A cruise hierarchical control method, system and electronic device based on driving behavior

Abstract

The present application provides a cruise hierarchical control method, system and electronic device based on driving behavior. In this method, an actual driving data set is obtained under different driving scenarios, and then a simulated driving data set is obtained according to the actual driving data set. Then, the simulated driving data is classified into driving behavior levels to obtain a simulated driving behavior classification data set. Then, the simulated driving behavior classification data set is imported into a neural network model to obtain groups of control parameters corresponding to different simulated driving behaviors. Finally, groups of control parameters corresponding to different simulated driving behaviors are output. Therefore, through the above method, the collection of actual driving behavior data can be reduced. By generating simulated driving behavior data, the sample size of driving behavior data and the training samples of the neural network model are increased, making the classification of driving behavior levels more accurate and improving the generalization ability of the clustering method.

Description

Technical Field

[0001] The present invention relates to the technical field of automotive autonomous driving, and particularly to a cruise hierarchical control method, system and electronic device based on driving behavior. Background Art

[0002] With the development of communication and automation technologies, more and more intelligent devices appear in life. With the support of 5G network, autonomous driving technology is becoming more and more mature.

[0003] Currently, the Adaptive Cruise Control (ACC) is applied to the new generation of automotive driver assistance systems. The ACC combines the traditional Cruise Control System (CCS) and the Front Collision Warning System (FCWS) to achieve automatic obstacle avoidance and automatic cruise control of the vehicle.

[0004] The traditional ACC system does not consider the driving habits and styles of different drivers, and uniformly configures control parameters according to the vehicle's own situation, resulting in poor driver experience.

[0005] However, if control parameters are configured separately according to the driving habits and styles of each driver, there are too many types of parameters that need to be calibrated for the ACC, which is difficult to implement in actual implementation, and the existing automotive intelligent system cannot fully upload the driving behavior-related data of all vehicles to the cloud for clustering analysis of driving behavior. Summary of the Invention

[0006] The present application provides a cruise hierarchical control method, system and electronic device based on driving behavior, which are used to generate simulated driving data, classify and train the simulated driving data, and output groups of control parameters corresponding to different simulated driving behaviors according to different driving behavior levels. The specific technical solutions are as follows:

[0007] In a first aspect, the present application provides a cruise hierarchical control method based on driving behavior, including:

[0008] Obtain an actual driving data set, where the actual driving data set includes data corresponding to each driving behavior in different scenarios;

[0009] Obtain a simulated driving data set according to the actual driving data set;

[0010] Perform driving behavior level division on the simulated driving data set to obtain a simulated driving behavior hierarchical data set;

[0011] Importing the simulated driving behavior classification data set into a neural network model to obtain groups of control parameters corresponding to different simulated driving behaviors;

[0012] Output the control parameters corresponding to different simulated driving behaviors.

[0013] Based on the above method, simulated driving data is generated according to actual driving data, which solves the problem of limited collection of actual driving behavior data; by generating simulated driving behavior data, the sample size of driving behavior data is increased, and the training samples of the neural network model are also increased, making the driving behavior level division more accurate and solving the problem of weak generalization ability of the clustering analysis method; the various hierarchical control parameters tested in the actual test environment are output based on the divided driving behavior level, which is more suitable for the public's driving behavior habits and meets the safety requirements of automobile driving.

[0014] In a possible design, after outputting the control parameters corresponding to different simulated driving behaviors, the following is further included:

[0015] Acquire driving behavior data of the current driver, and determine the driving behavior category of the current driver based on the driving behavior data, wherein the driving behavior data at least includes the distance between the vehicle and the obstacle, the vehicle speed, the brake, and the throttle information;

[0016] If the driving behavior category is determined to be a gentle driving behavior, outputting a first set of control parameters corresponding to the gentle driving behavior;

[0017] If the driving behavior category is determined to be a comfortable driving behavior, outputting a second set of control parameters corresponding to the comfortable driving behavior;

[0018] If the driving behavior category is determined to be an aggressive driving behavior, a third set of control parameters corresponding to the aggressive driving behavior is output.

[0019] Through the above-mentioned method, a set of control parameters corresponding to the driving behavior can be output according to the current driving behavior of the driver, and the set of control parameters is derived from actual tests, thereby ensuring the safety of vehicle driving.

[0020] In a possible design, obtaining a simulated driving data set according to the actual driving data set includes:

[0021] According to the data features of each actual driving data in the actual driving data set and the preset simulation data generation rules, simulated driving data are obtained;

[0022] Combining the simulated driving data with the actual driving data to obtain a first driving data set;

[0023] Calculate the probability that the data in the first driving data set comes from actual driving data;

[0024] According to the probability value, calculate the loss functions of the simulated driving behavior generation model and the simulated driving behavior discrimination model;

[0025] According to the loss functions of the simulated driving behavior generation model and the simulated driving behavior discrimination model, use the neural network algorithm to update the simulated driving data to obtain a simulated driving data set.

[0026] In the above way, simulated driving data is generated according to the characteristics of actual driving data, expanding the training samples of driving data.

[0027] In a possible design, perform driving behavior level classification on the simulated driving data set to obtain a simulated driving behavior classification data set, including:

[0028] Classify the data in the simulated driving data set according to the similarity between each data in the simulated driving data set. After n classifications, n types of simulated driving data are obtained, where n is an integer greater than 2;

[0029] According to the numerical intervals of the n types of simulated driving data, perform driving behavior level classification on the n types of simulated driving data to obtain a simulated driving behavior classification data set.

[0030] In the above way, the simulated driving data is increased, the accuracy of classifying the simulated driving data is improved, and at the same time, the generalization ability of the clustering analysis method is improved.

[0031] In a second aspect, the present application provides a hierarchical control system based on driving behavior, including:

[0032] An acquisition module for acquiring an actual driving data set, where the actual driving data set includes data corresponding to each driving behavior in different scenarios;

[0033] A generation module for obtaining a simulated driving data set according to the actual driving data set;

[0034] A classification module for performing driving behavior level classification on the simulated driving data set according to the similarity between each data in the simulated driving data set based on the similarity between each data to obtain a simulated driving behavior classification data set;

[0035] A network model training module for importing the simulated driving behavior classification data set into a neural network model to obtain groups of control parameters corresponding to different simulated driving behaviors;

[0036] An output module for outputting groups of control parameters corresponding to different simulated driving behaviors.

[0037] In a possible design, the obtaining module is further configured to obtain driving behavior data of the current driver;

[0038] The output module determines the driving behavior category of the current driver based on the driving behavior data, where the driving behavior data at least includes the distance between the vehicle and an obstacle, vehicle speed, braking, and throttle information;

[0039] If the driving behavior category is determined to be a gentle driving behavior, output the first set of control parameters corresponding to the gentle driving behavior;

[0040] If the driving behavior category is determined to be a comfortable driving behavior, output the second set of control parameters corresponding to the comfortable driving behavior;

[0041] If the driving behavior category is determined to be an aggressive driving behavior, output the third set of control parameters corresponding to the aggressive driving behavior.

[0042] In a possible design, the generating module is specifically configured to obtain simulated driving data according to the data characteristics of each actual driving data in the actual driving data set and a preset simulated data generation rule;

[0043] Combine the simulated driving data with the actual driving data to obtain a first driving data set;

[0044] Calculate the probability that the data in the first driving data set comes from the actual driving data;

[0045] According to the probability value, calculate the loss functions of the simulated driving behavior generation model and the simulated driving behavior discrimination model;

[0046] According to the loss functions of the simulated driving behavior generation model and the simulated driving behavior discrimination model, use a neural network algorithm to update the simulated driving data to obtain a simulated driving data set.

[0047] In a possible design, the classification module is specifically configured to classify the data in the simulated driving data set according to the similarity between the data. After n classifications, n types of simulated driving data are obtained, where n is an integer greater than 2;

[0048] According to the numerical intervals of the n types of simulated driving data, perform driving behavior level division on the n types of simulated driving data to obtain a simulated driving behavior classification data set.

[0049] In a third aspect, the present application provides an electronic device, including:

[0050] A memory for storing a computer program;

[0051] A processor, when executing the computer program stored on the memory, implements the method steps of the above-mentioned cruise hierarchical control based on driving behavior.

[0052] In a fourth aspect, the present application provides a computer-readable storage medium, in which a computer program is stored, and when the computer program is executed by a processor, the method steps of the above-mentioned cruise hierarchical control based on driving behavior are implemented.

[0053] For the various aspects in the second to fourth aspects above and the possible technical effects that each aspect may achieve, please refer to the description of the possible technical effects that can be achieved for the first aspect or various possible solutions in the first aspect above, and will not be repeated here. Description of the Drawings

[0054] Figure 1 It is a flowchart of a method for cruise hierarchical control based on driving behavior provided by the present application;

[0055] Figure 2 It is a schematic diagram of the architecture of a cruise hierarchical control system based on driving behavior provided by the present application;

[0056] Figure 3 It is a schematic diagram of the structure of a cruise hierarchical control system based on driving behavior provided by the present application;

[0057] Figure 4 It is a schematic diagram of the structure of an electronic device provided by the present application. Detailed Embodiments

[0058] In order to make the objectives, technical solutions, and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings. The specific operation methods in the method embodiments can also be applied to the device embodiments or system embodiments. It should be noted that in the description of the present application, "a plurality of" is understood as "at least two". "And / or" describes the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. The connection between A and B can represent: A is directly connected to B and A is connected to B through C. In addition, in the description of the present application, terms such as "first" and "second" are only used for the purpose of distinguishing descriptions, and cannot be understood as indicating or implying relative importance, nor can they be understood as indicating or implying an order.

[0059] The embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0060] With the development of network technology and the increasing maturity of autonomous driving technology, more and more intelligent devices and advanced control strategies are applied in the autonomous driving system of automobiles, making automobile autonomous driving more user-friendly. At present, the traditional adaptive cruise control system does not consider the driving habits and styles of different drivers, and uniformly configures control parameters according to the situation of the vehicle itself, resulting in a poor driving experience for drivers. However, if control parameters are configured separately according to the driving habits and styles of each driver, the types of parameters that need to be calibrated in the adaptive cruise control system are too many, and the driving habits and styles of actual drivers need to be collected, but the amount of collected data is limited. If only a small amount of data on driving habits and styles is used for the classification of different driving behaviors, the classification results of different driving behaviors obtained cannot accurately reflect the driving habits and styles of all drivers.

[0061] Therefore, the present application provides a cruise hierarchical control method based on driving behavior. In this method, first, the driving scenarios of vehicle automatic cruise are set, then the actual driving data set is obtained under different driving scenarios, and then, by comparing the characteristics of each data in the actual driving data set, a simulated driving data set is obtained. Then, the simulated driving data is classified into driving behavior levels to obtain a simulated driving behavior classification data set. Then, the simulated driving behavior classification data set is imported into a neural network model to obtain groups of control parameters corresponding to different simulated driving behaviors, and finally, groups of control parameters corresponding to different simulated driving behaviors are output.

[0062] Therefore, through the above method, the collection of actual driving behavior data can be reduced. By generating simulated driving behavior data, the sample size of driving behavior data is increased, and at the same time, the training samples of the neural network model are also increased, making the classification of driving behavior levels more accurate and improving the generalization ability of the clustering method.

[0063] Refer to Figure 1 The following shows a flowchart of a cruise hierarchical control method based on driving behavior provided by an embodiment of the present application. The method includes:

[0064] S1, obtain the actual driving data set;

[0065] First of all, the method provided by the present application can be applied to Figure 2In the shown system architecture, the system architecture includes a driving behavior data generation module, a clustering analysis module for driving behavior data, a training module for a neural network model, a driving classification control unit, a body control module (BCM), an engine control system (EMS), an electronic stability control system (ESC), and a motor control unit (MCU).

[0066] In actual application, the actual driving data of the driver respectively go through the generation of driving behavior data, the clustering analysis of driving behavior data, and the training module of the neural network model, and then obtain the groups of control parameters corresponding to different simulated driving behaviors. The driving classification control unit combines with the ACC system. After receiving the desired torque command, according to the currently determined driving behavior, it performs task allocation in the EMS, ESC, and MCU modules and inputs the corresponding control parameters to each module. The control parameters are derived from the actual test parameters of the vehicle and meet the safety requirements, and finally the vehicle realizes automatic cruise operation according to different driving behaviors.

[0067] In order to obtain simulated driving data, it is necessary to obtain actual driving data.

[0068] In the embodiment of this application, in order to ensure that the vehicle's automatic cruise can adapt to various scenarios and drive safely, it is necessary to obtain the actual driving data in each scenario. For example, two vehicles are driving in a row on a road. The following vehicle turns on the automatic cruise function and follows the leading vehicle at a certain distance. When the leading vehicle brakes, according to the driving habits of the driver of the following vehicle, the following vehicle should also brake to ensure a safe driving distance. At this time, the automatic cruise state of the vehicle changes from the following state to the braking state.

[0069] Therefore, it is necessary to obtain the actual driving data sets in different driving scenarios. Among them, different driving scenarios include but are not limited to acceleration, braking, lane change, and following scenarios. Here, 3 groups of actual driving data in different scenarios are obtained, and the obtained actual driving data sets can be shown in Table 1:

[0070]

[0071] Table 1

[0072] Through the above method, the actual driving data in each scenario can be obtained. Among them, the parameters in Table 1 include but are not limited to the current obstacles, the state information of the vehicle, braking, and throttle action information.

[0073] S2. Obtain a simulated driving dataset according to the actual driving dataset;

[0074] In the embodiment of the present application, the generation network randomly inputs noise data based on the actual driving data and generates simulated driving data corresponding to the characteristics of the actual driving data.

[0075] Combine the simulated driving data with the actual driving data to obtain a first driving dataset, and input the data in the first driving dataset into the discrimination network. The discrimination network outputs the probability value that the data in the first driving dataset comes from the actual driving dataset. In order to make the output result of the discrimination network more accurate, the Adam algorithm can also be used to optimize the discrimination network.

[0076] According to the probability value output by the discrimination network, calculate the loss functions of the simulated driving behavior generation network and the simulated driving behavior discrimination network. The value output by the loss function can be used to determine the similarity between the simulated driving data and the actual driving data. Based on the value output by the loss function, use the neural network algorithm to update the simulated driving data so that the simulated driving data has exactly the same data characteristics as the actual driving data.

[0077] Therefore, with the generated simulated driving data, on the premise of ensuring the acquisition of the data characteristics of the actual driving data, the amount of actual driving data is expanded, and finally a simulated driving dataset is obtained. Suppose n groups of simulated driving data are generated, where n is an integer greater than 2. In actual applications, tens of thousands of groups of simulated driving data will be generated. Here, n is taken as 6, and the obtained simulated driving dataset can be shown in Table 2:

[0078]

[0079] Table 2

[0080] Through the above method, simulated driving data generated based on the actual driving data in various scenarios can be obtained, increasing the sample size of the driving data.

[0081] Here, it should be noted that: corresponding to Table 2, a group of simulated driving data generated here consists of simulation parameter 1, simulation parameter 2, simulation parameter 3, and simulation parameter 4.

[0082] S3. Classify the driving behavior levels of the simulated driving dataset to obtain a simulated driving behavior classification dataset;

[0083] In the embodiment of the present application, in order to meet the different driving habits and styles of drivers, after obtaining the simulated driving dataset, it is necessary to perform clustering analysis on the data in the simulated driving dataset according to the similarity of each data in the simulated driving dataset. In order to ensure the classification effect and efficiency, the Parameter-free strategy is adopted in the clustering algorithm. The specific steps are as follows:

[0084] First, it is necessary to calculate the nearest neighbor corresponding to each data in each simulated driving dataset. Among them, the nearest neighbor is the shortest distance from the position where each driving data in the simulated driving dataset is distributed to the current data.

[0085] After obtaining the nearest neighbor of each data, calculate the adjacency matrix according to Formula 2.

[0086]

[0087] Among them, represents the nearest neighbor point of the i-th point; is the adjacency matrix of the data;

[0088] According to the calculation method of the adjacency matrix in Formula 2, it can be obtained that as long as the following conditions are met, the value in the adjacency matrix is 1:

[0089] : For the i-th point, the nearest neighbor of this point is the j-th point.

[0090] : For the i-th point, this point is the nearest neighbor of the j-th point.

[0091] : The nearest neighbor point of the i-th point and the nearest neighbor point of the j-th point are the same.

[0092] Then, obtain a directed graph according to the adjacency matrix to complete the first clustering analysis of the samples.

[0093] After n times of clustering analysis, a three-class simulated driving behavior classification dataset as shown in Table 3 can be obtained.

[0094] Here, it should be noted that: the sample size of the simulated driving behavior data remains unchanged. Therefore, in each scenario, each driving behavior category corresponds to multiple parameters, which are represented by the parameter set m here, and m is an integer from 1 to 12.

[0095]

[0096] Table 3

[0097] As shown in Table 3, according to the obtained simulated driving data in different scenarios, after clustering analysis, the parameter sets m corresponding to each driving behavior in different scenarios are obtained.

[0098] S4. Import the simulated driving behavior classification dataset into the neural network model to obtain each group of control parameters corresponding to different simulated driving behaviors;

[0099] In the embodiment of the present application, in order to obtain a grading data table more suitable for the driving behaviors of the general public, it is necessary to import each parameter set m corresponding to each driving behavior in different scenarios into a fully connected neural network model for training. The neural network model inputs obstacle distance, vehicle speed, braking, and throttle information. After the training is completed, softmax is used as the activation function, and the control parameter sets corresponding to each simulated driving behavior in each scenario are output as shown in Table 4.

[0100]

[0101] Table 4

[0102] As shown in Table 4, each parameter set corresponding to each driving behavior in different scenarios is imported into a fully connected neural network model for training to obtain a data table more suitable for the driving behaviors of the general public.

[0103] Here, it should be noted that: corresponding to Table 4, the control parameter set under the gentle driving behavior output here refers to the parameter set composed of parameter Z1, parameter Z2, parameter Z3, and parameter Z4.

[0104] S5. Output each group of control parameters corresponding to different simulated driving behaviors.

[0105] In the embodiment of the present application, in order to enable the vehicle to perform automatic grading cruise according to the driving behaviors of different drivers. Before outputting each group of control parameters corresponding to different simulated driving behaviors, the driving behavior data of the current driver is obtained, and the driving behavior category of the current driver is determined based on the driving behavior data.

[0106] If the current driving behavior category is determined to be a gentle driving behavior, as shown in Table 4, a group of control parameters composed of parameter Z1, parameter Z2, parameter Z3, and parameter Z4 corresponding to the gentle driving behavior are output.

[0107] If the current driving behavior category is determined to be a comfortable driving behavior, as shown in Table 4, a group of control parameters composed of parameter Z5, parameter Z6, parameter Z7, and parameter Z8 corresponding to the comfortable driving behavior are output.

[0108] If the current driving behavior category is determined to be an aggressive driving behavior, as shown in Table 4, a group of control parameters composed of parameter Z9, parameter Z10, parameter Z11, and parameter Z12 corresponding to the aggressive driving behavior are output.

[0109] Furthermore, the cruise control strategy based on driving behavior runs on the driving grading control unit. The driving grading control unit receives the switch quantity enabling signal of the BCM module through the CAN bus to determine whether to enable the cruise grading control function based on driving behavior.

[0110] As a braking execution unit, when the driving classification control unit receives a braking request, the ESC outputs corresponding control signals according to the received control parameters from the driving classification control unit and performs braking operations according to the driving behavior level.

[0111] As power output units, the EMS and MCU, when the driving classification control unit issues a power request, will also output torques according to the control parameters from the driving classification control unit and according to the driving behavior level.

[0112] In addition to receiving the above control signals, the driving classification control unit also receives driver intention information and power component information of the vehicle itself, and controls the vehicle to drive safely in line with the driving behavior category. The driver intention information refers to the operations of the driver, such as throttle, brake, and steering actions.

[0113] In terms of the above technical solution, the method provided by this application first generates simulated driving data according to actual driving data, solving the problem of limited collection of actual driving behavior data; by generating simulated driving behavior data, the sample size of driving behavior data is increased, and at the same time, the training samples of the neural network model are also increased, making the classification of driving behavior levels more accurate and solving the problem of weak generalization ability of the clustering analysis method; based on the divided driving behavior levels, various classification control parameters tested in the actual test environment are output, which are more suitable for the driving behavior habits of the public and meet the safety requirements of vehicle driving.

[0114] Based on the method provided in the above embodiments, an adaptive cruise control system based on driving behavior is also provided in the embodiments of this application, as Figure 3 shown in the structural schematic diagram of an adaptive cruise control system based on driving behavior in the embodiments of this application. The system includes:

[0115] An acquisition module 301, configured to acquire an actual driving data set, where the actual driving data set includes data corresponding to each driving behavior in different scenarios;

[0116] A generation module 302, configured to obtain a simulated driving data set according to the actual driving data set;

[0117] A classification module 303, configured to perform driving behavior level classification on the simulated driving data set to obtain a simulated driving behavior classification data set;

[0118] A network model training module 304, configured to import the simulated driving behavior classification data set into a neural network model to obtain groups of control parameters corresponding to different simulated driving behaviors;

[0119] An output module 305, configured to output groups of control parameters corresponding to different simulated driving behaviors.

[0120] In a possible design, the obtaining module 301 is further configured to obtain the driving behavior data of the current driver;

[0121] The output module 305 is configured to determine the driving behavior category of the current driver based on the driving behavior data, where the driving behavior data at least includes the distance between the vehicle and the obstacle, the vehicle speed, the braking, and the throttle information;

[0122] If the driving behavior category is determined to be a gentle driving behavior, output the first set of control parameters corresponding to the gentle driving behavior;

[0123] If the driving behavior category is determined to be a comfortable driving behavior, output the second set of control parameters corresponding to the comfortable driving behavior;

[0124] If the driving behavior category is determined to be an aggressive driving behavior, output the third set of control parameters corresponding to the aggressive driving behavior.

[0125] In a possible design, the generating module 302 is specifically configured to obtain simulated driving data according to the data characteristics of each actual driving data in the actual driving data set and a preset simulated data generation rule;

[0126] Combine the simulated driving data with the actual driving data to obtain a first driving data set;

[0127] Calculate the probability that the data in the first driving data set comes from the actual driving data;

[0128] According to the probability value, calculate the loss functions of the simulated driving behavior generation model and the simulated driving behavior discrimination model;

[0129] According to the loss functions of the simulated driving behavior generation model and the simulated driving behavior discrimination model, use the neural network algorithm to update the simulated driving data to obtain a simulated driving data set.

[0130] In a possible design, the data classification module 303 is specifically configured to classify the data in the simulated driving data set according to the similarity between the data. After n classifications, n types of simulated driving data are obtained, where n is an integer greater than 2;

[0131] According to the numerical intervals of the n types of simulated driving data, perform driving behavior level division on the n types of simulated driving data to obtain a simulated driving behavior classification data set.

[0132] The output module 305 is configured to output each set of control parameters corresponding to different simulated driving behaviors.

[0133] Based on the same inventive concept, an electronic device is further provided in an embodiment of the present application. The electronic device can implement the functions of the foregoing cruise hierarchical control method based on driving behavior. Refer to Figure 4 , the electronic device includes:

[0134] At least one processor 401 and a memory 402 connected to the at least one processor 401. In the embodiments of the present application, the specific connection medium between the processor 401 and the memory 402 is not limited. Figure 4 In [reference], it is taken as an example that the processor 401 and the memory 402 are connected through a bus 400. The bus 400 is Figure 4 represented by a thick line in [reference]. The connection manners between other components are only for illustrative purposes and are not limited thereto. The bus 400 can be divided into an address bus, a data bus, a control bus, etc. For the convenience of representation, Figure 4 it is only represented by a thick line in [reference], but it does not mean that there is only one bus or one type of bus. Alternatively, the processor 401 can also be referred to as a controller, and the name is not limited.

[0135] In the embodiments of the present application, the memory 402 stores instructions executable by the at least one processor 401. By executing the instructions stored in the memory 402, the at least one processor 401 can execute the cruise hierarchical control method based on driving behavior described above. The processor 401 can implement Figure 3 the functions of each module in the system shown in [reference].

[0136] Among them, the processor 401 is the control center of the device. It can connect various parts of the entire control device through various interfaces and lines. By running or executing the instructions stored in the memory 402 and calling the data stored in the memory 402, various functions of the device and process data, so as to monitor the device as a whole.

[0137] In a possible design, the processor 401 may include one or more processing units. The processor 401 may integrate an application processor and a modem processor. Among them, the application processor mainly processes the operating system, user interface, application programs, etc., and the modem processor mainly processes wireless communication. It can be understood that the above modem processor may not be integrated into the processor 401. In some embodiments, the processor 401 and the memory 402 can be implemented on the same chip. In some embodiments, they can also be implemented on separate chips respectively.

[0138] The processor 401 may be a general-purpose processor, such as a central processing unit (CPU), a digital signal processor, an application specific integrated circuit, a field programmable gate array or other programmable logic device, discrete gate or transistor logic device, discrete hardware component, and may implement or execute the various methods, steps and logic block diagrams disclosed in the embodiments of the present application. The general-purpose processor may be a microprocessor or any conventional processor, etc. The steps of the cruise hierarchical control method based on driving behavior disclosed in combination with the embodiments of the present application may be directly embodied as being executed by a hardware processor, or executed by a combination of hardware and software modules in the processor.

[0139] The memory 402, as a non-volatile computer-readable storage medium, can be used to store non-volatile software programs, non-volatile computer-executable programs and modules. The memory 402 may include at least one type of storage medium, for example, it may include flash memory, hard disk, multimedia card, card-type memory, random access memory (RAM), static random access memory (SRAM), programmable read only memory (PROM), read only memory (ROM), electrically erasable programmable read-only memory (EEPROM), magnetic memory, magnetic disk, optical disc, etc. The memory 402 is any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory 402 in the embodiments of the present application may also be a circuit or any other device capable of implementing a storage function, for storing program instructions and / or data.

[0140] By designing and programming the processor 401, the code corresponding to the cruise hierarchical control based on driving behavior introduced in the foregoing embodiments can be solidified into the chip, so that the chip can execute Figure 1 the steps of the cruise hierarchical control method based on driving behavior shown in the embodiments. How to design and program the processor 401 is a well-known technology to those skilled in the art and will not be elaborated here.

[0141] Based on the same inventive concept, the embodiments of the present application also provide a storage medium storing computer instructions, which when run on a computer, cause the computer to execute the cruise hierarchical control method based on driving behavior described above.

[0142] In some possible embodiments, various aspects of the cruise hierarchical control method based on driving behavior provided by the present application can also be implemented in the form of a program product, which includes program code. When the program product runs on a device, the program code is used to cause the control device to execute the steps in the cruise hierarchical control method based on driving behavior according to various exemplary embodiments of the present application described above in this specification.

[0143] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0144] The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams, can be realized by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for realizing the functions specified in one Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0145] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device realizes the functions specified in one Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0146] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for realizing the functions specified in one Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0147] Obviously, those skilled in the art can make various changes and modifications to this application without departing from the spirit and scope of this application. Thus, if these modifications and variations of this application fall within the scope of the claims of this application and their equivalent technologies, this application is also intended to include these modifications and variations.

Claims

1. A cruise hierarchical control method based on driving behavior, characterized in that, Including: Obtain an actual driving dataset, where the actual driving dataset includes data corresponding to various driving behaviors in different scenarios; According to the actual driving dataset, obtain a simulated driving dataset, including: According to the data characteristics of each actual driving data in the actual driving dataset and a preset simulated data generation rule, obtain simulated driving data; Combine the simulated driving data with the actual driving data to obtain a first driving dataset; Calculate the probability that the data in the first driving dataset comes from actual driving data; According to the probability value, calculate the loss functions of the simulated driving behavior generation model and the simulated driving behavior discrimination model; According to the loss functions of the simulated driving behavior generation model and the simulated driving behavior discrimination model, use a neural network algorithm to update the simulated driving data to obtain a simulated driving dataset; Perform driving behavior level classification on the simulated driving dataset to obtain a simulated driving behavior classification dataset; Import the simulated driving behavior classification dataset into a neural network model to obtain groups of control parameters corresponding to different simulated driving behaviors; Output groups of control parameters corresponding to different simulated driving behaviors.

2. The method according to claim 1, characterized in that, After outputting groups of control parameters corresponding to different simulated driving behaviors, it further includes: Obtain the driving behavior data of the current driver, and determine the driving behavior category of the current driver based on the driving behavior data, where the driving behavior data at least includes the distance between the vehicle and the obstacle, vehicle speed, braking, and throttle information; If the driving behavior category is determined to be a gentle driving behavior, output the first group of control parameters corresponding to the gentle driving behavior; If the driving behavior category is determined to be a comfortable driving behavior, output the second group of control parameters corresponding to the comfortable driving behavior; If the driving behavior category is determined to be an aggressive driving behavior, output the third group of control parameters corresponding to the aggressive driving behavior.

3. The method according to claim 1, characterized in that, Performing driving behavior level classification on the simulated driving dataset to obtain a simulated driving behavior classification dataset, including: Classify the data in the simulated driving dataset according to the similarity between the data. After n classifications, obtain n types of simulated driving data, where n is an integer greater than 2; According to the numerical intervals of the n types of simulated driving data, perform driving behavior level classification on the n types of simulated driving data to obtain a simulated driving behavior classification dataset.

4. A hierarchical control system based on driving behavior, characterized in that, Including: An acquisition module for obtaining an actual driving dataset, where the actual driving dataset includes data corresponding to various driving behaviors in different scenarios; A generation module for obtaining a simulated driving dataset according to the actual driving dataset, including: According to the data characteristics of each actual driving data in the actual driving dataset and a preset simulated data generation rule, obtain simulated driving data; Combine the simulated driving data with the actual driving data to obtain a first driving dataset; Calculate the probability that the data in the first driving dataset comes from actual driving data; According to the probability value, calculate the loss functions of the simulated driving behavior generation model and the simulated driving behavior discrimination model; According to the loss functions of the simulated driving behavior generation model and the simulated driving behavior discrimination model, use the neural network algorithm to update the simulated driving data to obtain a simulated driving data set; A classification module, configured to classify the driving behavior levels of the simulated driving data set according to the similarity of each data in the simulated driving data set, and obtain a simulated driving behavior classification data set; A network model training module, configured to import the simulated driving behavior classification data set into a neural network model to obtain groups of control parameters corresponding to different simulated driving behaviors; An output module, configured to output groups of control parameters corresponding to different simulated driving behaviors.

5. The system according to claim 4, characterized in that The obtaining module is further configured to obtain the driving behavior data of the current driver; An output module, which determines the driving behavior category of the current driver based on the driving behavior data, where the driving behavior data at least includes the distance between the vehicle and the obstacle, vehicle speed, braking, and throttle information; If the driving behavior category is determined to be a gentle driving behavior, output the first group of control parameters corresponding to the gentle driving behavior; If the driving behavior category is determined to be a comfortable driving behavior, output the second group of control parameters corresponding to the comfortable driving behavior; If the driving behavior category is determined to be an aggressive driving behavior, output the third group of control parameters corresponding to the aggressive driving behavior.

6. The system according to claim 4, wherein The data classification module is specifically configured to classify the data in the simulated driving data set according to the similarity between each data in the simulated driving data set. After n classifications, n types of simulated driving data are obtained, where n is an integer greater than 2; According to the numerical intervals of the n types of simulated driving data, classify the driving behavior levels of the n types of simulated driving data to obtain a simulated driving behavior classification data set.

7. An electronic device, characterized in that, Including: A memory for storing a computer program; A processor, when executing the computer program stored on the memory, implements the method steps described in any one of claims 1-3.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, it implements the method steps described in any one of claims 1-3.

Images