Vehicle auxiliary driving method and related equipment

Abstract

The invention discloses a vehicle auxiliary driving method and related equipment, and relates to the technical field of intelligent driving, and the method comprises the steps: determining a driving style label corresponding to a target vehicle terminal based on vehicle driving data, and updating the model parameters of a local intelligent driving model; updating information of the model parameters and the driving style label are uploaded to a server side; vehicle aided driving is carried out according to a target global intelligent driving model corresponding to the driving style label issued by the server side, and the target global intelligent driving model is a model obtained by carrying out parameter updating on an initial global intelligent driving model corresponding to the driving style label by the server side according to the updating information. According to the method, through personalized label driving, privacy protection, distributed incremental optimization and rapid iterative loop, the adaptability, safety, comfort and energy-saving performance of the intelligent driving model are comprehensively improved.

Description

Technical Field

[0001] The present application relates to the field of intelligent driving technology, and more specifically, to a vehicle assisted driving method and related equipment. Background Art

[0002] With the continuous development of intelligent driving technology, vehicle assisted driving has been widely used in the automotive industry as an important means to improve driving safety, comfort and energy efficiency; however, the diversified application scenarios and personalized needs of vehicle assisted driving systems make the optimization and adaptability of intelligent driving models particularly critical; therefore, how to achieve efficient updating and personalized optimization of the model under privacy protection has become an important technical challenge.

[0003] In related technologies, vehicle assisted driving systems usually use general models for driving decisions. Although this method has certain versatility, it ignores the individual differences of drivers and the diverse needs of driving scenarios, and it is difficult to meet the personalized preferences of different users for driving styles. In addition, traditional federated learning methods mainly rely on uploading full model parameters for server-side aggregation optimization. In large-scale model applications, this method has problems such as large data transmission volume, insufficient privacy protection, and high local device resource usage, resulting in low model update efficiency and insufficient personalized adaptation capabilities. Especially in dynamic driving environments, these technical bottlenecks may cause the assisted driving system to be unable to accurately adapt to user needs or complex scenarios. That is, the existing technology has technical problems in vehicle assisted driving, such as low model update efficiency, insufficient personalized adaptation, and limited privacy protection capabilities. Summary of the invention

[0004] A series of simplified concepts are introduced in the summary of the invention, which will be further described in detail in the detailed description. The summary of the invention of this application does not mean to attempt to define the key features and essential technical features of the technical solution claimed for protection, nor does it mean to attempt to determine the scope of protection of the technical solution claimed for protection.

[0005] The vehicle assisted driving method and related devices provided in this application can comprehensively improve the adaptability, safety, comfort and energy saving of the intelligent driving model through personalized tag drive, privacy protection, distributed incremental optimization and rapid iteration cycle.

[0006] In a first aspect, the present application provides a vehicle assisted driving method, which is applied to a target vehicle terminal, comprising: determining a driving style label corresponding to the target vehicle terminal based on vehicle driving data, and updating model parameters of a local intelligent driving model; uploading the update information of the model parameters and the driving style label to a server; performing vehicle assisted driving according to a target global intelligent driving model corresponding to the driving style label issued by the server, wherein the target global intelligent driving model is a model obtained by updating the parameters of an initial global intelligent driving model corresponding to the driving style label by the server according to the update information.

[0007] In a feasible implementation, the method of determining the driving style label corresponding to the target vehicle terminal based on the vehicle driving data and updating the model parameters of the local intelligent driving model includes: determining the driving style label corresponding to the target vehicle terminal based on the vehicle driving data or the driver's label selection; and updating the model parameters of the local intelligent driving model using a heuristic algorithm based on the vehicle driving data, wherein the heuristic algorithm includes a genetic algorithm and a simulated annealing algorithm.

[0008] In a feasible implementation, after updating the model parameters of the local intelligent driving model using a heuristic algorithm based on the vehicle driving data, the vehicle assisted driving method further includes: updating the model parameters of the local intelligent driving model in a gradient update manner based on the driver's preference selection.

[0009] In a feasible implementation manner, the vehicle assisted driving method also includes: uploading a first data amount of the vehicle driving data to the server side; the target global intelligent driving model is a model obtained by the server side updating the parameters of the initial global intelligent driving model corresponding to the driving style label according to the update information and the model parameter weight, wherein the model parameter weight is calculated based on the first data amount and the second data amount, and the second data amount is the data amount of the training data of the initial global intelligent driving model.

[0010] In a feasible implementation manner, the vehicle assisted driving method further includes: locally encrypting and storing the vehicle driving data, and setting read and write permissions to prohibit uploading to the server.

[0011] In a feasible implementation manner, the vehicle driving data includes vehicle dynamic behavior data, driver concentration level and vehicle type.

[0012] In the second aspect, the present application also provides a vehicle assisted driving device, comprising: a local parameter updating unit, used to determine the driving style label corresponding to the target vehicle terminal based on the vehicle driving data, and update the model parameters of the local intelligent driving model; a data uploading unit, used to upload the update information of the model parameters and the driving style label to the server; an assisted driving unit, used to perform vehicle assisted driving according to the target global intelligent driving model corresponding to the driving style label issued by the server, wherein the target global intelligent driving model is a model obtained by the server side updating the parameters of the initial global intelligent driving model corresponding to the driving style label according to the update information.

[0013] In a third aspect, the present application further provides an electronic device, comprising: a memory and a processor, wherein the processor is used to implement the steps of the vehicle assisted driving method described in the first aspect when executing a computer program stored in the memory.

[0014] In a fourth aspect, the present application further provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the vehicle assisted driving method described in the first aspect.

[0015] In a fifth aspect, the present application also provides a computer program product, including a computer program or computer executable instructions, which, when executed by a processor, implements the vehicle assisted driving method provided in an embodiment of the present application.

[0016] In the sixth aspect, the present application also provides a vehicle assisted driving system, including a vehicle terminal, used to execute the vehicle assisted driving method provided in the embodiment of the present application; a server side, used to receive update information and driving style labels of model parameters from the vehicle terminal, and send the target global intelligent driving model generated according to the update information and the driving style label to the vehicle terminal.

[0017] In summary, this application generates exclusive driving style labels (such as energy-saving driving, comfortable driving, etc.) for each driver through the analysis of vehicle driving data, and optimizes the intelligent driving model locally to make it more in line with individual driving behavior and preferences; only uploads model update information and driving style labels to the server, avoiding the transmission of original data and effectively reducing the risk of privacy leakage; the local terminal is responsible for local model training and updating, and only uploads incremental information, while the server optimizes and integrates the global model, significantly reducing the amount of data transmission and server computing burden, while forming a fast and efficient model optimization iteration cycle to ensure that the intelligent driving model can quickly adapt to dynamic needs; the optimized target global model is sent to the vehicle terminal through the cloud, and is used for assisted driving in real time, effectively improving driving safety, comfort and energy saving. In summary, the vehicle assisted driving method provided by this application comprehensively improves the adaptability, safety, comfort and energy saving of the intelligent driving model through personalized label drive, privacy protection, distributed incremental optimization and fast iteration cycle. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present specification. Also, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0019] Figure 1 A schematic diagram of a process flow of a vehicle assisted driving method provided in an embodiment of the present application;

[0020] Figure 2 A schematic diagram of the structure of a vehicle auxiliary driving device provided in an embodiment of the present application;

[0021] Figure 3 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0022] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, rather than to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "is" and "has" and any variations thereof involved in the present application are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0023] The term "module" or "unit" in this application refers to a computer program or a part of a computer program that has a predetermined function and works together with other related parts to achieve a predetermined goal, and can be implemented in whole or in part by using software, hardware (such as processing circuits or memories) or a combination thereof. Similarly, a processor (or multiple processors or memories) can be used to implement one or more modules or units. In addition, each module or unit can be part of an overall module or unit that includes the function of the module or unit.

[0024] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. In the following description, it is related to "some embodiments", which describes a subset of all possible embodiments, but it can be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments, and can be combined with each other without conflict.

[0025] See also Figure 1 , Figure 1 1 is a flow chart of a vehicle assisted driving method provided by an embodiment of the present application. The method may specifically include the following steps 101 to 103:

[0026] Step 101: determine a driving style label corresponding to a target vehicle terminal based on vehicle driving data, and update model parameters of a local intelligent driving model;

[0027] Specifically, vehicle driving data refers to various data collected by the target vehicle during driving, which may include but are not limited to vehicle speed, acceleration, braking force, steering angle, accelerator pedal position, GPS positioning, road slope, climate information and driver operation behavior. The target vehicle terminal is a terminal device on the target vehicle (i.e., the vehicle undergoing assisted driving), which may be a computing device, sensor or controller in the vehicle, etc., responsible for collecting vehicle driving data and running the local intelligent driving model. The driving style label is to classify the driver's driving style and assign a label, reflecting the driver's driving behavior characteristics within a certain period of time, such as "intense driving", "gentle driving", "smooth driving", etc.; for the driving style label, it can be determined based on the vehicle's acceleration, braking force, speed, steering angle and other dynamic driving data, combined with a certain algorithm model (such as cluster analysis, decision tree or neural network, etc.). The local intelligent driving model is an intelligent driving model running on the target vehicle terminal, which is used for real-time decision-making of assisted driving. According to the vehicle's sensor data, driving style label and other input data, it predicts and makes decisions on driving behavior in real time, and generates specific assisted driving instructions (such as acceleration, braking, steering, automatic parking, lane keeping, adaptive cruise control, etc.).

[0028] Through the implementation of step 101, the local intelligent driving model is optimized according to different driving styles, and can make more accurate auxiliary decisions during the driving process; for example, a driver with an aggressive driving style may require stronger acceleration support, while a driver with a gentle driving style may pay more attention to a smooth driving experience.

[0029] Step 102, uploading the updated information of the model parameters and the driving style label to the server;

[0030] Specifically, the update information of the model parameters is the parameter change data of the local intelligent driving model after being updated in step 101, which may specifically include the adjustment values ​​of the weights, biases, learning rates, control strategies, regularization parameters or other important parameters in the local intelligent driving model. The server side is a server system located in the cloud or a remote computing center, which is responsible for centrally processing and storing data uploaded from multiple vehicle terminals, and maintaining, updating and optimizing the global intelligent driving model.

[0031] Through the implementation of step 102, the server side can integrate driving data from different vehicles, perform data analysis and model optimization, and provide a more intelligent global intelligent driving model for all vehicles; through this information sharing, the overall effect of intelligent assisted driving can be improved.

[0032] Step 103, performing vehicle assisted driving according to the target global intelligent driving model corresponding to the driving style label issued by the server, wherein the target global intelligent driving model is a model obtained by updating the parameters of the initial global intelligent driving model corresponding to the driving style label by the server according to the update information;

[0033] Specifically, the target global intelligent driving model is a model obtained by adjusting and optimizing the initial global intelligent driving model based on the local model update information from each target vehicle terminal and the corresponding driving style labels. This model is a global intelligent driving system model optimized for specific driving styles (such as "aggressive driving", "smooth driving", etc.), which can adapt to a variety of driving styles and provide corresponding assisted driving decisions. The initial global intelligent driving model is a global intelligent driving model pre-set by the server, which can be constructed based on a large amount of historical data, traffic rules, driving behavior patterns and other information. This initial model is universal and suitable for most drivers and driving environments, but in actual applications it may not be able to fully adapt to the personalized needs of individual drivers, so it needs to be continuously optimized based on real-time data from each target vehicle terminal (such as local model update information and driving style labels).

[0034] Through the implementation of step 103, the target global intelligent driving model is issued according to the driving style label, which can assist driving based on an optimized global model. The global model can combine the data and optimization results of all vehicles to achieve a more intelligent driving strategy.

[0035] In summary, the embodiment of the present application generates exclusive driving style labels (such as energy-saving driving, comfortable driving, etc.) for each driver through the analysis of vehicle driving data, and optimizes the intelligent driving model locally to make it more in line with individual driving behavior and preferences; only the model update information and driving style labels are uploaded to the server, avoiding the transmission of original data and effectively reducing the risk of privacy leakage; the local terminal is responsible for local model training and updating, and only uploads incremental information, while the server optimizes and integrates the global model, significantly reducing the amount of data transmission and the server computing burden, while forming a fast and efficient model optimization iteration cycle to ensure that the intelligent driving model can quickly adapt to dynamic needs; the optimized target global model is sent to the vehicle terminal through the cloud, and is used for assisted driving in real time, effectively improving driving safety, comfort and energy saving. In summary, the vehicle assisted driving method provided by the embodiment of the present application comprehensively improves the adaptability, safety, comfort and energy saving of the intelligent driving model through personalized label drive, privacy protection, distributed incremental optimization and fast iteration cycle.

[0036] In some embodiments, the aforementioned step 101 may include: determining the driving style label corresponding to the target vehicle terminal based on the aforementioned vehicle driving data or the driver's label selection; based on the vehicle driving data, using a heuristic algorithm to update the model parameters of the local intelligent driving model, wherein the heuristic algorithm may include a genetic algorithm and a simulated annealing algorithm.

[0037] Specifically, heuristic algorithms are a type of algorithms that solve problems by using empirical rules or heuristic methods. These algorithms do not guarantee to find the optimal solution, but can find an approximate optimal solution or a more satisfactory solution within a reasonable time; in the embodiment of the present application, the heuristic algorithm will optimize the control strategy, weights and other important parameters of the local intelligent driving model based on the existing vehicle data and the target driving style label, so that the model can better adapt to the current driving style. Genetic algorithm is an optimization algorithm that simulates the natural selection and biological evolution process. It generates new solutions from a set of initial solutions through operations such as selection, crossover, and mutation, and continuously iterates and optimizes to eventually find a solution close to the optimal solution; simulated annealing algorithm is an optimization algorithm that simulates the physical annealing process. By simulating the state changes of the substance during the cooling process, the algorithm allows large changes in the initial stage (to avoid local optimal solutions), and gradually converges to the global optimal solution or approximate optimal solution as the search process proceeds.

[0038] Through the implementation of the above embodiments, the accuracy of label generation and the targetedness of model adjustment are improved by combining driving data and labels selected by the driver. Heuristic algorithms (such as genetic algorithms and simulated annealing algorithms) are introduced to effectively capture the nonlinear characteristics in complex driving scenarios, improve the accuracy of model updates, and better adapt to complex driving environments and diversified driving needs.

[0039] In some embodiments, after updating the model parameters of the local intelligent driving model using a heuristic algorithm based on the aforementioned vehicle driving data, the vehicle assisted driving method may also include: updating the model parameters of the local intelligent driving model in a gradient update manner based on the driver's preference.

[0040] Specifically, preference selection refers to the selection of specific driving strategies or characteristics based on the driver's personal preferences and needs; the driver's preference selection may involve different aspects such as acceleration smoothness, braking comfort, speed limit and fuel efficiency. Gradient update is an optimization technique based on the gradient descent method. In gradient update, the gradient of the model parameters (that is, the partial derivative of the loss function with respect to each parameter) can be calculated and the parameters can be updated in the opposite direction of the gradient to minimize the loss function (or error) and optimize the parameters of the local intelligent driving model. For example, if the driver has a high preference for acceleration smoothness, the acceleration and braking control model parameters of the vehicle can be adjusted according to the gradient update algorithm to make it smoother and more comfortable.

[0041] Through the implementation of the above embodiments, combined with the driver's real-time preferences and dynamic choices, more refined model adjustments are achieved, enabling the model to adaptively adjust under different driving conditions, further improving the level of personalization and intelligence; the global search capability of the heuristic algorithm is combined with the local optimization capability of the gradient update, thereby improving the convergence efficiency and accuracy of the model.

[0042] In some embodiments, the aforementioned vehicle assisted driving method may further include: uploading a first data amount of vehicle driving data to a server side; the target global intelligent driving model is a model obtained by the server side updating the parameters of the initial global intelligent driving model corresponding to the driving style label based on the update information and the model parameter weight, wherein the model parameter weight is calculated based on the first data amount and the second data amount, and the second data amount is the data amount of the training data of the initial global intelligent driving model.

[0043] Specifically, the first data volume refers to the data volume related to vehicle driving data, specifically, the total amount or sample amount of driving data collected during the driving process of the vehicle. The first data volume may include a series of driving data such as vehicle speed, acceleration, braking force, steering angle, etc. The second data volume refers to the amount of data used in the training process of the initial global intelligent driving model, that is, the amount of data used to train the initial global intelligent driving model. The model parameter weight refers to the weight coefficient calculated by the server based on the size of the first data volume and the second data volume uploaded when updating the global intelligent driving model. The model parameter weight determines the degree of influence of the update information uploaded by the vehicle on the adjustment of the global intelligent driving model during the model parameter update process; generally, the larger the model parameter weight coefficient, the more important the update information uploaded by the vehicle is in the model update.

[0044] Through the implementation of the above embodiments, weights are calculated for the first data amount and the training data amount, ensuring that the contribution of different terminals to the global model optimization matches their data amounts, avoiding excessive influence of small sample terminals on the global model optimization; the influence of data amount is comprehensively considered using model parameter weights, making the generated target global model more applicable in diversified driving scenarios; the server side makes full use of uploaded data when calculating weights and updating models, achieving more efficient resource allocation and calculation.

[0045] In some embodiments, the aforementioned vehicle assisted driving method may further include: locally encrypting and storing the vehicle driving data, and setting read and write permissions to prohibit uploading to the server.

[0046] Specifically, local encrypted storage means encrypting the vehicle driving data in the target vehicle terminal and storing the encrypted data in a local device (such as a computer, sensor or storage medium in the vehicle) to protect the privacy and security of the data and ensure that the data content is not read by unauthorized personnel or systems during the storage process; and applying access control policies to the stored vehicle driving data on the target vehicle terminal to prohibit uploading this data to the server.

[0047] Through the implementation of the above embodiments, the leakage of sensitive data is completely avoided, the driver's privacy is ensured to be not violated, and the user's trust and acceptance of intelligent driving technology is improved.

[0048] In some embodiments, the aforementioned vehicle driving data may include vehicle dynamic behavior data, driver concentration level and vehicle type.

[0049] Specifically, vehicle dynamic behavior data refers to data that reflects the dynamic performance of the vehicle during driving, including the vehicle's acceleration, braking, steering and other behaviors, as well as the distance to the front and rear vehicles, the driving route, etc.; the driver's concentration level refers to whether the driver is focused during driving, and the degree of concentration. The driver's eyes, head movements, facial expressions, etc. can be monitored through sensors or cameras in the vehicle (such as facial recognition, eye tracking technology, etc.), or the driver's attention state can be inferred by analyzing driving behavior (such as reaction time, operation accuracy, etc.); vehicle type refers to the basic classification or type of vehicle, which is usually divided according to the vehicle's size, purpose, power system, drive mode and other attributes; for example, vehicle types can be divided into sedans, SUVs, trucks, electric vehicles and hybrid vehicles.

[0050] It should be noted that different vehicle types will affect their driving characteristics and driving behavior; for example, SUVs may be more likely to roll over when making sharp turns due to their larger body and higher center of gravity, while sedans are more suitable for high-speed driving on urban roads; by identifying the vehicle type, the intelligent driving model and driving assistance strategies can be optimized according to the characteristics of the vehicle, such as adjusting acceleration, braking, and steering response, etc.; in addition, the vehicle type can also affect the driver's operating method, so adjusting the system's response method according to the vehicle type can improve the overall driving experience and safety.

[0051] Through the implementation of the above embodiments, the combination of multi-dimensional data sources improves the generation accuracy of driving style labels, enables the model to more accurately reflect the driver's real behavior, and improves the adaptability of the intelligent driving model in complex environments and diversified driving scenarios.

[0052] Furthermore, as an implementation of the aforementioned method embodiment, the present application also provides a vehicle assisted driving device for implementing the aforementioned method embodiment. The device embodiment corresponds to the aforementioned method embodiment. For ease of reading, the present vehicle assisted driving device embodiment will no longer repeat the details of the aforementioned method embodiment one by one, but it should be clear that the device in the embodiment of the present application can correspond to and implement all the contents of the aforementioned method embodiment. Figure 2 As shown, the vehicle assisted driving device 20 includes: a local parameter updating unit 201, a data uploading unit 202 and an assisted driving unit 203, wherein the local parameter updating unit 201 is used to determine the driving style label corresponding to the target vehicle terminal based on the vehicle driving data, and update the model parameters of the local intelligent driving model; the data uploading unit 202 is used to upload the update information of the model parameters and the driving style label to the server side; the assisted driving unit 203 is used to perform vehicle assisted driving according to the target global intelligent driving model corresponding to the driving style label issued by the server side, wherein the target global intelligent driving model is a model obtained by updating the parameters of the initial global intelligent driving model corresponding to the driving style label according to the update information.

[0053] In some embodiments, the local parameter update unit 201 is also used to determine the driving style label corresponding to the target vehicle terminal based on vehicle driving data or the driver's label selection; based on the vehicle driving data, the model parameters of the local intelligent driving model are updated using a heuristic algorithm, wherein the heuristic algorithm includes a genetic algorithm and a simulated annealing algorithm.

[0054] In some embodiments, the local parameter updating unit 201 is also used to update the model parameters of the local intelligent driving model in a gradient updating manner based on the driver's preference selection.

[0055] In some embodiments, the data upload unit 202 is also used to upload a first data amount of vehicle driving data to the server side; the target global intelligent driving model is a model obtained by the server side updating the parameters of the initial global intelligent driving model corresponding to the driving style label based on the update information and the model parameter weight, wherein the model parameter weight is calculated based on the first data amount and the second data amount, and the second data amount is the data amount of the training data of the initial global intelligent driving model.

[0056] In some embodiments, the vehicle assisted driving device 20 also includes a local management unit for locally encrypting and storing the vehicle driving data and setting read and write permissions to prohibit uploading to the server.

[0057] In some embodiments, the aforementioned vehicle driving data includes vehicle dynamic behavior data, driver concentration level and vehicle type.

[0058] The present application also provides a computer-readable storage medium, which stores computer-executable instructions or computer programs. When the computer-executable instructions or computer programs are executed by a processor, the processor will be caused to execute any step of the vehicle assisted driving method provided in the present application.

[0059] In some embodiments, the computer-readable storage medium may be a memory such as RAM, read-only memory (ROM), flash memory, magnetic surface memory, optical disk, or compact disc read-only memory (CD-ROM); or it may be various devices including one or any combination of the above memories.

[0060] In some embodiments, computer executable instructions may be in the form of a program, software, software module, script or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and may be deployed in any form, including as a stand-alone program or as a module, component, subroutine or other unit suitable for use in a computing environment.

[0061] In some embodiments, computer-executable instructions may, but need not, correspond to a file in a file system, may be stored as part of a file storing other programs or data, for example, in one or more scripts in a HyperText Markup Language (HTML) document, in a single file dedicated to the program in question, or in multiple coordinated files (e.g., files storing one or more modules, subroutines, or code portions).

[0062] In some embodiments, computer executable instructions may be deployed to be executed on one electronic device, or on multiple electronic devices located at one site, or on multiple electronic devices distributed at multiple sites and interconnected by a communication network.

[0063] like Figure 3 As shown, the present application also provides an electronic device 30, including a memory 310, a processor 320, and a computer program 311 stored in the memory 310 and executable on the processor, and when the processor 320 executes the computer program 311, any step of the above-mentioned vehicle assisted driving method is implemented.

[0064] The present application also provides a computer program product, which includes a computer program or a computer executable instruction, and the computer program or the computer executable instruction is stored in a computer-readable storage medium. The processor of the electronic device reads the computer program or the computer executable instruction from the computer-readable storage medium, and the processor executes the computer program or the computer executable instruction, so that the electronic device performs any step of the vehicle assisted driving method described above in the present application.

[0065] The present application also provides a vehicle assisted driving system, including a vehicle terminal, used to execute the vehicle assisted driving method provided in the embodiment of the present application; a server side, used to receive update information and driving style labels of model parameters from the vehicle terminal, and send the target global intelligent driving model generated according to the update information and driving style labels to the vehicle terminal.

[0066] In some embodiments, the server side can also update the parameters of the initial global intelligent driving model corresponding to the driving style label according to the update information and the model parameter weight to obtain the target global intelligent driving model, wherein the model parameter weight is calculated based on the first data amount and the second data amount, and the second data amount is the data amount of the training data of the initial global intelligent driving model.

[0067] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A vehicle assisted driving method, applied to a target vehicle terminal, characterized in that: include: Based on the vehicle driving data, determine the driving style label corresponding to the target vehicle terminal, and update the model parameters of the local intelligent driving model; Uploading the updated information of the model parameters and the driving style label to a server; Vehicle assisted driving is performed according to the target global intelligent driving model corresponding to the driving style label issued by the server, wherein the target global intelligent driving model is a model obtained by the server by updating the parameters of the initial global intelligent driving model corresponding to the driving style label according to the update information.

2. The vehicle assisted driving method according to claim 1, characterized in that: The determining, based on the vehicle driving data, a driving style label corresponding to the target vehicle terminal, and updating model parameters of the local intelligent driving model includes: Determining a driving style label corresponding to the target vehicle terminal based on the vehicle driving data or the driver's label selection; Based on the vehicle driving data, a heuristic algorithm is used to update the model parameters of the local intelligent driving model, wherein the heuristic algorithm includes a genetic algorithm and a simulated annealing algorithm.

3. The vehicle assisted driving method according to claim 2, characterized in that: After updating the model parameters of the local intelligent driving model using a heuristic algorithm based on the vehicle driving data, the vehicle assisted driving method further includes: Based on the driver's preference, the model parameters of the local intelligent driving model are updated in a gradient update manner.

4. The vehicle assisted driving method according to claim 1, characterized in that: The vehicle assisted driving method further includes: Uploading a first amount of data of the vehicle driving data to the server; The target global intelligent driving model is a model obtained by the server side updating the parameters of the initial global intelligent driving model corresponding to the driving style label according to the update information and the model parameter weight, wherein the model parameter weight is calculated based on the first data amount and the second data amount, and the second data amount is the data amount of the training data of the initial global intelligent driving model.

5. The vehicle assisted driving method according to claim 1, characterized in that: The vehicle assisted driving method further includes: The vehicle driving data is locally encrypted and stored, and read and write permissions are set to prohibit uploading to the server.

6. The vehicle assisted driving method according to any one of claims 1 to 5, characterized in that: The vehicle driving data includes vehicle dynamic behavior data, driver concentration level and vehicle type.

7. A vehicle driving assistance device, characterized in that: include: A local parameter updating unit, configured to determine a driving style label corresponding to the target vehicle terminal based on vehicle driving data, and to update model parameters of a local intelligent driving model; A data uploading unit, used for uploading the update information of the model parameters and the driving style label to a server; An assisted driving unit is used to perform vehicle assisted driving according to a target global intelligent driving model corresponding to the driving style label issued by the server, wherein the target global intelligent driving model is a model obtained by the server by updating the parameters of an initial global intelligent driving model corresponding to the driving style label according to the update information.

8. A vehicle assisted driving system, characterized in that: include: A vehicle terminal, configured to execute the vehicle assisted driving method according to any one of claims 1 to 6; The server side is used to receive update information of model parameters and driving style labels from the vehicle terminal, and send the target global intelligent driving model generated according to the update information and the driving style label to the vehicle terminal.

9. An electronic device, comprising: A memory and a processor, characterized in that the processor is used to implement the steps of the vehicle assisted driving method as described in any one of claims 1 to 6 when executing the computer program stored in the memory.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the vehicle assisted driving method as described in any one of claims 1 to 6 are implemented.

Images