Driving style identification method and system based on entropy weight method

Abstract

The invention belongs to the technical field of artificial intelligence, and provides a driving style recognition method and system based on an entropy weight method.The driving style recognition method based on the entropy weight method comprises the steps that processing is conducted on the basis of original driving data of a target vehicle and an artificial neural network model, obtaining a predicted driving cycle label of each time slice and a predicted driving style label of each time slice; storing the predicted driving cycle label of each time slice and the predicted driving style label of each time slice; calculating based on the time number and the predicted driving style label to obtain an average style score; calculating based on the predicted driving cycle label and the predicted driving style label to obtain an entropy weight; and performing weighted calculation based on the average style score and the entropy weight to obtain a target driving style score, and determining the target driving style based on the target driving style score and a preset score, thereby significantly improving the objectivity and reproducibility of the score.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a driving style recognition method and system based on an entropy weight method. Background Art

[0002] In recent years, as cars have become increasingly intelligent, users have increasingly demanded more adaptive, intelligent control. This has led to the emergence of driving style recognition technology. This technology primarily involves identifying data features based on collected data and training models based on machine learning or deep learning algorithms and data features to accurately identify the driver's driving style.

[0003] Currently, driving style recognition technology often uses the Analytic Hierarchy Process (AHP) to assign importance to driving cycles. However, AHP requires domain experts to set a pairwise comparison matrix based on their personal experience or knowledge. The judgment criteria provided by different experts vary, resulting in highly subjective weights and difficulty ensuring consistency across different environments. Once the AHP judgment matrix is ​​established, its weights are fixed and cannot be dynamically adjusted to varying driving environments or driver behavior, making it difficult to adapt to the diversity and complexity of actual driving. Furthermore, when there are many different cycles or complex judgment criteria, the manual construction of the judgment matrix by experts is prone to logical inconsistencies, requiring consistency checks, increasing design complexity and the probability of errors.

[0004] Therefore, how to improve the objectivity and consistency of the scoring results in driving style recognition technology is an urgent problem that technicians in this field need to solve. Summary of the Invention

[0005] In view of this, embodiments of the present invention provide a driving style recognition method and system based on the entropy weight method to solve the problem of how to improve the objectivity and consistency of scoring results in driving style recognition technology in the existing technology; that is, embodiments of the present invention can significantly improve the objectivity and reproducibility of scoring.

[0006] According to one aspect of the present invention, a driving style identification method based on an entropy weight method is provided. The driving style identification method based on the entropy weight method includes: processing raw driving data of a target vehicle and an artificial neural network model to obtain a predicted driving cycle label and a predicted driving style label for each time segment; storing the predicted driving cycle label and the predicted driving style label for each time segment, and recording the time number corresponding to each time segment; counting the number of stored predicted driving cycle labels or predicted driving style labels for each time segment, and determining whether the stored number is equal to or greater than a preset storage number; when the stored number is equal to or greater than the preset storage number, calculating based on the time number and the predicted driving style label to obtain an average style score; when the stored number is equal to or greater than the preset storage number, calculating based on the predicted driving cycle label and the predicted driving style label to obtain an entropy weight; performing a weighted calculation based on the average style score and the entropy weight to obtain a target driving style score, and determining a target driving style based on the target driving style score and the preset score.

[0007] In one embodiment, the processing of the original driving data of the target vehicle and the artificial neural network model to obtain a predicted driving cycle label for each time segment and a predicted driving style label for each time segment includes: acquiring the original driving data of the target vehicle in real time, and performing cleaning processing based on the original driving data to obtain target driving data, wherein the original driving data includes training original driving data and non-training original driving data; dividing the target driving data into time segments based on a preset driving duration to obtain target driving data for at least one time segment; performing feature extraction based on the target driving data of each time segment to obtain driving cycle features and driving style for the corresponding time segment. characteristics; performing dimensionality reduction processing based on the driving cycle characteristics and the driving style characteristics to obtain driving principal component characteristics, wherein the driving principal component characteristics include training driving principal component characteristics and non-training driving principal component characteristics; training an artificial neural network model based on the training driving principal component characteristics to obtain a target driving cycle artificial neural network model and a target driving style artificial neural network model, respectively; calling the target driving cycle artificial neural network model, performing recognition processing based on the non-training driving principal component characteristics to obtain the predicted driving cycle label, and calling the target driving style artificial neural network model, performing recognition processing based on the non-training driving principal component characteristics to obtain the predicted driving style label.

[0008] In one embodiment, the training of the artificial neural network model based on the training driving principal component features to obtain a target driving cycle artificial neural network model and a target driving style artificial neural network model, respectively, includes: performing clustering processing based on the training driving principal component features to obtain corresponding real driving cycle labels and corresponding real driving style labels; and training the artificial neural network model based on the training driving principal component features, the real driving cycle labels, and the real driving style labels to obtain a target driving cycle artificial neural network model and a target driving style artificial neural network model, respectively.

[0009] In one embodiment, the training of the artificial neural network model based on the training driving principal component features, the real driving cycle labels, and the real driving style labels to obtain a target driving cycle artificial neural network model and a target driving style artificial neural network model, respectively, includes: obtaining a prediction probability based on the training driving principal component features; calculating a target loss value based on the prediction probability, the real driving cycle labels, and the real driving style labels; and optimizing model parameters in the artificial neural network model in a direction of reducing the target loss value to obtain the target driving cycle artificial neural network model and the target driving style artificial neural network model.

[0010] In one embodiment, when the storage quantity is equal to or greater than the preset storage quantity, a calculation is performed based on the time number and the predicted driving style label to obtain an average style score, including: performing a calculation based on the time number to obtain a time decay weight; and performing a weighted calculation based on the predicted driving style label and the time decay weight to obtain the average style score.

[0011] In one embodiment, performing a weighted calculation based on the predicted driving style label and the time decay weight to obtain the average style score includes: performing a calculation based on the predicted driving cycle label and the predicted driving style label to obtain the frequency of occurrence of each type of driving style label; performing a calculation based on the frequency of occurrence of each type of driving style label to obtain an entropy value; and performing a calculation based on the entropy value to obtain the entropy weight.

[0012] In one embodiment, the target driving style score is obtained by performing weighted calculation based on the average style score and the entropy weight, and the target driving style score is:

[0013]

[0014] Among them, w i is the entropy weight, i is the type of predicted driving cycle label, avg_score i(t) is the average style score.

[0015] According to another aspect of the present invention, a driving style recognition system based on an entropy weight method is provided, and the driving style recognition system based on the entropy weight method includes: a data processing module, a data storage module, a comparison module, a first calculation module, a second calculation module and a style determination module, wherein the data processing module is used to process the original driving data of the target vehicle and the artificial neural network model to obtain a predicted driving cycle label of each time segment and a predicted driving style label of each time segment; the data storage module is used to store the predicted driving cycle label of each time segment and the predicted driving style label of each time segment, and record the time number corresponding to each time segment; the comparison module is used to store the predicted driving cycle label of each time segment and the predicted driving style label of each time segment, and record the time number corresponding to each time segment; the comparison module is used to compare the stored The number of predicted driving cycle labels for a time segment or the predicted driving style labels for each time segment is counted, and it is determined whether the stored number is equal to or greater than a preset stored number; when the stored number is equal to or greater than the preset stored number, a first calculation module is used to perform calculation based on the time number and the predicted driving style label to obtain an average style score; when the stored number is equal to or greater than the preset stored number, a second calculation module is used to perform calculation based on the predicted driving cycle label and the predicted driving style label to obtain an entropy weight; a style determination module is used to perform weighted calculation based on the average style score and the entropy weight to obtain a target driving style score, and determine a target driving style based on the target driving style score and the preset score.

[0016] In one embodiment, the data processing module includes a data acquisition module, a time segment division module, a feature extraction module, a feature processing module, a model training module and a label generation module, wherein the data acquisition module is used to acquire the original driving data of the target vehicle in real time, and perform cleaning processing based on the original driving data to obtain target driving data, wherein the original driving data includes training original driving data and non-training original driving data; the time segment division module is used to divide the target driving data into time segments based on a preset driving duration to obtain target driving data of at least one time segment; the feature extraction module is used to perform feature extraction based on the target driving data of each time segment to obtain driving cycle characteristics and driving style characteristics of the corresponding time segment; The feature processing module is configured to perform dimensionality reduction processing based on the driving cycle features and the driving style features to obtain driving principal component features, wherein the driving principal component features include training driving principal component features and non-training driving principal component features. The model training module is configured to train an artificial neural network model based on the training driving principal component features to obtain a target driving cycle artificial neural network model and a target driving style artificial neural network model, respectively. The label generation module is configured to call the target driving cycle artificial neural network model, perform identification processing based on the non-training driving principal component features to obtain the predicted driving cycle label, and call the target driving style artificial neural network model, perform identification processing based on the non-training driving principal component features to obtain the predicted driving style label.

[0017] In one embodiment, the first calculation module includes a first calculation module and a second calculation module, wherein the first calculation module is used to perform calculation based on the time number to obtain a time decay weight; the second calculation module is used to perform weighted calculation based on the predicted driving style label and the time decay weight to obtain the average style score.

[0018] In summary, in an embodiment of the present invention, the original driving data of the target vehicle and the artificial neural network model are processed to obtain a predicted driving cycle label and a predicted driving style label for each time segment. The predicted driving cycle label and the predicted driving style label for each time segment are stored, and the time number corresponding to each time segment is recorded. The number of the stored predicted driving cycle labels or the predicted driving style labels for each time segment is counted, and it is determined whether the stored number is equal to or greater than a preset storage number. When the stored number is equal to or greater than a preset storage number, the predicted driving cycle label or the predicted driving style label for each time segment is counted. When the number of stored times is greater than the preset number, a calculation is performed based on the time number and the predicted driving style label to obtain an average style score. When the number of stored times is equal to or greater than the preset number, a calculation is performed based on the predicted driving cycle label and the predicted driving style label to obtain an entropy weight. A weighted calculation is performed based on the average style score and the entropy weight to obtain a target driving style score. A target driving style is determined based on the target driving style score and the preset score. This not only significantly improves the objectivity and reproducibility of the score, but also can flexibly adapt to the actual characteristics of different drivers, different road conditions, or time periods, and has strong environmental adaptability and generalization capabilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Further details, features and advantages of the present invention are disclosed in the following description of exemplary embodiments in conjunction with the accompanying drawings, in which:

[0020] Figure 1 A schematic diagram of a process of a driving style identification method based on entropy weight method disclosed in an embodiment of the present application is shown;

[0021] Figure 2 Shown Figure 1 The schematic diagram of the step flow of step S110 is shown;

[0022] Figure 3 Shown Figure 2 The schematic diagram of the step flow of step S115 is shown;

[0023] Figure 4 Shown Figure 3 The schematic diagram of the step flow of step S1152 is shown;

[0024] Figure 5 Shown Figure 1 The schematic diagram of the step flow of step S140 is shown;

[0025] Figure 6 Shown Figure 1 The schematic diagram of the step flow of step S150 is shown;

[0026] Figure 7A schematic structural diagram of a driving style recognition system based on entropy weight method disclosed in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0027] Embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be construed as limited to the embodiments described herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of the present invention. It should be understood that the drawings and embodiments of the present invention are for illustrative purposes only and are not intended to limit the scope of protection of the present invention.

[0028] It should be understood that the various steps described in the method embodiments of the present invention may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this respect.

[0029] The term "including" and its variations used in this document are open inclusions, that is, "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one other embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description. It should be noted that the concepts of "first", "second", etc. mentioned in the present invention are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0030] It should be noted that the modifications of "one" and "multiple" mentioned in the present invention are illustrative rather than restrictive. Those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0031] The names of the messages or information exchanged between multiple devices in the embodiments of the present invention are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0032] It should be noted that the execution subject of the driving style recognition method based on the entropy weight method provided in the embodiment of the present invention can be one or more electronic devices, which is not limited by the present invention. Among them, the electronic device can be a terminal (i.e., a client) or a server. Then, when the execution subject includes multiple electronic devices, and the multiple electronic devices include at least one terminal and at least one server, the driving style recognition method based on the entropy weight method provided in the embodiment of the present invention can be jointly executed by the terminal and the server. Accordingly, the terminals mentioned here can include but are not limited to: smartphones, tablets, laptops, desktop computers, smart watches, intelligent voice interaction devices, smart home appliances, vehicle terminals, aircraft, etc. The server mentioned here can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN (Content Delivery Network), and basic cloud computing services such as big data and artificial intelligence platforms, etc.

[0033] Based on the above description, embodiments of the present invention provide a driving style recognition method based on an entropy weighting method. This method can be performed by the aforementioned electronic device (terminal or server); alternatively, this method can be performed jointly by a terminal and a server. For ease of illustration, the following description uses an electronic device performing this method as an example.

[0034] See also Figure 1 , which is a flow chart of a driving style recognition method based on entropy weight method disclosed in the embodiment of the present application. The problem of how to improve the objectivity and consistency of the scoring results in the driving style recognition technology is solved by a driving style recognition method based on entropy weight method. It should be noted that the driving style recognition method based on entropy weight method in the embodiment of the present application is not limited to Figure 1 The steps and order in the flowchart shown. According to different needs, the steps in the flowchart shown can be added, removed, or changed in order. In the embodiment of the present application, Figure 1 As shown, a process of a driving style identification method based on entropy weight method includes at least the following steps.

[0035] S110 , processing is performed based on the original driving data of the target vehicle and the artificial neural network model to obtain a predicted driving cycle label for each time segment and a predicted driving style label for each time segment.

[0036] like Figure 2As shown, in an embodiment of the present invention, Figure 2 The step S110 includes at least the following steps:

[0037] S111. Acquire original driving data of a target vehicle in real time, and perform cleaning processing based on the original driving data to obtain target driving data, wherein the original driving data includes training original driving data and non-training original driving data.

[0038] In an embodiment of the present invention, the original driving data of a vehicle may include at least vehicle speed data, pedal signal data, time series data, abnormal data, fluctuation data, etc. Among them, the vehicle speed data may include at least vehicle speed, acceleration, and jerk; the pedal signal data may include at least accelerator pedal opening, accelerator pedal change rate, brake pedal opening, and brake pedal change rate; the time series data may include at least segments with discontinuous or missing sampling time; and the abnormal data may include at least acceleration absolute values ​​exceeding 10 m / s. 2 The abnormal value of the fluctuation data may at least include the fluctuation noise of the vehicle speed and the fluctuation noise of the pedal data.

[0039] The raw driving data is cleaned to obtain the target driving data. Specifically, first, due to sensor errors or environmental interference, the raw driving data may contain noise, which affects the accuracy of data analysis. Second, time discontinuities or data loss may occur during the data sampling process, which requires interpolation to supplement. Third, extreme values ​​(i.e., abnormal acceleration) may appear during the data sampling process, thereby distorting the statistical results and requiring elimination or correction. Fourth, data fluctuations (i.e., sudden changes in vehicle speed) require smoothing through filtering to reflect the actual driving situation. Fifth, the target driving data obtained through cleaning can more accurately represent the driving cycle and driving style, thereby ensuring the stability and accuracy of the artificial neural network (ANN) model.

[0040] S112: Divide the target driving data into time segments based on a preset driving duration to obtain target driving data of at least one time segment.

[0041] In an embodiment of the present invention, the target driving data is divided into time segments based on a preset driving duration to obtain target driving data for at least one time segment. The preset driving duration may be 40 seconds. It is understood that the preset driving duration can be set based on actual needs and is not limited in this embodiment.

[0042] S113 . Perform feature extraction based on the target driving data of each time segment to obtain driving cycle features and driving style features of the corresponding time segment.

[0043] In an embodiment of the present invention, feature extraction is performed based on the target driving data for each time segment to obtain drive cycle features and driving style features for the corresponding time segment. The drive cycle features can be extracted based on the vehicle speed, acceleration, and time distribution in the target driving data, while the driving style features can be extracted based on the pedal data and jerk in the target driving data.

[0044] S114. Perform dimensionality reduction processing based on the driving cycle characteristics and the driving style characteristics to obtain driving principal component characteristics, wherein the driving principal component characteristics include training driving principal component characteristics and non-training driving principal component characteristics.

[0045] In an embodiment of the present invention, principal component analysis (PCA) dimensionality reduction is performed based on driving cycle characteristics and driving style characteristics to obtain driving principal component characteristics. Specifically, the driving cycle characteristics and driving style characteristics are projected into a low-dimensional space through an orthogonal transformation, thereby preserving the direction of maximum variance. PCA dimensionality reduction can reduce computational complexity, generate uncorrelated principal components through orthogonal transformation to avoid duplication of information that interferes with the ANN model, make the processed data more compact, and significantly improve the efficiency of the ANN model while ensuring model accuracy.

[0046] S115 . Training the artificial neural network model based on the training driving principal component features to obtain a target driving cycle artificial neural network model and a target driving style artificial neural network model.

[0047] like Figure 3 As shown, in an embodiment of the present invention, Figure 3 The step S115 at least includes the following steps:

[0048] S1151 . Perform clustering processing based on the training driving principal component features to obtain corresponding real driving cycle labels and corresponding real driving style labels.

[0049] In this embodiment of the present invention, Kmeans clustering is performed based on the principal component features of the training driving data to obtain corresponding real driving cycle labels and corresponding real driving style labels. The driving cycle labels can be divided into four types, and the driving style labels can be divided into three types. Specifically, the four driving cycle labels can include at least 0-3, where 0 represents congestion, 1 represents low speed, 2 represents medium speed, and 3 represents high speed. The three driving style labels can include at least 0-2, where 0 represents conservative, 1 represents normal, and 2 represents aggressive.

[0050] S1152: Train the artificial neural network model based on the training driving principal component features, the real driving cycle label, and the real driving style label to obtain a target driving cycle artificial neural network model and a target driving style artificial neural network model, respectively.

[0051] like Figure 4 As shown, in an embodiment of the present invention, Figure 4 The step S1152 at least includes the following steps:

[0052] S1152a. Obtain a predicted probability based on the training driving principal component features.

[0053] In the embodiment of the present invention, the calculation is performed by forward propagation, and the calculation process is as follows:

[0054] data h =ω1·data PAC +b1 formula (1)

[0055] data out =ω2·data ReLU +b2 Formula (2)

[0056] Among them, data h is the output data of the first model, ω1 is the first weight coefficient, data PAC To train the driving principal component features, b1 is the first bias, data out is the output data of the second model, ω2 is the second weight coefficient, data ReLU is the input data processed by the activation function, and b2 is the second bias.

[0057] The ReLU activation function is as follows:

[0058]

[0059] Where x is the input data.

[0060] Then, based on the Softmax normalization, the probability of the model prediction is output. The probability of the model prediction is as follows:

[0061]

[0062] Among them, data out,i Output data of the second model corresponding to the i-th data, is the second model output data corresponding to the j-th data, and k is 4 (that is, the number of types of driving cycle labels) or 3 (that is, the number of types of driving style labels).

[0063] S1152b. Calculate a target loss value based on the predicted probability, the actual driving cycle label, and the actual driving style label.

[0064] In this embodiment of the present invention, a target loss value is obtained by calculating the predicted probability, the actual driving cycle label, and the actual driving style label using cross-entropy loss (CEE). The target loss value is calculated as follows:

[0065]

[0066] Among them, q i is the real driving cycle label or real driving style label, prob i is the predicted probability.

[0067] S1152c. Optimize the model parameters in the artificial neural network model in a direction of reducing the target loss value to obtain the target driving cycle artificial neural network model and the target driving style artificial neural network model.

[0068] S116: Calling the target driving cycle artificial neural network model to perform recognition processing based on the non-training driving principal component features to obtain the predicted driving cycle label, and calling the target driving style artificial neural network model to perform recognition processing based on the non-training driving principal component features to obtain the predicted driving style label.

[0069] S120: Store the predicted driving cycle label and the predicted driving style label of each time segment, and record the time number corresponding to each time segment.

[0070] In this embodiment of the present invention, the predicted driving cycle label and predicted driving style label for each time segment are stored in a sliding window, and the time number of each time segment is recorded. Specifically, taking the first time segment as an example, its time number is 1, and its corresponding predicted driving cycle label and predicted driving style label are stored.

[0071] S130: Count the number of the stored predicted driving cycle labels or predicted driving style labels for each time segment, and determine whether the stored number is equal to or greater than a preset storage number.

[0072] In the embodiment of the present invention, when the storage quantity is less than the preset storage quantity, accumulation continues. It is understandable that the preset storage quantity can be set based on actual needs, and this embodiment does not limit this.

[0073] S140: When the stored number is equal to or greater than the preset stored number, perform calculation based on the time number and the predicted driving style label to obtain an average style score.

[0074] like Figure 5 As shown, in an embodiment of the present invention, Figure 5 The step S140 at least includes the following steps:

[0075] S141. Perform calculation based on the time number to obtain a time decay weight.

[0076] In this embodiment of the present invention, the time segments with the same predicted driving cycle labels are put into the time segment set In , the calculation process of the time decay weight of each time segment is as follows:

[0077]

[0078] Where λ is the time attenuation coefficient, λ>0, f ik is the time number of the kth time segment in the time segment set, and t is the time number of the current time segment.

[0079] S142: Perform weighted calculation based on the predicted driving style label and the time decay weight to obtain the average style score.

[0080] In the embodiment of the present invention, the time decay weight of each time segment is based on And the style score v corresponding to the predicted driving style label for each time segment ik Perform weighted calculation to get the average style score. Average style score avg_score i The calculation process of (t) is as follows:

[0081]

[0082] Among them, v ik Predict the style score corresponding to the driving style label for each time segment, is the time attenuation weight of each time segment, and k is the time number of the time segment.

[0083] S150: When the stored quantity is equal to or greater than the preset stored quantity, perform calculation based on the predicted driving cycle label and the predicted driving style label to obtain an entropy weight.

[0084] like Figure 6 As shown, in an embodiment of the present invention, Figure 6 The step S150 at least includes the following steps:

[0085] S151 : Calculate based on the predicted driving cycle label and the predicted driving style label to obtain the frequency of occurrence of each type of driving style label.

[0086] In an embodiment of the present invention, in a time segment of each type of predicted driving cycle label, the frequency of occurrence of each type of predicted driving style label is calculated to obtain the frequency of occurrence of each type of driving style label.

[0087] S152: Calculate the frequency of each type of driving style label to obtain an entropy value.

[0088] In this embodiment of the present invention, an entropy value is calculated based on the frequency of occurrence of each type of driving style label. The entropy value reflects the degree of dispersion or uncertainty of the predicted driving style label within the time segment of each type of predicted driving cycle label. The smaller the entropy value, the more concentrated the cycle style, the stronger the discrimination, and the higher the weight; the larger the entropy value, the more mixed the cycle style, the weaker the discrimination, and the lower the weight. The entropy value calculation process is as follows:

[0089]

[0090] Among them, i is the type of predicted driving cycle label, k is the normalization coefficient, and s ij is the frequency of occurrence of the jth predicted driving style label in the time segment of the ith predicted driving cycle label.

[0091] S153. Perform calculation based on the entropy value to obtain the entropy weight.

[0092] In the embodiment of the present invention, the entropy weight is obtained by calculation based on the entropy value. The calculation process of the entropy weight is as follows:

[0093]

[0094] Among them, 1-E i The effective amount of information in the time segment of the driving cycle label predicted for the i-th class.

[0095] S160 : Perform weighted calculation based on the average style score and the entropy weight to obtain a target driving style score, and determine a target driving style based on the target driving style score and a preset score.

[0096] In this embodiment of the present invention, a target driving style score is obtained by performing a weighted summation based on the average style score and the entropy weight. The target driving style score is calculated as follows:

[0097]

[0098] When the target driving style score Scoretotal >85, the target driving style is determined to be conservative; when 75<target driving style score total When the target driving style score is ≤85, the target driving style is determined to be conventional; when the target driving style score is total When ≤75, the target driving style is determined to be aggressive.

[0099] In summary, in a driving style recognition method based on the entropy weight method of the present application, the original driving data of the target vehicle and the artificial neural network model are processed to obtain a predicted driving cycle label and a predicted driving style label for each time segment, the predicted driving cycle label and the predicted driving style label for each time segment are stored, and the time number corresponding to each time segment is recorded. The number of the stored predicted driving cycle labels or the predicted driving style labels for each time segment is counted, and it is determined whether the stored number is equal to or greater than the preset storage number. When the predicted driving cycle labels or the predicted driving style labels for each time segment are stored, the predicted driving cycle labels or the predicted driving style labels for each time segment are counted. When the storage quantity is equal to or greater than the preset storage quantity, a calculation is performed based on the time number and the predicted driving style label to obtain an average style score. When the storage quantity is equal to or greater than the preset storage quantity, a calculation is performed based on the predicted driving cycle label and the predicted driving style label to obtain an entropy weight. A weighted calculation is performed based on the average style score and the entropy weight to obtain a target driving style score. A target driving style is determined based on the target driving style score and the preset score. This not only significantly improves the objectivity and reproducibility of the score, but also can flexibly adapt to the actual characteristics of different drivers, different road conditions, or time periods, and has strong environmental adaptability and generalization capabilities.

[0100] See also Figure 7 , which is a schematic diagram of the structure of a driving style recognition system based on entropy weight method disclosed in an embodiment of the present application. In one embodiment, Figure 7 As shown, the present application provides a driving style recognition system 100 based on an entropy weight method. A driving style recognition system 100 based on an entropy weight method may include at least: a data processing module 110, a data storage module 130, a comparison module 150, a first calculation module 170, a second calculation module 180, and a style determination module 190. There is information exchange between the data processing module 110 and the data storage module 130, between the data storage module 130 and the comparison module 150, between the comparison module 150 and the first calculation module 170, between the first calculation module 170 and the second calculation module 180, and between the second calculation module 180 and the style determination module 190.

[0101] Data processing module 110 processes the target vehicle's raw driving data and the artificial neural network model to obtain predicted drive cycle labels and driving style labels for each time segment. Data processing module 110 may include at least a data acquisition module 111, a time segment segmentation module 113, a feature extraction module 115, a feature processing module 116, a model training module 117, and a label generation module 118.

[0102] The data acquisition module 111 is used to acquire the original driving data of the target vehicle in real time, and to obtain the target driving data by cleaning the original driving data, wherein the original driving data includes training original driving data and non-training original driving data. Specifically, the original driving data of the vehicle may include at least vehicle speed data, pedal signal data, time series data, abnormal data, fluctuation data, etc. Among them, the vehicle speed data may include at least vehicle speed, acceleration and jerk, the pedal signal data may include at least accelerator pedal opening, accelerator pedal change rate, brake pedal opening and brake pedal change rate, time series data may include at least segments with discontinuous or missing sampling time, and abnormal data may include at least acceleration absolute value exceeding 10m / s 2 The abnormal values ​​of the fluctuation data may at least include the fluctuation noise of the vehicle speed and the fluctuation noise of the pedal data. The original driving data is cleaned to obtain the target driving data. Specifically, first, due to sensor errors or environmental interference, the original driving data may contain noise, and the noise will affect the accuracy of data analysis; second, time discontinuity or data loss may occur during the data sampling process, which needs to be supplemented by interpolation; third, extreme values ​​(that is, abnormal acceleration) may appear during the data sampling process, thereby distorting the statistical results and need to be eliminated or corrected; fourth, data fluctuations (that is, sudden changes in vehicle speed) need to be smoothed by filtering to reflect the actual driving situation; fifth, the target driving data obtained through cleaning can more accurately characterize the driving cycle and driving style, thereby ensuring the stability and accuracy of the artificial neural network (ANN) model.

[0103] The time segment division module 113 is configured to divide the target driving data into time segments based on a preset driving duration to obtain target driving data for at least one time segment. Specifically, the time segment division module 113 divides the target driving data into time segments based on a preset driving duration to obtain target driving data for at least one time segment. The preset driving duration may be 40 seconds. It will be appreciated that the preset driving duration may be set based on actual needs and is not limited in this embodiment.

[0104] The feature extraction module 115 is configured to extract features based on the target driving data for each time segment to obtain driving cycle features and driving style features for the corresponding time segment. Specifically, the feature extraction module 115 extracts features based on the target driving data for each time segment to obtain driving cycle features and driving style features for the corresponding time segment. The driving cycle features can be extracted based on the vehicle speed, acceleration, and time distribution in the target driving data, and the driving style features can be extracted based on the pedal data and jerk in the target driving data.

[0105] The feature processing module 116 is used to perform dimensionality reduction processing based on the driving cycle features and the driving style features to obtain driving principal component features, wherein the driving principal component features include training driving principal component features and non-training driving principal component features. Specifically, the feature processing module 116 performs principal component analysis (PCA) dimensionality reduction based on the driving cycle features and the driving style features to obtain driving principal component features. Specifically, the driving cycle features and the driving style features are projected into a low-dimensional space through an orthogonal transformation, thereby retaining the direction of the maximum variance. PCA dimensionality reduction can reduce computational complexity, and can generate unrelated principal components through orthogonal transformation to avoid repeated information interfering with the ANN model. This can make the data obtained after processing more compact, and can also significantly improve the efficiency of the ANN model while ensuring model accuracy.

[0106] The model training module 117 is used to train the artificial neural network model based on the training driving principal component features to obtain a target driving cycle artificial neural network model and a target driving style artificial neural network model. The model training module 117 may include at least: a clustering processing unit 1171 and a training unit 1173.

[0107] The clustering processing unit 1171 is configured to perform clustering based on the training driving principal component features to obtain corresponding real driving cycle labels and corresponding real driving style labels. Specifically, the clustering processing unit 1171 performs Kmeans clustering based on the training driving principal component features to obtain corresponding real driving cycle labels and corresponding real driving style labels. The driving cycle labels can be divided into four types of driving cycle labels, and the driving style labels can be divided into three types of driving style labels. Specifically, the four driving cycle labels can include at least 0-3, where 0 represents congestion, 1 represents low speed, 2 represents medium speed, and 3 represents high speed; the three driving style labels can include at least 0-2, where 0 represents conservative, 1 represents normal, and 2 represents aggressive.

[0108] The training unit 1173 is configured to train the artificial neural network model based on the training driving principal component features, the real driving cycle labels, and the real driving style labels, thereby obtaining a target driving cycle artificial neural network model and a target driving style artificial neural network model, respectively. Specifically, the training unit 1173 obtains a predicted probability based on the training driving principal component features. In this embodiment of the present invention, the calculation is performed via forward propagation, as shown in the following formula:

[0109] data h =ω1·data PAC +b1 formula (1)

[0110] data out =ω2·data ReLU +b2 Formula (2)

[0111] Among them, data h is the output data of the first model, ω1 is the first weight coefficient, data PAC To train the driving principal component features, b1 is the first bias, data out is the output data of the second model, ω2 is the second weight coefficient, data ReLU is the input data processed by the activation function, and b2 is the second bias.

[0112] The ReLU activation function is as follows:

[0113]

[0114] Where x is the input data.

[0115] Then, based on the Softmax normalization, the probability of the model prediction is output. The probability of the model prediction is as follows:

[0116]

[0117] Among them, data out,i Output data of the second model corresponding to the i-th data, is the second model output data corresponding to the j-th data, and k is 4 (that is, the number of types of driving cycle labels) or 3 (that is, the number of types of driving style labels).

[0118] The training unit 1173 calculates a target loss value based on the predicted probability, the actual driving cycle label, and the actual driving style label. In an embodiment of the present invention, the target loss value is calculated using a cross-entropy loss (CEE) on the predicted probability, the actual driving cycle label, and the actual driving style label. The target loss value is calculated as follows:

[0119]

[0120] Among them, q i is the real driving cycle label or real driving style label, prob i is the predicted probability.

[0121] The training unit 1173 optimizes the model parameters in the artificial neural network model in the direction of reducing the target loss value to obtain the target driving cycle artificial neural network model and the target driving style artificial neural network model.

[0122] The label generation module 118 is used to call the target driving cycle artificial neural network model, perform identification processing based on the non-training driving principal component features to obtain the predicted driving cycle label, and call the target driving style artificial neural network model, perform identification processing based on the non-training driving principal component features to obtain the predicted driving style label.

[0123] The data storage module 130 is configured to store the predicted driving cycle label and the predicted driving style label for each time segment, and to record the time number corresponding to each time segment. Specifically, the data storage module 130 stores the predicted driving cycle label and the predicted driving style label for each time segment in a sliding window, while also recording the time number for each time segment. For example, the first time segment, whose time number is 1, is used to store the corresponding predicted driving cycle label and predicted driving style label.

[0124] Comparison module 150 is configured to count the number of stored predicted driving cycle labels or predicted driving style labels for each time segment and determine whether the stored number is equal to or greater than a preset storage number. Specifically, if the stored number is less than the preset storage number, accumulation continues. It will be appreciated that the preset storage number can be set based on actual needs and is not limited in this embodiment.

[0125] When the stored number is equal to or greater than the preset stored number, the first calculation module 170 is configured to calculate based on the time number and the predicted driving style label to obtain an average style score. The first calculation module 170 may include at least: a first calculation module 171 and a second calculation module 173.

[0126] The first calculation module 171 is used to calculate based on the time number to obtain the time decay weight. Specifically, the first calculation module 171 puts the time segments with the same predicted driving cycle label into the time segment set In , the calculation process of the time decay weight of each time segment is as follows:

[0127]

[0128] Where λ is the time attenuation coefficient, λ>0, f ik is the time number of the kth time segment in the time segment set, and t is the time number of the current time segment.

[0129] The second calculation module 173 is used to perform weighted calculation based on the predicted driving style label and the time decay weight to obtain the average style score. Specifically, the second calculation module 173 is based on the time decay weight of each time segment. And the style score v corresponding to the predicted driving style label for each time segment ik Perform weighted calculation to get the average style score. Average style score avg_score i The calculation process of (t) is as follows:

[0130]

[0131] Among them, v ik Predict the style score corresponding to the driving style label for each time segment, is the time attenuation weight of each time segment, and k is the time number of the time segment.

[0132] When the stored number is equal to or greater than the preset number, second calculation module 180 is configured to perform a calculation based on the predicted driving cycle label and the predicted driving style label to obtain an entropy weight. Second calculation module 180 may include at least a third calculation module 181, a fourth calculation module 183, and a fifth calculation module 185.

[0133] The third calculation module 181 is configured to calculate, based on the predicted driving cycle labels and the predicted driving style labels, the frequency of occurrence of each type of driving style label. Specifically, the frequency of occurrence of each type of predicted driving style label is calculated within the time segment of each type of predicted driving cycle label to obtain the frequency of occurrence of each type of driving style label.

[0134] The fourth calculation module 183 is used to calculate based on the frequency of occurrence of each type of driving style label to obtain an entropy value. Specifically, the fourth calculation module 183 calculates based on the frequency of occurrence of each type of driving style label to obtain an entropy value. The entropy value reflects the degree of dispersion or uncertainty of the predicted driving style label in the time segment of each type of predicted driving cycle label. The smaller the entropy value, the more concentrated the cycle style is, the stronger the discrimination is, and the higher the weight is; the larger the entropy value, the more mixed the cycle style is, the weaker the discrimination is, and the lower the weight is. The calculation process of the entropy value is as follows:

[0135]

[0136] Among them, i is the type of predicted driving cycle label, k is the normalization coefficient, and s ij is the frequency of occurrence of the jth predicted driving style label in the time segment of the ith predicted driving cycle label.

[0137] The fifth calculation module 185 is used to calculate based on the entropy value to obtain the entropy weight. Specifically, the fifth calculation module 185 calculates based on the entropy value to obtain the entropy weight. The calculation process of the entropy weight is as follows:

[0138]

[0139] Among them, 1-E i The effective amount of information in the time segment of the driving cycle label predicted for the i-th class.

[0140] The style determination module 190 is configured to perform a weighted calculation based on the average style score and the entropy weight to obtain a target driving style score, and then determine the target driving style based on the target driving style score and a preset score. Specifically, the style determination module 190 performs a weighted summation based on the average style score and the entropy weight to obtain the target driving style score. The target driving style score is calculated as follows:

[0141]

[0142] When the target driving style score Score total >85, the target driving style is determined to be conservative; when 75<target driving style score total When the target driving style score is ≤85, the target driving style is determined to be conventional; when the target driving style score is total When ≤75, the target driving style is determined to be aggressive.

[0143] In summary, in a driving style recognition system based on the entropy weight method of the present application, the data processing module 110 processes the original driving data of the target vehicle and the artificial neural network model to obtain a predicted driving cycle label for each time segment and a predicted driving style label for each time segment. The data storage module 130 stores the predicted driving cycle label for each time segment and the predicted driving style label for each time segment, and records the time number corresponding to each time segment. The comparison module 150 counts the number of stored predicted driving cycle labels for each time segment or predicted driving style labels for each time segment, and determines whether the stored number is equal to or greater than the preset storage number. When the number is greater than the preset storage number, the comparison module 150 counts the number of stored predicted driving cycle labels for each time segment or predicted driving style labels for each time segment. When the storage quantity is equal to or greater than the preset storage quantity, the first calculation module 170 performs calculation based on the time number and the predicted driving style label to obtain an average style score. When the storage quantity is equal to or greater than the preset storage quantity, the second calculation module 180 performs calculation based on the predicted driving cycle label and the predicted driving style label to obtain an entropy weight. The style determination module 190 performs weighted calculation based on the average style score and the entropy weight to obtain a target driving style score, and determines the target driving style based on the target driving style score and the preset score. This not only significantly improves the objectivity and reproducibility of the score, but also can flexibly adapt to the actual characteristics of different drivers, different road conditions or time periods, and has strong environmental adaptability and generalization capabilities.

[0144] Throughout this specification, reference to terms such as "one embodiment," "some embodiments," "example," "specific example," "one implementation," "a preferred implementation," or "some examples" means that the specific features, structures, materials, or characteristics described in conjunction with that embodiment or example are included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0145] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to the embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the claims and their equivalents.

Claims

1. A driving style recognition method based on entropy weight method, characterized in that: The driving style recognition method based on the entropy weight method includes: Based on the original driving data of the target vehicle and the artificial neural network model, the predicted driving cycle label and the predicted driving style label of each time segment are obtained; storing the predicted driving cycle label and the predicted driving style label of each time segment, and recording the time number corresponding to each time segment; counting the number of the stored predicted driving cycle labels or the predicted driving style labels for each time segment, and determining whether the stored number is equal to or greater than a preset storage number; When the stored number is equal to or greater than the preset stored number, performing calculation based on the time number and the predicted driving style label to obtain an average style score; When the stored number is equal to or greater than the preset stored number, performing calculation based on the predicted driving cycle label and the predicted driving style label to obtain an entropy weight; A target driving style score is obtained by performing a weighted calculation based on the average style score and the entropy weight, and a target driving style is determined based on the target driving style score and a preset score.

2. The driving style identification method based on entropy weight method according to claim 1, characterized in that: The raw driving data of the target vehicle and the artificial neural network model are processed to obtain a predicted driving cycle label for each time segment and a predicted driving style label for each time segment, including: Acquiring raw driving data of the target vehicle in real time, and performing cleaning processing on the raw driving data to obtain target driving data, wherein the raw driving data includes training raw driving data and non-training raw driving data; Dividing the target driving data into time segments based on a preset driving duration to obtain target driving data of at least one time segment; Extracting features based on the target driving data of each time segment to obtain driving cycle features and driving style features of the corresponding time segment; Performing dimensionality reduction processing based on the driving cycle characteristics and the driving style characteristics to obtain driving principal component characteristics, wherein the driving principal component characteristics include training driving principal component characteristics and non-training driving principal component characteristics; Training the artificial neural network model based on the training driving principal component features to obtain a target driving cycle artificial neural network model and a target driving style artificial neural network model respectively; The target driving cycle artificial neural network model is called to perform identification processing based on the non-training driving principal component features to obtain the predicted driving cycle label, and the target driving style artificial neural network model is called to perform identification processing based on the non-training driving principal component features to obtain the predicted driving style label.

3. The driving style identification method based on entropy weight method according to claim 2, characterized in that: The artificial neural network model is trained based on the training driving principal component features to obtain a target driving cycle artificial neural network model and a target driving style artificial neural network model, respectively, including: Performing clustering processing based on the training driving principal component features to obtain corresponding real driving cycle labels and corresponding real driving style labels; The artificial neural network model is trained based on the training driving principal component features, the real driving cycle label, and the real driving style label to obtain a target driving cycle artificial neural network model and a target driving style artificial neural network model, respectively.

4. The driving style identification method based on entropy weight method according to claim 3, characterized in that: The artificial neural network model is trained based on the training driving principal component features, the real driving cycle label, and the real driving style label to obtain a target driving cycle artificial neural network model and a target driving style artificial neural network model, respectively, including: Obtaining a predicted probability based on the training driving principal component features; Calculating a target loss value based on the predicted probability, the actual driving cycle label, and the actual driving style label; In the direction of reducing the target loss value, the model parameters in the artificial neural network model are optimized to obtain the target driving cycle artificial neural network model and the target driving style artificial neural network model.

5. The driving style identification method based on entropy weight method according to claim 4 is characterized in that: When the stored number is equal to or greater than the preset stored number, performing calculation based on the time number and the predicted driving style label to obtain an average style score includes: Calculating based on the time number to obtain a time decay weight; A weighted calculation is performed based on the predicted driving style label and the time decay weight to obtain the average style score.

6. The driving style identification method based on entropy weight method according to claim 5, characterized in that: The performing weighted calculation based on the predicted driving style label and the time decay weight to obtain the average style score includes: Calculating based on the predicted driving cycle labels and the predicted driving style labels to obtain the frequency of occurrence of each type of driving style label; Calculate the frequency of each driving style label to obtain an entropy value; Calculation is performed based on the entropy value to obtain the entropy weight.

7. The driving style identification method based on entropy weight method according to claim 6, characterized in that: The target driving style score is obtained by performing weighted calculation based on the average style score and the entropy weight, and the target driving style score is: Among them, w i is the entropy weight, i is the type of predicted driving cycle label, avg_score i (t) is the average style score.

8. A driving style recognition system based on entropy weight method, characterized in that: The driving style recognition system based on the entropy weight method includes: a data processing module, a data storage module, a comparison module, a first calculation module, a second calculation module and a style determination module, wherein: The data processing module is used to process the original driving data of the target vehicle and the artificial neural network model to obtain the predicted driving cycle label and the predicted driving style label of each time segment; The data storage module is used to store the predicted driving cycle label and the predicted driving style label of each time segment, and record the time number corresponding to each time segment; The comparison module is configured to count the number of the stored predicted driving cycle labels or the predicted driving style labels for each time segment, and determine whether the stored number is equal to or greater than a preset storage number; When the stored number is equal to or greater than the preset stored number, the first calculation module is configured to perform calculation based on the time number and the predicted driving style label to obtain an average style score; When the storage quantity is equal to or greater than the preset storage quantity, the second calculation module is used to calculate based on the predicted driving cycle label and the predicted driving style label to obtain an entropy weight; The style determination module is configured to perform weighted calculation based on the average style score and the entropy weight to obtain a target driving style score, and determine the target driving style based on the target driving style score and a preset score.

9. The driving style recognition system based on entropy weight method according to claim 8, characterized in that: The data processing module includes a data acquisition module, a time segment division module, a feature extraction module, a feature processing module, a model training module and a label generation module, wherein: The data acquisition module is used to acquire the original driving data of the target vehicle in real time, and to clean the original driving data to obtain target driving data, wherein the original driving data includes training original driving data and non-training original driving data; The time segment division module is used to divide the target driving data into time segments based on a preset driving duration to obtain target driving data of at least one time segment; The feature extraction module is used to extract features based on the target driving data of each time segment to obtain driving cycle features and driving style features of the corresponding time segment; The feature processing module is used to perform dimensionality reduction processing based on the driving cycle feature and the driving style feature to obtain a driving principal component feature, wherein the driving principal component feature includes a training driving principal component feature and a non-training driving principal component feature; The model training module is used to train the artificial neural network model based on the training driving principal component characteristics to obtain a target driving cycle artificial neural network model and a target driving style artificial neural network model respectively; The label generation module is configured to call the target driving cycle artificial neural network model, perform identification processing based on the non-training driving principal component features to obtain the predicted driving cycle label, and call the target driving style artificial neural network model, perform identification processing based on the non-training driving principal component features to obtain the predicted driving style label.

10. The driving style recognition system based on entropy weight method according to claim 9, characterized in that: The first computing module includes a first computing module and a second computing module, wherein: The first calculation module is used to perform calculation based on the time number to obtain a time decay weight; The second calculation module is configured to perform weighted calculation based on the predicted driving style label and the time decay weight to obtain the average style score.

Images