance, but the difference between the two is not significant. The headlight control command can be to display yellow text (e.g., "Increase following distance, pay attention to safety") and control the taillights to flash at a low frequency.

[0096] For example, if the difference between the recommended following distance and the actual following distance is greater than or equal to the specified distance, it means that the actual following distance is much smaller than the recommended following distance. The headlight control command can be to display red text (e.g., "Slow down as soon as possible!" or "Emergency brake!" etc.) and control the taillights to flash at a high frequency.

[0097] As another implementation, the vehicle can determine the corresponding headlight control command based on the ratio between the recommended following distance and the actual following distance. Since the ratio more accurately reflects the numerical value, it provides a more intuitive explanation of the following vehicle's behavior, ensuring the vehicle can issue more accurate headlight control commands. Specifically, step S3320 may include steps S3321 and S3323.

[0098] Step S3321: Determine the ratio between the actual following distance and the recommended following distance.

[0099] Step S3323: Determine the headlight control command based on the ratio.

[0100] In this embodiment, the urgency of the headlight control command and its ratio are negatively correlated. As one implementation, the vehicle can pre-store a headlight control command mapping table, which represents the correspondence between different ratios and different headlight control commands. For ease of explanation below, the ratio will be denoted as K.

[0101] As an example, when K ≥ 1.2, the headlight control command could be: displaying green text (e.g., "Maintain a distance of 50m"). When 1.0 ≤ K < 1.2, the headlight control command could be: displaying blue text (e.g., "Maintain a distance of 35m"). When 0.7 ≤ K < 1.0, the headlight control command could be: displaying yellow text (e.g., "Maintain a distance of 25m!") and controlling the taillights to flash at a low frequency. When K < 0.7, the headlight control command could be: displaying red text (e.g., "Emergency braking!") and controlling the taillights to flash at a high frequency.

[0102] It should be noted that the ratio range and headlight control commands in this embodiment are merely exemplary. Researchers can formulate corresponding headlight control commands based on the actual situation of the vehicle, and this embodiment does not impose any limitations.

[0103] Step S3330: Control the vehicle's taillights to operate based on the headlight control command.

[0104] In some possible embodiments, step S3330 may include steps S3332 and S3324.

[0105] Step S3332: Determine the flashing frequency of the headlights based on the headlight control command.

[0106] In this embodiment, the headlight flashing frequency and the ratio are negatively correlated. That is, the smaller the ratio, the greater the actual following distance is compared to the recommended following distance. In this case, there is a high risk of a rear-end collision, and the headlights can be controlled to flash at a high frequency (e.g., a flashing frequency greater than or equal to 2Hz). Conversely, the larger the ratio, the greater the actual following distance is compared to the recommended following distance. In this case, the following vehicle is at a relatively safe following distance, and the headlights can be controlled to flash at a low frequency (e.g., a flashing frequency less than or equal to 1Hz) or not flash at all.

[0107] As one implementation method, the vehicle can directly read the headlight flashing frequency from the headlight control commands. For example, the headlight flashing frequency can be 0Hz, 1Hz, 2Hz, 4Hz, etc.

[0108] Step S3334: Control the vehicle's taillights to flash according to the headlight flashing frequency.

[0109] This application provides a vehicle control method. The vehicle in this method employs an "active guidance" following warning strategy, which acquires real-time driving data (e.g., vehicle speed, distance to the vehicle in front, steering wheel angle, etc.) and determines recommended following parameters (e.g., recommended following distance and speed for the following vehicle) based on a personalized prediction model. Clear following guidance is provided to the following vehicle through the taillights. Therefore, this application can proactively assist the driver of the following vehicle in adjusting their following strategy at an early stage of potential risk, preventing following hazards before they occur and improving following safety.

[0110] Please see Figure 4 This document illustrates a vehicle control method provided in the third embodiment of this application, which specifically describes another control strategy corresponding to the taillights. The method may include the following steps.

[0111] Step S410: Obtain the vehicle's current driving data.

[0112] Step S420: Based on the current driving data, determine the recommended following parameters through a personalized prediction model.

[0113] Specifically, the implementation methods of steps S410 and S420 can be found in the relevant descriptions in the first and second embodiments above the specification, and will not be repeated here.

[0114] Step S430: Based on the recommended following parameters, control the vehicle's taillights to operate.

[0115] In this embodiment, the recommended following parameters may include a recommended following speed. Specifically, step S430 may include step S4310.

[0116] Step S4310: Control the vehicle's taillights to display the recommended following speed.

[0117] In this embodiment, the vehicle displays the recommended following speed determined by the personalized prediction model directly through the taillights, allowing the driver of the following vehicle to intuitively see the recommended following speed and adjust the actual following speed accordingly, thereby improving following safety. For example, the text displayed on the taillights could be "Recommended speed: XX km / h".

[0118] In some possible embodiments, the vehicle's taillights have image projection capabilities, and the vehicle's taillights include an array of LEDs. Specifically, step S4310 may include steps S4312 to S4316.

[0119] Step S4312: Obtain the actual following distance between the following vehicle and the vehicle behind.

[0120] Specifically, the implementation method of step S4312 can be found in the relevant description in step S3310, and will not be repeated here.

[0121] Step S4314: When the actual following distance is greater than or equal to the preset distance, control the vehicle's taillights to project the recommended following speed onto the road surface.

[0122] In this embodiment, the preset distance can be a default value, which can be derived by researchers based on a large amount of test data. For example, the preset distance could be 15m, 20m, etc. Specifically, when the actual following distance is greater than or equal to the preset distance, it indicates that the distance between the following vehicle and the vehicle is relatively large. In this case, if the text corresponding to the recommended following speed is directly displayed on the taillights, it is difficult to ensure that the driver of the following vehicle can see or even clearly see the text on the taillights in time. Therefore, in this embodiment, when the distance between the two is large, the text corresponding to the recommended following speed is projected onto the road surface using the taillights to ensure that the driver of the following vehicle can see the text in time and adjust the actual following speed accordingly.

[0123] Step S4316: When the actual following distance is less than the preset distance, control the LED array to display the recommended following speed.

[0124] In this embodiment, when the actual following distance is less than a preset distance, the LED array is controlled to display the recommended following speed, so that the driver of the following vehicle can intuitively see the recommended following speed.

[0125] In some possible embodiments, steps S440 and S450 may be included after step S430.

[0126] Step S440: In response to the vehicle's engine shutdown command, acquire the vehicle's current driving data.

[0127] In this embodiment, the vehicle responds to a shutdown command, indicating that the vehicle's power system is off and the driver currently has no need to drive. In this situation, the vehicle acquires the driving data for this specific driving task.

[0128] Step S450: Based on the driving data, fine-tune and train the personalized prediction model.

[0129] In this embodiment, the vehicle uses the current driving data as a new set of historical driving data to fine-tune and train the personalized prediction model again. Specifically, the implementation methods for fine-tuning and training the personalized prediction model can be found in the relevant description above in the instruction manual, and will not be elaborated upon here.

[0130] It's easy to understand that as the target driver's driving experience increases, their driving habits and style will also change. Therefore, in this embodiment, after each driving task is completed, the personalized prediction model is continuously fine-tuned and trained based on the driving data from that task. This ensures that the recommended following parameters predicted by the personalized prediction model match the target driver's current driving habits and style, thereby ensuring that the prediction results given by the model are more accurate and reliable.

[0131] This application provides a vehicle control method. The vehicle in this method employs an "active guidance" following warning strategy, which acquires real-time driving data (e.g., vehicle speed, distance to the vehicle in front, steering wheel angle, etc.) and determines recommended following parameters (e.g., recommended following distance and speed for the following vehicle) based on a personalized prediction model. Clear following guidance is provided to the following vehicle through the taillights. Therefore, this application can proactively assist the driver of the following vehicle in adjusting their following strategy at an early stage of potential risk, preventing following hazards before they occur and improving following safety.

[0132] Please see Figure 5This illustrates a vehicle control method provided in the fourth embodiment of this application, which may include the following steps.

[0133] Step B10: Obtain the vehicle's current driving data.

[0134] Step B20: Obtain the target driver's historical driving data.

[0135] Step B21: Based on historical driving data, fine-tune the trained prediction model to determine the personalized prediction model.

[0136] Step B22: Based on the current driving data, determine the recommended following parameters using a personalized prediction model.

[0137] In this embodiment, the recommended following parameters include the recommended following distance.

[0138] Step B30: Calculate the ratio K of the actual following distance to the recommended following distance between the following vehicles.

[0139] Step B31: When K ≥ 1.2, control the rear lights to display green text.

[0140] Step B32: When 1.0 ≤ K < 1.2, control the rear lights to display blue text.

[0141] Step B33: When 0.7 ≤ K < 1.0, control the taillights to display yellow text and flash at a low frequency.

[0142] Step B34: When K < 0.7, control the taillights to display red text and flash at high frequency.

[0143] Step B40: Control the vehicle's taillights to operate.

[0144] This application provides a vehicle control method. The vehicle in this method employs an "active guidance" following warning strategy, which acquires real-time driving data (e.g., vehicle speed, distance to the vehicle in front, steering wheel angle, etc.) and determines recommended following parameters (e.g., recommended following distance and speed for the following vehicle) based on a personalized prediction model. Clear following guidance is provided to the following vehicle through the taillights. Therefore, this application can proactively assist the driver of the following vehicle in adjusting their following strategy at an early stage of potential risk, preventing following hazards before they occur and improving following safety.

[0145] Please see Figure 6This illustration shows a vehicle control device 600 according to an embodiment of this application. The control device 600 may include an acquisition module 610, a determination module 620, and a control module 630. The acquisition module 610 acquires the vehicle's current driving data. The determination module 620 determines recommended following parameters based on the current driving data using a personalized prediction model; wherein the personalized prediction model is obtained by fine-tuning a trained prediction model based on the target driver's historical driving data. The control module 630 controls the vehicle's taillights to operate based on the recommended following parameters.

[0146] In some possible embodiments, the control device 600 may further include a model fine-tuning module (not shown in the figure), which is used to acquire historical driving data of the target driver, including vehicle driving parameters and driver operation parameters; filter the historical driving data to determine target driving data; the target driving data refers to the driving data corresponding to the absence of a braking command within a preset time period; and fine-tune the trained prediction model based on the target driving data to determine a personalized prediction model; wherein the prediction model is trained based on the driving data of multiple drivers and is used to determine recommended following parameters; the recommended following parameters include at least one of recommended following distance and recommended following speed.

[0147] In some possible embodiments, the recommended following parameters include a recommended following distance, which is a recommended value for the distance between the following vehicle and the vehicle itself. The control module 630 is specifically used to obtain the actual following distance between the following vehicle and the vehicle itself; determine the headlight control command based on the recommended following distance and the actual following distance; and control the vehicle's taillights to operate based on the headlight control command.

[0148] In some possible embodiments, the control module 630 is specifically used to determine the ratio between the actual following distance and the recommended following distance; based on the ratio, to determine the headlight control command; wherein the urgency of the headlight control command is negatively correlated with the ratio.

[0149] In some possible embodiments, the control module 630 is specifically used to determine the headlight flashing frequency based on the headlight control command; the headlight flashing frequency and the ratio are negatively correlated; and control the vehicle's taillights to flash according to the headlight flashing frequency.

[0150] In some possible embodiments, the recommended following parameter also includes a recommended following speed, which is a recommended value for the speed of the following vehicle. The control module 630 is specifically used to control the vehicle's taillights to display the recommended following speed.

[0151] In some possible embodiments, the vehicle's taillights have image projection capabilities, and the taillights include an array of LEDs. The control module 630 is specifically used to acquire the actual following distance between the vehicle and the following vehicle; if the actual following distance is greater than or equal to a preset distance, it controls the vehicle's taillights to project a recommended following speed onto the road surface; if the actual following distance is less than the preset distance, it controls the LED array to display the recommended following speed.

[0152] In some possible embodiments, the control device 600 may further include a model fine-tuning module (not shown in the figure), and the acquisition module 610 is further configured to acquire the vehicle's current driving data in response to the vehicle's engine shutdown command. The model fine-tuning module is also configured to fine-tune and train a personalized prediction model based on the current driving data.

[0153] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the above-described device and module can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0154] In the several embodiments provided in this application, the coupling between modules can be electrical, mechanical, or other forms of coupling.

[0155] Furthermore, the functional modules in the various embodiments of this application can be integrated into a single control module, or each module can exist physically separately, or two or more modules can be integrated into a single module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0156] This application provides a vehicle control device that employs an "active guidance" following warning strategy. It acquires real-time driving data (e.g., vehicle speed, distance to the vehicle in front, steering wheel angle, etc.) and determines recommended following parameters (e.g., recommended following distance and speed for the following vehicle) based on a personalized prediction model. It then provides clear following guidance to the following vehicle through taillight signals. Therefore, this application can proactively assist the driver of the following vehicle in adjusting their following strategy at an early stage of potential risk, preventing following hazards before they occur and improving following safety.

[0157] Please see Figure 7 The illustration shows a vehicle 700 provided in an embodiment of this application. The vehicle 700 may include one or more processors 710, a memory 720, and one or more application programs. The one or more application programs are stored in the memory 720 and configured to be executed by the one or more processors 710. The one or more application programs are configured to perform the methods described in the above embodiments.

[0158] The processor 710 may include one or more processing cores. The processor 710 connects to various parts of the entire battery management system using various interfaces and lines, and performs various functions and processes data of the battery management system by running or executing instructions, programs, code sets, or instruction sets stored in the memory 720, and by calling data stored in the memory 720. Optionally, the processor 710 may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor 710 may integrate one or a combination of several of the following: Central Processing Unit (CPU), Graphics Processing Unit (GPU), and modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the displayed content; and the modem handles wireless communication. It is understood that the modem may also not be integrated into the processor 710 and may be implemented separately using a communication chip.

[0159] The memory 720 may include random access memory (RAM) or read-only memory (ROM). The memory 720 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 720 may include a program storage area and a data storage area. The program storage area may store instructions for implementing an operating system, instructions for implementing at least one function (e.g., touch functionality, sound playback functionality, image playback functionality, etc.), and instructions for implementing the various method embodiments described above. The data storage area may also store data created during the use of the electronic device (e.g., phonebook, audio / video data, chat log data, etc.).

[0160] This application also provides a computer-readable storage medium storing computer program instructions that can be invoked by a processor to execute the methods described in the above embodiments.

[0161] Computer-readable storage media can be, for example, flash memory, electrically erasable programmable read-only memory (EEPROM), electrically programmable read-only memory (EPROM), hard disk, or read-only memory (ROM). Optionally, computer-readable storage media includes non-transitory computer-readable storage medium. The computer-readable storage medium has storage space for computer program instructions that perform any of the method steps described above. These computer program instructions can be read from or written to one or more computer program products.

[0162] In this application, "multiple" refers to two or more.

[0163] In this application, unless otherwise expressly defined, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0164] The terms “first,” “second,” “third,” “fourth,” etc., in this application (if present) are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0165] In this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, in this application, the character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0166] Unless otherwise specified, all steps in this application may be performed sequentially or randomly. For example, if the method includes steps A and B, it means that the method may include steps A and B performed sequentially, or it may include steps B and A performed sequentially. For example, if the method may also include step C, it means that step C may be added to the method in any order. For example, the method may include steps A, B, and C, or it may include steps A, C, and B, or it may include steps C, A, and B, etc.

[0167] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for controlling a vehicle, characterized in that, include: Obtain the vehicle's current driving data; Based on the current driving data, recommended following parameters are determined through a personalized prediction model; wherein, the personalized prediction model is obtained by fine-tuning the trained prediction model based on the target driver's historical driving data. Based on the recommended following parameters, the vehicle's taillights are controlled to operate.

2. The method according to claim 1, characterized in that, The training process of the personalized prediction model includes: Acquire the target driver's historical driving data, which includes vehicle driving parameters and driver operation parameters; The historical driving data is filtered to determine the target driving data; the target driving data refers to the driving data that has not triggered a braking command within a preset time period. Based on the target driving data, the trained prediction model is fine-tuned to determine the personalized prediction model; wherein, the prediction model is trained based on the driving data of multiple drivers and is used to determine the recommended following parameters; the recommended following parameters include at least one of recommended following distance and recommended following speed.

3. The method according to claim 1 or 2, characterized in that, The recommended following parameters include a recommended following distance, which is a recommended value for the distance between the following vehicle and the vehicle. The step of controlling the vehicle's taillights to operate based on the recommended following parameters includes: Obtain the actual following distance between the following vehicle and the vehicle in question; Based on the recommended following distance and the actual following distance, determine the headlight control command; The vehicle's taillights are controlled to operate based on the headlight control commands.

4. The method according to claim 3, characterized in that, The step of determining the headlight control command based on the recommended following distance and the actual following distance includes: Determine the ratio between the actual following distance and the recommended following distance; Based on the ratio, the vehicle light control command is determined; wherein the urgency of the vehicle light control command is negatively correlated with the ratio. The control of the vehicle's taillights to operate based on the headlight control commands includes: The headlight flashing frequency is determined based on the headlight control command; the headlight flashing frequency and the ratio are negatively correlated. The vehicle's taillights are controlled to flash at the same frequency as the vehicle lights.

5. The method according to claim 1 or 2, characterized in that, The recommended following parameters also include a recommended following speed, which is a recommended value for the speed of the following vehicle. The step of controlling the vehicle's taillights to operate based on the recommended following parameters includes: The vehicle's taillights are controlled to display the recommended following speed.

6. The method according to claim 5, characterized in that, The vehicle's taillights have an image projection function, and the vehicle's taillights include an array of LEDs; controlling the vehicle's taillights to display the recommended following speed includes: Obtain the actual following distance between the following vehicle and the vehicle in question; When the actual following distance is greater than or equal to the preset distance, the vehicle's taillights are controlled to project the recommended following speed onto the road surface. If the actual following distance is less than the preset distance, the LED array is controlled to display the recommended following speed.

7. The method according to claim 1 or 2, characterized in that, The method further includes: In response to the vehicle's engine shutdown command, acquire the vehicle's current driving data; Based on the driving data, the personalized prediction model is fine-tuned and trained.

8. A vehicle control device, characterized in that, include: The acquisition module is used to acquire the vehicle's current driving data; The determination module is used to determine recommended following parameters based on the current driving data using a personalized prediction model; wherein the personalized prediction model is obtained by fine-tuning the trained prediction model based on the target driver's historical driving data. The control module is used to control the taillights of the vehicle to operate based on the recommended following parameters.

9. A vehicle, characterized in that, include: One or more processors; Memory; as well as One or more applications, wherein the one or more said applications are stored in the memory and configured to be executed by one or more said processors and configured to perform the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer program instructions that can be invoked by a processor to perform the method as described in any one of claims 1 to 7.

Images