Driving scene recognition method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention provides a driving scene recognition method and device, electronic equipment and a storage medium. Key driving characteristic parameters are extracted from vehicle data; determining the driving environment of the current vehicle based on at least one parameter dimension of the key driving characteristic parameters; acquiring a parameter weight corresponding to each parameter dimension of the key driving characteristic parameters in the current driving environment; and determining a driving scene of the current vehicle based on the key driving characteristic parameters and the parameter weights, wherein the driving scene is used for determining a vehicle mode of the current vehicle. The driving environment with a large range is determined based on the key driving characteristic parameters, and then the more specific driving scene is determined based on the key driving characteristic parameters and the parameter weights, so that the calculation process can be simplified, and the accuracy of driving scene judgment can be improved.

Description

Technical Field

[0001] This application relates to the field of vehicle control, and more particularly to a driving scenario recognition method, apparatus, electronic device, and storage medium. Background Art

[0002] Nowadays, new energy vehicles can switch different driving modes for different driving scenarios to assist drivers in driving the vehicle. Existing driving scenario recognition methods usually determine the driving scenario based on the real-time vehicle speed, and then determine the driving mode. However, the real-time vehicle speed only represents the vehicle state at that moment and may not necessarily reflect the road state. The judgment of the driving scenario by this solution is not accurate. Moreover, the judgment process of this method is relatively cumbersome, and the execution of complex algorithms requires a large amount of computing time, occupying too much of the controller load. Summary of the Invention

[0003] In view of this, this application provides a driving scenario recognition method, apparatus, electronic device, and storage medium to facilitate solving the problems of inaccurate driving scenario judgment and cumbersome process in the prior art.

[0004] In a first aspect, an embodiment of this application provides a driving scenario recognition method, including: Extracting key driving feature parameters from vehicle data; Determining the driving environment where the current vehicle is located based on at least one parameter dimension of the key driving feature parameters; Obtaining the parameter weights corresponding to each parameter dimension of the key driving feature parameters in the current driving environment; Determining the driving scenario of the current vehicle based on the key driving feature parameters and the parameter weights.

[0005] In an optional embodiment, the parameter dimensions of the key driving feature parameters include: speed parameter; The determining the driving environment where the current vehicle is located based on at least one parameter dimension of the key driving feature parameters includes: Determining the driving environment where the current vehicle is located based on the speed parameter, and the driving environment includes: urban congestion environment, urban unobstructed environment, highway environment, or suburban environment.

[0006] In an optional embodiment, before extracting the key driving feature parameters from the vehicle data, the method further includes: Obtaining historically collected vehicle data, and extracting the driving feature parameters from the historically collected vehicle data; Dividing the driving feature parameters into multiple data sets according to different driving environments, and each data set corresponds to one driving environment; Determining the parameter weights in each driving environment based on the driving feature parameters of each data set; Store the parameter weights in each driving environment.

[0007] In an alternative embodiment, determining the parameter weights corresponding to each driving environment based on the driving characteristic parameters of each data set includes: Divide the driving characteristic parameters in each data set into multiple data groups, and determine the clustering center of each data group as the first representation value; Determine the second representation value based on the difference degree of the corresponding parameter dimensions in different first representation values; Process the second representation value based on the principal component analysis algorithm to obtain the third representation value and the corresponding parameter weights, and each third representation value corresponds to a driving scenario.

[0008] In an alternative embodiment, dividing the driving characteristic parameters in each data set into multiple data groups includes: Determine the corresponding number of clustering centers based on the set number of clusters; Calculate the Euclidean distance between each driving parameter and each clustering center, and divide each driving parameter into the class to which the clustering center with the smallest Euclidean distance belongs; Recalculate the clustering center of each class; If the number of iterations reaches the iteration threshold or the clustering centers of each class no longer change, end the clustering process, otherwise execute the step of calculating the Euclidean distance between each driving parameter and each clustering center.

[0009] In an alternative embodiment, determining the second representation value based on the difference degree of the corresponding parameter dimensions in different first representation values includes: For any parameter dimension in the first representation value, if the numerical difference of the current parameter dimension in different first representations exceeds the preset difference threshold, retain the current parameter dimension, otherwise discard the current parameter dimension; Generate the second representation value based on the parameter dimensions retained in the first representation value.

[0010] In an alternative embodiment, processing the second representation value based on the principal component analysis algorithm to obtain the third representation value and the corresponding parameter weights includes: Construct the covariance matrix of the second representation value; Determine the eigenvalues and eigenvectors of the covariance matrix; Sort the eigenvalues from largest to smallest, determine the third representation value based on the first N eigenvalues, and determine the parameter weights based on the eigenvectors corresponding to the first N eigenvalues.

[0011] In an alternative embodiment, extracting the key driving characteristic parameters from the vehicle data includes: Obtain the vehicle data of the current vehicle in real time; Extract driving characteristic parameters from the vehicle data; Extract key driving characteristic parameters with the same parameter dimension as the third characterization value from the driving characteristic parameters.

[0012] In an optional embodiment, determining the driving scenario of the current vehicle based on the key driving characteristic parameters and the parameter weights includes: Process the key driving characteristic parameters and the parameter weights corresponding to each parameter dimension based on a distance judgment algorithm to determine the weighted Euclidean distance between the current key driving characteristic parameters and each third characterization value; Determine the driving scenario corresponding to the third characterization value with the minimum weighted Euclidean distance as the driving scenario of the current vehicle.

[0013] In an optional embodiment, the method further includes: When it is detected that the newly determined driving scenario is different from the previous driving scenario, determine whether the duration of the previous driving scenario exceeds a preset duration threshold; If it is determined that the duration of the previous driving scenario exceeds the duration threshold, perform a driving scenario switch on the current vehicle. If it is determined that the duration of the previous driving scenario does not exceed the duration threshold, do not perform a driving scenario switch.

[0014] In a second aspect, an embodiment of the present application provides a driving scenario recognition device, including: A first acquisition module, configured to extract key driving characteristic parameters from vehicle data; A first determination module, configured to determine the driving environment in which the current vehicle is located based on at least one parameter dimension of the key driving characteristic parameters; A second acquisition module, configured to acquire the parameter weights corresponding to each parameter dimension of the key driving characteristic parameters in the current driving environment; A second determination module, configured to determine the driving scenario of the current vehicle based on the key driving characteristic parameters and the parameter weights.

[0015] In a third aspect, an embodiment of the present application provides an electronic device, including a memory for storing computer program instructions and a processor for executing the program instructions. When the computer program instructions are executed by the processor, the electronic device is triggered to execute the method according to any one of the first aspects above.

[0016] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, where the computer-readable storage medium includes a stored program. When the program runs, it controls the device where the computer-readable storage medium is located to execute the method according to any one of the first aspects above.

[0017] Fifth aspect, an embodiment of the present application provides a computer program product, which includes executable instructions. When the executable instructions are executed on a computer, the computer is caused to execute the method according to any one of the first aspect.

[0018] By adopting the solution provided by the embodiment of the present application, key driving feature parameters are extracted from vehicle data; a driving environment where the current vehicle is located is determined based on at least one parameter dimension of the key driving feature parameters; parameter weights corresponding to each parameter dimension of each key driving feature parameter in the current driving environment are obtained; a driving scenario of the current vehicle is determined based on the key driving feature parameters and the parameter weights, and the driving scenario is used to determine the vehicle mode of the current vehicle. Compared with determining the driving scenario only by vehicle speed, the embodiment of the present application comprehensively determines based on driving feature parameters in multiple dimensions, improving the accuracy of driving scenario determination. Compared with directly processing the collected driving feature parameters through a model to determine the driving scenario, the embodiment of the present application first determines a relatively large driving environment based on the key driving feature parameters, and then determines a more specific driving scenario based on the key driving feature parameters and the parameter weights, simplifying the overall calculation amount and improving the determination efficiency of the driving scenario. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the accompanying drawings required for use in the embodiments will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present application, and those of ordinary skill in the art can also obtain other accompanying drawings without creative efforts based on these drawings.

[0020] Figure 1 It is a schematic flowchart of a driving scenario recognition method provided by an embodiment of the present application; Figure 2 It is a schematic flowchart of another driving scenario recognition method provided by an embodiment of the present application; Figure 3 It is a schematic flowchart of another driving scenario recognition method provided by an embodiment of the present application; Figure 4 It is a schematic structural diagram of a driving scenario recognition device provided by an embodiment of the present application; Figure 5 It is a schematic structural diagram of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0021] In order to better understand the technical solutions of the present application, the embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0022] It should be clear that the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of the present application.

[0023] The terms used in the embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to limit the present application. The singular forms of "a", "the" and "said" used in the embodiments of the present application and the appended claims are also intended to include the plural forms, unless the context clearly indicates otherwise.

[0024] It should be understood that the term " / and" used herein is only a relational expression describing associated objects, indicating that there can be three relationships. For example, A / and B can mean: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " herein generally indicates that the associated objects before and after are in an "or" relationship.

[0025] Aiming at the problem that the existing method has low accuracy in judging driving scenarios, the embodiments of the present application provide a driving scenario recognition method. First, a relatively large driving environment is determined based on key driving feature parameters, and then a more specific driving scenario is determined based on the key driving feature parameters and parameter weights, which can simplify the calculation process, improve the judgment efficiency of driving scenarios, and improve the accuracy of driving scenario judgment. This method can be applied to a processing device. Optionally, the processing device can specifically refer to an intelligent device installed in a vehicle for assisting the user in driving the vehicle, or the processing device can also refer to devices such as a computer or a server.

[0026] The method of the embodiments of the present application mainly includes: a data processing stage before vehicle driving and a scenario recognition stage during vehicle driving. The process of the data processing stage will be described first below.

[0027] The data processing stage mainly includes steps such as dataset division, clustering processing, principal component analysis, and determining parameter weights.

[0028] (1) Dataset division The embodiments of the present application can divide the driving environment where the vehicle is located into: urban congestion environment, urban unobstructed environment, highway environment, or suburban environment, and each driving environment can be further divided into multiple driving scenarios. During vehicle driving, vehicle data can be collected in real time and summarized to a big data platform, and the processing device can download the vehicle data from the big data platform for subsequent analysis and processing.

[0029] After the vehicle data is acquired, the continuous vehicle data is first divided into several data segments with a fixed time length, and the vehicle data is divided into four data sets according to the vehicle speed signal and the Global Positioning System (GPS) signal for the above four driving environments, and each data set corresponds to one driving environment. In the vehicle data, the processing device can extract driving feature parameters. Optionally, the driving feature parameters can include multiple parameter dimensions, such as speed parameters (such as vehicle speed or vehicle acceleration), battery power, accelerator pedal, brake pedal, motor power, engine power, etc.

[0030] (2) Clustering processing For the driving scenarios further divided for each driving environment, clustering processing is performed on the driving feature parameters in each data set, and multiple data groups can be obtained. The clustering center of each data group corresponds to a driving scenario. In the embodiments of the present application, the driving scenario can be used to characterize the environment where the vehicle is located and the driving state of the vehicle itself. For example, when the vehicle is driving normally in a suburban environment, the driving scenario of the vehicle can be determined as "high-speed driving", and when the vehicle is parked in a suburban environment, the driving scenario of the vehicle can be determined as "automatic parking".

[0031] Specifically, for the driving feature parameters of any driving environment, the processing device can use the k-means clustering algorithm to cluster the driving feature parameters into K driving scenarios, and the number of K is determined in advance by CH index evaluation. For example, the suburban environment can be further divided into three driving scenarios: high-speed driving, automatic parking, and multimedia playing, then the value of K can be set to 3, and the vehicle data will be clustered into 3 data groups. The initial clustering centers of the k-means clustering algorithm are randomly generated, but it is greatly affected by the initial clustering centers, so the algorithm is randomly calculated multiple times, and the clustering results are selected.

[0032] In an optional embodiment, the maximum number of iterations of the algorithm can be set to 100, the number of repeated clusterings is 50 times, and the distance calculation method in the multi-dimensional space is selected as the Euclidean distance. Different K values will be selected for each repeated clustering, and based on different K values, iterative processing will be performed on the driving feature parameters. The steps of the iterative processing can include the following: first determine K clustering centers, then calculate the Euclidean distance between each driving feature parameter and each clustering center respectively, and classify the vehicle data into the class to which the clustering center with the closest Euclidean clustering belongs. Then, recalculate the clustering center of each class based on the Euclidean distance, and repeat the classification of the vehicle data into the corresponding class. Each recalculation of the clustering center can be regarded as one iteration, and the iteration can end when the number of iterations exceeds 100 times or the clustering center no longer changes. 50 repeated clusterings will select 50 different K values and obtain 50 different results. Through user manual evaluation, the most reasonable K value and the corresponding clustering result are selected from the 50 clustering results.

[0033] After clustering is completed, the cluster center coordinates of each data group (i.e., the values of each parameter dimension of the driving feature parameters) are the representation values of the current data group. The processing device calculates the difference degrees of the corresponding parameter dimensions in different representation values, uses the parameter dimensions with larger difference degrees as the judgment basis for driving scenario recognition, and discards the parameter dimensions with smaller difference degrees.

[0034] For example, among the 3 data groups in the suburban environment, the cluster center coordinates (i.e., the representation values) of the first data group are (speed parameter A, motor power A, engine power A), the cluster center coordinates of the second data group are (speed parameter B, motor power B, engine power B), and the cluster center coordinates of the third data group are (speed parameter C, motor power C, engine power C). Suppose the difference degrees among speed parameter A, speed parameter B, and speed parameter C are large (e.g., greater than 25%), the difference degrees among motor power A, motor power B, and motor power C are also large, while the difference degrees among engine power A, engine power B, and engine power C are small. The speed parameter and the motor power can be used as the judgment basis for scenario recognition, while the engine power cannot be used as the judgment basis for scenario recognition. The processing device can discard the parameter dimensions with smaller difference degrees between different representation values and only retain the parameter dimensions with larger difference degrees, and the parameter dimensions of the cluster center coordinates (representation values) of each data group will be reduced. Referring to the above example, the cluster center coordinates (representation values) of the first data group in the suburban environment will become (speed parameter A, motor power A), and the same applies to others.

[0035] Through clustering processing and difference degree recognition, the processing device can obtain the representation values (i.e., cluster center coordinates) corresponding to each driving scenario, and use this as the judgment basis for driving scenarios.

[0036] (3)Principal component analysis On the basis of having initially reduced the dimension of the representation values through the difference degrees of the corresponding parameter dimensions between different representation values, further dimension reduction processing can be performed on the representation values based on principal component analysis. Referring to the above example, after the processing device performs principal component analysis on the 3 data groups in the suburban environment, the motor power is discarded, and only the speed parameter is used as the basis for identifying different driving scenarios. Specifically, the processing device constructs the covariance matrix between different parameter dimensions in the representation values, and obtains the eigenvalues and eigenvectors of the covariance matrix based on the eigenvalue decomposition method. Sort the eigenvalues from large to small, determine N key parameter dimensions in the representation values based on the first N eigenvalues, discard the remaining parameter dimensions, and complete the further dimension reduction processing of the representation values.

[0037] (4)Determine parameter weights Based on the eigenvectors corresponding to the above-mentioned first N eigenvalues, the parameter weights of each parameter dimension can be obtained. Since different driving environments divide different data sets, the parameter weights obtained by processing each data set are only applicable to their respective corresponding driving environments, that is, each driving environment has its own corresponding parameter weights.

[0038] As shown in Table 1, in different driving environments, the parameter weights of each key driving feature parameter are different.

[0039]

[0040] Table 1 The above process can be referred to Figure 1 as shown, and mainly includes: Step 101, obtain vehicle data and extract driving feature parameters.

[0041] Step 102, divide the driving feature parameters into data sets based on the driving environment.

[0042] Step 103, group the driving feature parameters through a clustering algorithm and obtain multiple characterization values.

[0043] Based on the clustering algorithm, the processing device can divide the driving feature parameters in each data set into multiple data groups, and each data group corresponds to a driving scenario.

[0044] Step 104, retain the parameter dimensions with obvious differences in the characterization values.

[0045] Step 105, perform dimensionality reduction processing by principal component analysis.

[0046] Based on the principal component analysis algorithm, process each characterization value, further discard the unimportant parameter dimensions in the characterization value, and achieve dimensionality reduction processing.

[0047] Step 106, determine the parameter weights.

[0048] Through the above process, the processing device first divides four driving environments, each driving environment is further divided into multiple driving scenarios, and each driving scenario has a corresponding characterization value as the recognition basis.

[0049] The following explains the recognition of driving scenarios during vehicle driving.

[0050] The process of driving scenario recognition can be referred to Figure 2 as shown, and mainly includes: Step 201, extract key driving feature parameters from vehicle data.

[0051] Step 202, determine the driving environment where the current vehicle is located based on at least one parameter dimension of the key driving feature parameters.

[0052] Step 203: Obtain the parameter weights corresponding to each parameter dimension of the key driving feature parameters in the current driving environment.

[0053] Step 204: Determine the driving scenario of the current vehicle based on the key driving feature parameters and the parameter weights.

[0054] In an alternative embodiment, before step 201, the processing device needs to first obtain the parameter weights in each driving environment based on the historically collected vehicle data and store them. The specific steps may include: The processing device obtains the historically collected vehicle data and extracts the driving feature parameters in the historically collected vehicle data; divides the driving feature parameters into multiple data sets according to different driving environments, and each data set corresponds to a driving environment; determines the parameter weights in each driving environment based on the driving feature parameters of each data set; stores the parameter weights in each driving environment.

[0055] In an alternative embodiment, the specific steps for the processing device to determine the parameter weights in each driving environment based on the driving feature parameters of each data set may include: (1) Divide the driving feature parameters in each data set into multiple data groups based on the clustering algorithm, and determine the clustering center of each data group as the first representation value.

[0056] After the user sets the initial number of clusters, the processing device first determines the corresponding number of cluster centers. In each iteration, the processing device calculates the Euclidean distance between each driving parameter and each cluster center, and divides each driving parameter into the class to which the cluster center with the smallest Euclidean distance belongs. Then, recalculate the cluster center of each class, usually obtained from the mean of the driving feature parameters in each class. If the number of iterations reaches the iteration threshold or the cluster centers of each class no longer change, the clustering process ends; otherwise, enter the next iteration and re-execute the step of "calculating the Euclidean distance between each driving parameter and each cluster center".

[0057] (2) Determine the second representation value based on the difference degree of the corresponding parameter dimensions among different first representation values.

[0058] For any parameter dimension in the first characterization value, if the numerical difference of the current parameter dimension in different first characterizations exceeds a preset difference threshold, the current parameter dimension is retained; otherwise, the current parameter dimension is discarded. A second characterization value is generated based on the parameter dimensions retained in the first characterization value. For example, if the first characterization value A is (A1, A2, A3) and the second characterization value B is (B1, B2, B3), the "numerical difference between A1 and B1", the "numerical difference between A2 and B2", or the "numerical difference between A3 and B3" can all be regarded as the difference degree of the corresponding parameter dimensions in different characterization values. If the numerical difference between A1 and B1 is small and does not exceed the preset difference threshold, and the numerical differences between A2 and B2 and between A3 and B3 are large and exceed this difference threshold, then A1 and B1 can be discarded, and only A2, A2, B3 are retained. The second characterization value A is (A2, A3), and the second characterization value B is (B2, B3).

[0059] (3) Process the second characterization value based on the principal component analysis algorithm to obtain a third characterization value and the corresponding parameter weights. Each third characterization value corresponds to a driving scenario.

[0060] The processing device constructs the covariance matrix of the second characterization value and calculates the eigenvalues and eigenvectors of the covariance matrix. Each eigenvalue can represent a parameter dimension in the second characterization value. The eigenvalues are sorted from largest to smallest, and the third characterization value is determined based on the first N eigenvalues, and the parameter weights are determined based on the eigenvectors corresponding to the first N eigenvalues. For example, if the second characterization value contains 6 parameter dimensions, constructing the covariance matrix can obtain 6 eigenvalues and 6 eigenvectors. After sorting the 6 eigenvalues, the processing device can take the first 4 eigenvalues to generate the third characterization value, and the third characterization value contains 4 parameter dimensions. The parameter weights of each parameter dimension can be obtained based on the eigenvectors corresponding to the 4 eigenvalues.

[0061] Through the above process, the processing device can obtain the parameter weights in each driving environment and the third characterization value corresponding to each driving scenario. The third characterization value is used for the identification of driving scenarios.

[0062] During the vehicle driving process, the processing device obtains vehicle data in real time, and the vehicle data contains various types of driving characteristic parameters. Since there are many driving characteristic parameters, in order to reduce the computational amount, the processing device can first screen out the key driving characteristic parameters with relatively high importance for the judgment of the driving scenario from the driving characteristic parameters. Specifically, the processing device first extracts the driving characteristic parameters from the vehicle data, and the parameter dimensions of the driving characteristic parameters are numerous. Then, the processing device extracts the key driving characteristic parameters with the same parameter dimension as the third characterization value from the driving characteristic parameters. The parameter dimension of the key driving characteristic parameters is determined by the third characterization value of the above-mentioned driving scenario. For example, if the third characterization value of the driving scenario is (speed parameter, brake pedal, engine power), the processing device can extract the speed parameter, brake pedal parameter, and engine power parameter from the vehicle data obtained in real time as the parameter dimensions of the key driving characteristic parameters.

[0063] If the vehicle data of the above four driving environments is divided by speed data and positioning data, during the offline driving process of the vehicle, the processing device can determine the driving environment where the current vehicle is located based on the speed parameter. In the area where the vehicle can connect to the network, the processing device can also comprehensively determine the driving environment where the current vehicle is located based on the speed parameter and the positioning data.

[0064] After determining the driving environment, the processing device can directly obtain the parameter weights corresponding to each parameter dimension of the key driving characteristic parameters in the current driving environment from the storage space. Then, based on the distance judgment algorithm, the key driving characteristic parameters and the parameter weights corresponding to each parameter dimension are processed to determine the weighted Euclidean distance between the current key driving characteristic parameters and each third characterization value. The driving scenario corresponding to the third characterization value with the smallest weighted Euclidean distance is determined as the driving scenario of the current vehicle. For example, the processing device has determined that the vehicle is in a suburban environment, and the suburban environment contains 3 specific driving scenarios, and each driving scenario corresponds to a third characterization value. The key driving characteristic parameters and each third characterization value can both be regarded as a multi-dimensional coordinate, and the dimensions of the key driving characteristic parameters and the third characterization values are the same. The processing device calculates the weighted Euclidean distances between the key driving characteristic parameters and the three third characterization values respectively, and determines the driving scenario corresponding to the third characterization value with the smallest weighted Euclidean distance as the driving scenario of the current vehicle.

[0065] For different driving scenarios, the processing device can switch to different vehicle modes to assist the driver, thereby improving the driver's driving experience. For example, the vehicle modes can include: economy mode, sport mode, or comfort mode. In economy mode, the throttle response is relatively sluggish, and the gearbox shifts up actively to reduce the engine speed, thereby reducing fuel consumption. This mode is suitable for driving scenarios in urban congested traffic conditions. The sport mode makes the vehicle's power output more rapid, with sensitive throttle response and delayed gearbox upshifting to provide stronger acceleration performance. This mode is suitable for driving scenarios such as quickly overtaking on the highway or requiring strong power on mountain roads. The comfort mode provides a relatively smooth and comfortable driving experience. The suspension system is tuned to be relatively soft and has a strong ability to filter road bumps. This mode is suitable for driving scenarios such as normal driving on urban roads and gentle sections in the suburbs.

[0066] During the actual driving process, some vehicle data may have abnormal values, resulting in incorrect identification of driving scenarios, and then causing frequent switching of vehicle modes, which affects the user experience. To avoid frequent switching of vehicle modes, the processing device can preset a duration threshold, and the interval between every two scene switches cannot be lower than this duration threshold. Specifically, when the processing device detects that the newly determined driving scenario is different from the previous driving scenario, it determines whether the duration of the previous driving scenario exceeds the preset duration threshold; if it is determined that the duration of the previous driving scenario exceeds the duration threshold, then perform a driving scenario switch on the current vehicle, and if it is determined that the duration of the previous driving scenario does not exceed the duration threshold, then do not perform a driving scenario switch. The driving scenario switch is to change the vehicle mode to match the current driving scenario of the vehicle.

[0067] The above steps can be referred to Figure 3 as shown in the flowchart, specifically including: Step 301, obtain vehicle data in real time.

[0068] Step 302, determine the key driving characteristic parameters.

[0069] Step 303, determine the driving scenario based on the key driving characteristic parameters.

[0070] Step 304, detect that the driving scenario is different from the previous driving scenario.

[0071] Step 305, determine whether the cumulative duration of the previous driving scenario exceeds the duration threshold. If so, enter step 306; otherwise, return to step 301.

[0072] Step 306, perform the driving scenario switch operation.

[0073] In the embodiments of the present application, by presetting driving scenarios and dividing vehicle data into different data sets first, it helps to cluster the vehicle data of each driving scenario more accurately, avoids the interference of noise data, and can improve the accuracy of driving scenario judgment. By comprehensively determining based on key driving feature parameters and parameter weights, it does not rely on a single signal for judgment, and situations such as temporary deceleration on the highway will not cause changes in scenario recognition. By training with offline big data, a driving scenario recognition model based on parameter weights is built. Only simple processing of vehicle data is required, and the driving scenario can be quickly judged according to the preset rules, without relying on the computing power of the controller and with rapid recognition.

[0074] Figure 4 It is a schematic structural diagram of a driving scenario recognition device provided by an embodiment of the present application. As Figure 4 shown, the device may include: A first acquisition module 410, configured to extract key driving feature parameters from vehicle data.

[0075] A first determination module 420, configured to determine the driving environment where the current vehicle is located based on at least one parameter dimension of the key driving feature parameters.

[0076] A second acquisition module 430, configured to acquire the parameter weights corresponding to each parameter dimension of the key driving feature parameters in the current driving environment.

[0077] A second determination module 440, configured to determine the driving scenario of the current vehicle based on the key driving feature parameters and the parameter weights.

[0078] Corresponding to the above embodiments, the present application also provides an electronic device. Figure 5 It is a schematic structural diagram of an electronic device provided by an embodiment of the present application. The electronic device 500 may include: a processor 501, a memory 502, and a communication unit 503. These components communicate through one or more buses. Those skilled in the art can understand that the structure of the electronic device shown in the figure does not constitute a limitation on the embodiments of the present application. It can be a bus structure, a star structure, and may also include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0079] Among them, the communication unit 503 is configured to establish a communication channel, so that the electronic device can communicate with other devices. Receive user data sent by other devices or send user data to other devices.

[0080] The processor 501 is the control center of the electronic device, connecting various parts of the entire electronic device through various interfaces and circuits. By running or executing software programs, instructions, and / or modules stored in the memory 502, and by invoking the data stored in the memory, it executes various functions of the electronic device and / or processes data. The processor may be composed of an integrated circuit (IC). For example, it may be composed of a single packaged IC, or it may be composed of multiple packaged ICs with the same or different functions connected together. For example, the processor 501 may only include a central processing unit (CPU). In the embodiments of the present application, the CPU may be a single-core processor or may include multiple cores.

[0081] The memory 502 is used to store the execution instructions of the processor 501. The memory 502 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, a magnetic disk, or an optical disc.

[0082] When the execution instructions in the memory 502 are executed by the processor 501, the electronic device 500 can execute some or all of the steps in the above embodiments.

[0083] In specific implementation, the present application further provides a computer storage medium. The computer storage medium can store a program, and when the program is executed, it may include some or all of the steps in the embodiments of the driving scenario recognition method provided by the present application. The storage medium may be a magnetic disk, an optical disc, a read-only memory (ROM), or a random access memory (RAM), etc.

[0084] In specific implementation, the present application further provides a computer program product. The computer program product contains executable instructions, and when the executable instructions are executed on a computer, the computer executes some or all of the steps in the embodiments of the driving scenario recognition method provided by the present application.

[0085] The embodiments of the present application further provide a non-temporary computer-readable storage medium. The non-temporary computer-readable storage medium stores computer instructions, and the computer instructions cause the computer to execute the driving scenario recognition method provided by the embodiments of the present application.

[0086] The above-mentioned non-transitory computer-readable storage medium may adopt any combination of one or more computer-readable media. The computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples (a non-exhaustive list) of the computer-readable storage medium include: an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (hereinafter referred to as: ROM), an erasable programmable read-only memory (hereinafter referred to as: EPROM), or a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, the computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0087] The computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which the computer-readable program code is carried. Such a propagated data signal may take various forms, including - but not limited to - an electromagnetic signal, an optical signal, or any suitable combination of the above. The computer-readable signal medium may also be any computer-readable medium other than the computer-readable storage medium, which can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0088] The program code contained on the computer-readable medium may be transmitted by any appropriate medium, including - but not limited to - wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0089] Those skilled in the art can clearly understand that the technology in the embodiments of the present application can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution in the embodiments of the present application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments of the present application.

[0090] For the same or similar parts among the various embodiments in this specification, reference can be made to each other. In particular, for the apparatus embodiments and the terminal embodiments, since they are basically similar to the method embodiments, they are described relatively simply, and for the relevant parts, reference can be made to the descriptions in the method embodiments.

Claims

1. A driving scenario recognition method, characterized in that, Including: Extracting key driving feature parameters from vehicle data; Determining the driving environment where the current vehicle is located based on at least one parameter dimension of the key driving feature parameters; Obtaining the parameter weights corresponding to each parameter dimension of the key driving feature parameters in the current driving environment; Determining the driving scenario of the current vehicle based on the key driving feature parameters and the parameter weights.

2. The method according to claim 1, characterized in that The parameter dimensions of the key driving feature parameters include: speed parameter; The determining the driving environment where the current vehicle is located based on at least one parameter dimension of the key driving feature parameters includes: Determining the driving environment where the current vehicle is located based on the speed parameter, and the driving environment includes: urban congestion environment, urban unobstructed environment, highway environment or suburban environment.

3. The method according to claim 1, wherein Before extracting the key driving feature parameters from the vehicle data, the method further includes: Obtaining the historically collected vehicle data and extracting the driving feature parameters in the historically collected vehicle data; Dividing the driving feature parameters into multiple data sets according to different driving environments, and each data set corresponds to a driving environment; Determining the parameter weights in each driving environment based on the driving feature parameters of each data set; Storing the parameter weights in each driving environment.

4. The method according to claim 3, characterized in that, The determining the parameter weights corresponding to each driving environment based on the driving feature parameters of each data set includes: Dividing the driving feature parameters in each data set into multiple data groups, and determining the clustering center of each data group as the first representation value; Determining the second representation value based on the difference degree of the corresponding parameter dimensions in different first representation values; Processing the second representation value based on the principal component analysis algorithm to obtain the third representation value and the corresponding parameter weights, and each third representation value corresponds to a driving scenario.

5. The method according to claim 4, characterized in that, The dividing the driving feature parameters in each data set into multiple data groups includes: Determining the corresponding number of clustering centers based on the set number of clusters; Calculating the Euclidean distance between each driving parameter and each clustering center, and dividing each driving parameter into the class where the clustering center with the smallest Euclidean distance is located; Recalculating the clustering center of each class; If the number of iterations reaches the iteration threshold or the clustering center of each class no longer changes, end the clustering process, otherwise execute the step of calculating the Euclidean distance between each driving parameter and each clustering center.

6. The method according to claim 4, characterized in that The determining the second representation value based on the difference degree of the corresponding parameter dimensions in different first representation values includes: For any parameter dimension in the first representation value, if the numerical difference of the current parameter dimension in different first representations exceeds the preset difference threshold, retain the current parameter dimension, otherwise discard the current parameter dimension; Generating the second representation value based on the parameter dimensions retained in the first representation value.

7. The method according to claim 4, characterized in that The processing the second representation value based on the principal component analysis algorithm to obtain the third representation value and the corresponding parameter weights includes: Constructing the covariance matrix of the second representation value; Determining the eigenvalues and eigenvectors of the covariance matrix; Sorting the eigenvalues from large to small, determining the third representation value based on the first N eigenvalues, and determining the parameter weights based on the eigenvectors corresponding to the first N eigenvalues.

8. The method according to claim 4, characterized in that The extracting the key driving feature parameters from the vehicle data includes: Obtain the vehicle data of the current vehicle in real time; Extract the driving characteristic parameters from the vehicle data; Extract the key driving characteristic parameters whose parameter dimensions are the same as the third characterization value from the driving characteristic parameters.

9. The method according to claim 4, wherein The determining the driving scenario of the current vehicle based on the key driving characteristic parameters and the parameter weights includes: Processing the key driving characteristic parameters and the parameter weights corresponding to each parameter dimension based on the distance judgment algorithm to determine the weighted Euclidean distance between the current key driving characteristic parameters and each third characterization value; Determine the driving scenario corresponding to the third characterization value with the smallest weighted Euclidean distance as the driving scenario of the current vehicle.

10. The method according to claim 1, wherein The method further includes: When it is detected that the newly determined driving scenario is different from the previous driving scenario, judge whether the duration of the previous driving scenario exceeds a preset duration threshold; If it is determined that the duration of the previous driving scenario exceeds the duration threshold, perform a driving scenario switch on the current vehicle. If it is determined that the duration of the previous driving scenario does not exceed the duration threshold, do not perform a driving scenario switch.

11. A driving scenario recognition device, characterized in that, including: A first acquisition module for extracting key driving characteristic parameters from vehicle data; A first determination module for determining the driving environment in which the current vehicle is located based on at least one parameter dimension of the key driving characteristic parameters; A second acquisition module for acquiring the parameter weights corresponding to each parameter dimension of the key driving characteristic parameters in the current driving environment; A second determination module for determining the driving scenario of the current vehicle based on the key driving characteristic parameters and the parameter weights.

12. An electronic device, characterized in that, including a memory for storing computer program instructions and a processor for executing the program instructions, wherein when the computer program instructions are executed by the processor, the electronic device executes the method according to any one of claims 1 to 10.

13. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, wherein when the program runs, it controls the device where the computer-readable storage medium is located to execute the method according to any one of claims 1 to 10.

Images