Vehicle traveling condition prediction method, controller, vehicle and cloud server

Abstract

A vehicle traveling condition prediction method, a controller, a vehicle and a cloud server. The method comprises: on the basis of real-time traveling data of a target vehicle and a first vehicle traveling data sample set, obtaining a target first vehicle traveling data sample which matches the real-time traveling data (S11); and on the basis of the target first vehicle traveling data sample, predicting a target traveling condition of the target vehicle (S13).

Description

Vehicle driving condition prediction method, controller, vehicle and cloud server

[0001] This application claims priority to Chinese Patent Application No. 202411004530.3, filed on July 25, 2024, the entire contents of which are incorporated herein by reference. TECHNICAL FIELD

[0002] The present disclosure relates to the technical field of intelligent vehicles, and in particular to a vehicle driving condition prediction method, a controller, a vehicle and a cloud server. BACKGROUND

[0003] With the advancement of intelligent driving technology, it is particularly important to optimize the vehicle energy management strategy based on the vehicle driving condition. SUMMARY

[0004] The present disclosure provides a vehicle driving condition prediction method, a controller, a vehicle and a cloud server, which can solve the problem of limited application scenarios and low reliability in predicting the vehicle driving condition in the related art.

[0005] In a first aspect, a vehicle driving condition prediction method is provided, the method comprising:

[0006] Based on the real-time driving data of the target vehicle and the vehicle driving data sample set, at least one target vehicle driving data sample matching the real-time driving data is obtained, wherein the vehicle driving data sample set comprises a plurality of first vehicle driving data samples.

[0007] Based on the at least one target vehicle driving data sample, a target driving condition of the target vehicle is predicted.

[0008] In some embodiments, the method further comprises:

[0009] Based on the target driving condition, target driving condition speed data corresponding to the target driving condition is determined.

[0010] The target driving condition speed data is output, or a set of motion parameters predicted based on the target driving condition speed data is output.

[0011] In some embodiments, the vehicle driving data sample set comprises a plurality of sample sets, each sample set of the plurality of sample sets comprises at least two first vehicle driving data samples, each sample set has a corresponding condition label configured to identify the driving condition corresponding to the at least two first vehicle driving data samples in the each sample set, and the at least one target vehicle driving data sample comprises at least two target vehicle driving data samples.

[0012] The target driving condition of the target vehicle is predicted based on the at least one target vehicle driving data sample, including:

[0013] For each target vehicle driving data sample in the at least two target vehicle driving data samples, a sample set including the target vehicle driving data sample is determined;

[0014] The condition label corresponding to the target sample set is determined as the target driving condition of the target vehicle, wherein the target sample set is the sample set with the largest number of target vehicle driving data samples.

[0015] In some embodiments, the at least one target vehicle driving data sample matching the real-time driving data is obtained based on the real-time driving data and the vehicle driving data sample set, including:

[0016] The feature extraction is performed on the real-time driving data to obtain a feature vector corresponding to the real-time driving data;

[0017] Based on the similarity between the feature vector and the feature vector of each first vehicle driving data sample in the vehicle driving data sample set, the first vehicle driving data samples corresponding to the top N similarities in descending order of similarity are determined as the target vehicle driving data samples, wherein N is an integer greater than or equal to 2.

[0018] In some embodiments, the similarity is determined based on the Euclidean distance.

[0019] In some embodiments, the condition label is determined based on the relevant parameter information of the first vehicle driving data sample in each sample set, wherein the relevant parameter information includes at least one of the following: speed-related feature parameters and acceleration-related feature parameters.

[0020] In some embodiments, the plurality of first vehicle driving data samples are obtained by dividing historical vehicle driving data samples according to a preset time length; and the plurality of sample sets are obtained by performing clustering analysis on the plurality of first vehicle driving data samples based on the feature vectors corresponding to the plurality of first vehicle driving data samples.

[0021] In some embodiments, the plurality of sample sets are obtained by performing clustering analysis based on the feature vectors after standardizing each parameter of the feature vectors corresponding to the plurality of first vehicle driving data samples.

[0022] In some embodiments, the condition speed data corresponding to the driving condition is constructed based on historical vehicle driving data samples.

[0023] In some embodiments, the working condition speed data corresponding to the driving working condition is obtained by splicing second vehicle driving data samples, which are obtained by dividing the historical vehicle driving data samples according to vehicle speed.

[0024] In some embodiments, the working condition speed data corresponding to the driving working condition is obtained by splicing second vehicle driving data samples after screening the second vehicle driving data samples based on preset conditions, wherein the preset conditions include at least one of the following:

[0025] If the second vehicle driving data sample has data frame loss, the second vehicle driving data sample is deleted.

[0026] If the running time of the second vehicle driving data sample is less than a preset time length, the second vehicle driving data sample is deleted.

[0027] In some embodiments, the working condition speed data corresponding to the driving working condition is obtained by splicing second vehicle driving data samples belonging to the same driving working condition, wherein the driving working condition corresponding to the second vehicle driving data sample is determined based on a parameter threshold of the driving working condition, and the second vehicle driving data sample is obtained by screening the second vehicle driving data sample based on preset conditions.

[0028] In some embodiments, the parameter threshold of the driving working condition is determined based on related parameter information of the first vehicle driving data sample in each sample set.

[0029] In a second aspect, a vehicle driving working condition prediction device is provided, which includes a processing module and a prediction module.

[0030] The processing module is configured to obtain a target vehicle driving data sample matching real-time driving data of a target vehicle based on the real-time driving data and a vehicle driving data sample set, wherein the vehicle driving data sample set includes a plurality of first vehicle driving data samples.

[0031] The prediction module is configured to predict a target driving working condition of the target vehicle based on the target vehicle driving data sample.

[0032] In some embodiments, the prediction module is further configured to determine target driving working condition speed data corresponding to the target driving working condition based on the target driving working condition, and output the target driving working condition speed data or a group of motion parameters predicted based on the target driving working condition speed data.

[0033] In some embodiments, the vehicle driving data sample set comprises: a plurality of sample sets, each sample set in the plurality of sample sets comprising at least two first vehicle driving data samples, each sample set having a corresponding working condition label configured to identify a driving working condition corresponding to the first vehicle driving data samples in the sample set; and the at least one target vehicle driving data sample comprising at least two target vehicle driving data samples.

[0034] The prediction module is specifically configured to, for each target vehicle driving data sample, determine a sample set comprising the target vehicle driving data sample, and determine a working condition label corresponding to the target sample set as a target driving working condition of the target vehicle, the target sample set being a sample set comprising the target vehicle driving data sample in the largest quantity.

[0035] In some embodiments, the processing module is specifically configured to perform feature extraction on the real-time driving data to obtain a feature vector corresponding to the real-time driving data, and determine, according to similarities between the feature vector and feature vectors of each first vehicle driving data sample in the vehicle driving data sample set, in a descending order of the similarities, the first vehicle driving data samples corresponding to the top N similarities as the target vehicle driving data sample, N being an integer greater than or equal to 2.

[0036] In some embodiments, the similarity is determined based on Euclidean distance.

[0037] In some embodiments, the working condition label is determined based on relevant parameter information of the first vehicle driving data samples in each sample set, wherein the relevant parameter information comprises at least one of a speed-related feature parameter and an acceleration-related feature parameter.

[0038] In some embodiments, the plurality of first vehicle driving data samples are obtained by dividing historical vehicle driving data samples according to a preset time length, and the plurality of sample sets are obtained by performing clustering analysis on the plurality of first vehicle driving data samples based on feature vectors corresponding to the plurality of first vehicle driving data samples.

[0039] In some embodiments, the plurality of sample sets are obtained by performing clustering analysis based on the feature vectors after performing standardization processing on each parameter of the feature vectors corresponding to the plurality of first vehicle driving data samples.

[0040] In some embodiments, the working condition speed data corresponding to the driving working condition is constructed based on historical vehicle driving data samples.

[0041] In some embodiments, the working condition speed data corresponding to the driving working condition is obtained by splicing second vehicle driving data samples, which are obtained by dividing the historical vehicle driving data samples according to vehicle speed.

[0042] In some embodiments, the working condition speed data corresponding to the driving working condition is obtained by splicing second vehicle driving data samples after screening the second vehicle driving data samples according to preset conditions, wherein the preset conditions include at least one of the following:

[0043] If the second vehicle driving data sample has data frame loss, the second vehicle driving data sample is deleted.

[0044] If the running time of the second vehicle driving data sample is less than a preset time length, the second vehicle driving data sample is deleted.

[0045] In some embodiments, the working condition speed data corresponding to the driving working condition is obtained by splicing second vehicle driving data samples belonging to the same driving working condition, the driving working condition corresponding to the second vehicle driving data sample is determined based on a parameter threshold of the driving working condition, and the second vehicle driving data sample is obtained by screening the second vehicle driving data sample according to preset conditions.

[0046] In some embodiments, the parameter threshold of the driving working condition is determined based on the related parameter information of the first vehicle driving data sample in each sample set.

[0047] In a third aspect, a controller is provided, comprising a processor and a memory, wherein the memory stores programs or instructions executable on the processor, and the programs or instructions are executed by the processor to implement the vehicle driving working condition prediction method according to any one of the first aspect.

[0048] In a fourth aspect, a vehicle is provided, comprising a processor and a memory, wherein the memory stores programs or instructions executable on the processor, and the programs or instructions are executed by the processor to implement the vehicle driving working condition prediction method according to the first aspect.

[0049] In a fifth aspect, a cloud server is provided, comprising a processor and a memory, wherein the memory stores programs or instructions executable on the processor, and the programs or instructions are executed by the processor to implement the vehicle driving working condition prediction method according to the first aspect.

[0050] In a sixth aspect, a computer readable storage medium is provided, and the computer readable storage medium stores a program or instructions, and the program or instructions are executed by a processor to implement the vehicle driving condition prediction method according to the first aspect.

[0051] In a seventh aspect, a computer program product is provided, and the computer program product is executed by a processor of a vehicle or a cloud server to implement the vehicle driving condition prediction method according to the first aspect.

[0052] The vehicle driving condition prediction method, the controller, the vehicle and the cloud server provided by some embodiments of the present disclosure are based on real-time driving data of a target vehicle and a vehicle driving data sample set to obtain a target vehicle driving data sample matched with the real-time driving data, and based on the target vehicle driving data sample, a target driving condition of the target vehicle is predicted. That is, some embodiments of the present disclosure predict the target driving condition of the target vehicle through the real-time driving data of the target vehicle, and therefore, the reliability is improved without being limited by a scene. BRIEF DESCRIPTION OF DRAWINGS

[0053] FIG. 1 is a flowchart of a vehicle driving condition prediction method according to some embodiments of the present disclosure;

[0054] FIG. 2 is a schematic diagram of real-time driving data according to some embodiments of the present disclosure;

[0055] FIG. 3 is a flowchart of another vehicle driving condition prediction method according to some embodiments of the present disclosure;

[0056] FIG. 4 is a flowchart of still another vehicle driving condition prediction method according to some embodiments of the present disclosure;

[0057] FIG. 5 is a flowchart of still another vehicle driving condition prediction method according to some embodiments of the present disclosure;

[0058] FIG. 6 is a flowchart of still another vehicle driving condition prediction method according to some embodiments of the present disclosure;

[0059] FIG. 7 is a flowchart of still another vehicle driving condition prediction method according to some embodiments of the present disclosure;

[0060] FIG. 8A is a schematic diagram of condition speed data according to some embodiments of the present disclosure;

[0061] FIG. 8B is another schematic diagram of condition speed data according to some embodiments of the present disclosure;

[0062] FIG. 8C is still another schematic diagram of condition speed data according to some embodiments of the present disclosure;

[0063] FIG. 8D is still another schematic diagram of condition speed data according to some embodiments of the present disclosure;

[0064] FIG. 8E is another schematic diagram of work condition speed data according to some embodiments of the present disclosure;

[0065] FIG. 9 is a block diagram of a vehicle driving condition prediction device according to some embodiments of the present disclosure;

[0066] FIG. 10 is a block diagram of a controller according to some embodiments of the present disclosure;

[0067] FIG. 11 is a block diagram of a vehicle according to some embodiments of the present disclosure;

[0068] FIG. 12 is a block diagram of a cloud server according to some embodiments of the present disclosure. DETAILED DESCRIPTION

[0069] The technical solutions in some embodiments of the present disclosure will be clearly described below with reference to the accompanying drawings in some embodiments of the present disclosure. Obviously, the described embodiments are part of the embodiments of the present disclosure, rather than all the embodiments. Based on some embodiments of the present disclosure, all other embodiments obtained by those of ordinary skill in the art belong to the scope of protection of the present disclosure.

[0070] In order to optimize the whole vehicle energy management strategy, in the related art, by establishing the correspondence between the traffic and road parameters and the vehicle driving condition, based on the driving route selected by the user, the traffic and road parameters corresponding to the driving route are obtained through high-precision maps and vehicle networks, the vehicle driving condition is predicted by relying on high-precision maps and vehicle networks, and then the whole vehicle energy management strategy is optimized based on the vehicle driving condition. However, the related art relies heavily on high-precision maps and vehicle networks, and in some scenarios, such as when the driver does not turn on the navigation mode, does not select a specific driving route, the map accuracy of the vehicle is not enough, the map is not updated in time, or the vehicle is in a network-out state, the vehicle driving condition cannot be predicted, which limits the prediction of the vehicle driving condition and reduces the reliability.

[0071] In order to reduce the limitation of the prediction of the vehicle driving condition by the scene and improve the reliability, some embodiments of the present disclosure provide a vehicle driving condition prediction method, which can predict the vehicle driving condition based on the real-time driving data of the vehicle.

[0072] Further, in order to improve the optimization effect of the vehicle energy management strategy, some embodiments of the present disclosure further construct driving condition speed data corresponding to different vehicle driving conditions. The driving condition speed data can be represented in the form of a driving condition speed curve. After predicting the vehicle driving condition, the future driving speed and acceleration of the vehicle and other motion parameters can be predicted based on the driving condition speed data corresponding to the vehicle driving condition, or a set of motion parameters predicted based on the driving condition speed data can be directly output. The motion parameters can be output in sequence, which can better optimize the vehicle energy management strategy.

[0073] The steps of the above-mentioned embodiments of the present disclosure can be executed by a vehicle, a cloud server, or a combination of a vehicle and a cloud server. The combination of a vehicle and a cloud server includes but is not limited to the following cases: in some cases, the vehicle is in an off-network state, and the steps are executed by the vehicle alone; in some cases, the vehicle is in an on-network state, and the steps are executed by the cloud server; in some cases, the cloud server sends the execution result to the vehicle controller, so that the vehicle controller controls the vehicle based on the execution result. The present disclosure does not limit the execution subject, which can be determined based on the actual situation such as the computing power of the vehicle, the application scenario of the vehicle, and the type of the vehicle.

[0074] The vehicle driving conditions related to some embodiments of the present disclosure include but are not limited to urban congestion, urban semi-congestion, urban free-flow, suburban free-flow, suburban semi-congestion, suburban congestion, highway free-flow, highway congestion, and highway semi-congestion.

[0075] Some embodiments of the present disclosure are described below.

[0076] FIG. 1 is a flowchart of a vehicle driving condition prediction method according to some embodiments of the present disclosure. As shown in FIG. 1, the method comprises the following steps:

[0077] S11: obtaining target vehicle driving data samples matched with the real-time driving data of the target vehicle based on the real-time driving data of the target vehicle and the vehicle driving data sample set.

[0078] The target vehicle refers to a vehicle whose driving condition is to be predicted.

[0079] The real-time driving data of the target vehicle refers to the driving data in a continuous time period before the current time. For example, if the time period is set to 5 minutes and the current time is 5:00 pm, the real-time driving data obtained is the driving data in the time period of 4:55-5:00. Alternatively, the real-time driving data of the target vehicle can be the driving data in a continuous time period before a time close to the current time, for example, 4:54-4:59.

[0080] The real-time driving data includes speed information. The real-time driving data can be in the form of a driving speed-time curve, for example. As shown in FIG. 2, which is a schematic diagram of real-time driving data according to some embodiments of the present disclosure, the horizontal axis represents time information, and the vertical axis represents speed information. In some embodiments, the real-time driving data in this step can also be represented in the form of speed, acceleration, and other motion parameters, that is, the speed, acceleration, and other motion parameters corresponding to the driving speed-time curve are directly obtained.

[0081] In some embodiments, the real-time driving data of the target vehicle can be obtained by a speed sensor.

[0082] In some embodiments, the real-time driving data can be matched with the vehicle driving data sample set to obtain target vehicle driving data samples matched with the real-time driving data.

[0083] In some embodiments, the vehicle driving data sample set can include, but is not limited to, the following possible implementation manners:

[0084] One implementation manner is that the vehicle driving data sample set includes a plurality of first vehicle driving data samples, and the driving condition corresponding to each first vehicle driving data sample is known. For example, when the first vehicle driving data sample is collected, the corresponding perception information is also collected at the same time, and the perception information can represent the driving condition corresponding to the first vehicle driving data sample.

[0085] Another implementation manner is that the vehicle driving data sample set includes a plurality of sample sets, each sample set includes a plurality of first vehicle driving data samples, and each sample set has a corresponding condition label, which is used to identify the driving condition corresponding to the plurality of first vehicle driving data samples in the sample set. That is, the driving conditions of the plurality of first vehicle driving data samples included in the same sample set are the same, and the driving condition of the first vehicle driving data sample can be determined based on the condition label corresponding to the sample set to which the first vehicle driving data sample belongs. The condition label of the sample set can be determined based on the motion parameter information of the plurality of first vehicle driving data samples in the sample set.

[0086] One implementation manner of this step is described in FIG. 3, which is a flowchart of another vehicle driving condition prediction method according to some embodiments of the present disclosure. FIG. 3 is a description of how to perform matching to determine the target vehicle driving data sample matched with the real-time driving data based on the embodiment shown in FIG. 1:

[0087] S111: Feature extraction is performed on the real-time driving data to obtain a feature vector corresponding to the real-time driving data.

[0088] The feature extraction on the real-time driving data can be the extraction of motion features, which include but are not limited to at least one of the following: speed-related feature parameters, acceleration-related feature parameters, distance-related feature parameters, and the like.

[0089] For example, the speed-related feature parameters include but are not limited to at least one of the following: average vehicle speed, maximum vehicle speed, vehicle speed standard deviation, minimum vehicle speed, real-time vehicle speed, and the like.

[0090] The acceleration-related feature parameters include but are not limited to at least one of the following: acceleration change rate, average acceleration, maximum acceleration, minimum acceleration, and acceleration standard deviation.

[0091] The distance-related feature parameters include but are not limited to at least one of the following: total number of stops, average number of stops per kilometer, and driving distance, and the like.

[0092] Based on the extracted motion feature parameters, a feature vector corresponding to the real-time driving data is formed. For example, if the average vehicle speed, maximum vehicle speed, vehicle speed standard deviation, minimum vehicle speed, real-time vehicle speed, acceleration change rate, average acceleration, maximum acceleration, minimum acceleration, acceleration standard deviation, total number of stops, average number of stops per kilometer, and driving distance are extracted, then the feature vector is a 13-dimensional feature vector, and the 13 parameters in the feature vector correspond one-to-one to the above 13 feature parameters. The parameters included in the feature vector need to be consistent with the parameters included in the feature vector of the first vehicle driving data sample.

[0093] S112: Based on the similarity between the feature vector and the feature vector of each first vehicle driving data sample in the vehicle driving data sample set, the first vehicle driving data samples corresponding to the top N similarities in descending order of similarity are determined as the target vehicle driving data samples, and N is an integer greater than or equal to 2.

[0094] In some embodiments, the number of target vehicle driving data samples can be preset, and the number of target first vehicle data samples is at least 2, for example, it can be set to 5. Then, based on the similarity between the feature vector and the feature vector of each first vehicle driving data sample in the vehicle driving data sample set, the top 5 first vehicle driving data samples with the highest similarity are determined as the target vehicle driving data samples.

[0095] The feature vector and each first vehicle driving data sample can obtain a similarity value by calculation. The first vehicle driving data samples corresponding to the top 5 feature vectors in terms of similarity are obtained as the target vehicle driving data samples.

[0096] In some embodiments, the similarity can be determined based on the Euclidean distance, and the smaller the Euclidean distance between two feature vectors, the higher the similarity.

[0097] S13: predicting the target driving condition of the target vehicle based on the target vehicle driving data sample.

[0098] This step includes but is not limited to the following implementation manners:

[0099] One implementation manner is that if the driving condition of the first vehicle driving data sample in the vehicle driving data sample set is known, for each target vehicle driving data sample, the condition label corresponding to the target vehicle driving data sample is determined as the candidate driving condition of the target vehicle, so that at least two candidate driving conditions can be obtained, and the same candidate driving condition with the largest number in the at least two candidate driving conditions is determined as the target driving condition of the target vehicle. For example, 5 target vehicle driving data samples are obtained through similarity calculation, 2 target vehicle driving data samples correspond to driving condition 1, 1 target vehicle driving data sample corresponds to driving condition 2, 1 target vehicle driving data sample corresponds to driving condition 3, and 1 target vehicle driving data sample corresponds to driving condition 4. Driving conditions 1, 2, 3 and 4 are all candidate driving conditions, the number of target vehicle driving data samples corresponding to driving condition 1 is 2, and driving condition 1 is determined as the target driving condition of the target vehicle.

[0100] Another implementation manner is that for each target vehicle driving data sample, a sample set including the target vehicle driving data sample is determined, and the condition label corresponding to the target sample set is determined as the target driving condition of the target vehicle, and the target sample set is the sample set including the largest number of target vehicle driving data samples. In this implementation manner, the driving condition of the target vehicle driving data sample is determined based on the condition label of the sample set including the sample. For example, 5 target vehicle driving data samples are obtained through similarity calculation, sample set 1 includes 2 target vehicle driving data samples, the condition label of sample set 1 is driving condition 1, sample set 2 includes 1 target vehicle driving data sample, the driving condition corresponding to sample set 2 is driving condition 2, sample set 3 includes 1 target vehicle driving data sample, the driving condition corresponding to sample set 3 is driving condition 3, sample set 4 includes 1 target vehicle driving data sample, and the driving condition corresponding to sample set 4 is driving condition 4. Since sample set 1 includes 2 target vehicle driving data samples, it is determined that sample set 1 includes the largest number of target vehicle driving data samples, and sample set 1 is determined as the target sample set. The driving condition 1 corresponding to sample set 1 is the target driving condition of the target vehicle.

[0101] In some embodiments of the present disclosure, a target vehicle driving data sample matching the real-time driving data of the target vehicle is obtained based on the real-time driving data of the target vehicle and the vehicle driving data sample set, and a target driving condition of the target vehicle is predicted based on the target vehicle driving data sample. Therefore, in some embodiments of the present disclosure, the target driving condition of the target vehicle is predicted based on the real-time driving data of the target vehicle, and thus the reliability is improved without being limited by the scene.

[0102] FIG. 4 is a flowchart of another vehicle driving condition prediction method according to some embodiments of the present disclosure. FIG. 4 further describes the manner of obtaining the vehicle driving data sample set based on the embodiment shown in FIG. 1. The plurality of first vehicle driving data samples are obtained by dividing the historical vehicle driving data sample according to a preset time length. The plurality of sample sets are obtained by performing clustering analysis on the plurality of first vehicle driving data samples based on the feature vectors corresponding to the plurality of first vehicle driving data samples. The process is as follows:

[0103] S31: Obtain a historical vehicle driving data sample, and divide the historical vehicle driving data sample according to a preset time length to obtain a plurality of first vehicle driving data samples.

[0104] The historical vehicle driving data sample can be obtained by a device such as a vehicle event data recorder. The historical vehicle driving data sample can be represented in the form of a speed-time curve.

[0105] Since the historical vehicle driving data sample is usually a long period of driving record of a vehicle, it may include scenes of multiple vehicle driving conditions at the same time. The number of vehicle driving conditions included in each sample can be reduced by dividing the segments, such as short trip condition segment division, so as to facilitate subsequent clustering analysis. For example, the historical vehicle driving data sample is divided according to a preset time length, so that a plurality of first vehicle driving data samples can be obtained.

[0106] S32: Perform clustering analysis on the plurality of first vehicle driving data samples based on the feature vectors corresponding to the plurality of first vehicle driving data samples to obtain a plurality of sample sets.

[0107] An implementation of this step is shown in FIG. 5, which is a flowchart of another vehicle driving condition prediction method according to some embodiments of the present disclosure.

[0108] S321: Obtain the feature vectors corresponding to the plurality of first vehicle driving data samples.

[0109] The feature extraction is performed on each first vehicle driving data sample, and the motion characteristic parameters are extracted. Taking three groups of motion characteristic parameters as examples, the characteristic parameters related to the vehicle speed, the characteristic parameters related to the acceleration, and the characteristic parameters related to the distance are taken as examples. The characteristic parameters related to the vehicle speed are set to five, which are the average vehicle speed, the maximum vehicle speed, the vehicle speed standard deviation, the minimum vehicle speed, and the real-time vehicle speed. The characteristic parameters related to the acceleration are set to five, which are the acceleration change rate, the average acceleration, the maximum acceleration, the minimum acceleration, and the acceleration standard deviation. The characteristic parameters related to the distance are set to three, which are the total number of stops, the average number of stops per kilometer, and the driving distance. In this way, the feature vector corresponding to each first vehicle driving data sample is a 13-dimensional feature vector, and the 13 parameters in the feature vector correspond to the 13 characteristic parameters described above one by one. For example, the feature vector is [x 11 , x 12 , x 13 , x 14 , x 15 , x 16 , x 17 , x 18 , x 19 , x 110 , x 111 , x 112 , x 113 ]. Then, x 11 corresponds to the average vehicle speed, x 12 corresponds to the maximum vehicle speed, x 13 corresponds to the vehicle speed standard deviation, x 14 corresponds to the minimum vehicle speed, x 15 corresponds to the real-time vehicle speed, x 16 corresponds to the acceleration change rate, x 17 corresponds to the average acceleration, x 18 corresponds to the maximum acceleration, x 19 corresponds to the minimum acceleration, x 110 corresponds to the acceleration standard deviation, x 111 corresponds to the total number of stops, x 112 corresponds to the average number of stops per kilometer, and x 113 corresponds to the driving distance.

[0110] The plurality of sample sets are obtained by performing standardization processing on each parameter of the feature vectors corresponding to the plurality of first vehicle driving data samples, and performing clustering analysis based on the feature vectors after the standardization processing, such as S322.

[0111] S322: Perform standardization processing on each parameter in the feature vector, and perform clustering analysis based on the feature vector after the standardization processing to obtain a plurality of sample sets.

[0112] Since the number of each parameter is different in dimension, before clustering analysis, each parameter in the feature vector can be standardized, for example, Z-score standardization can be performed:

[0113] wherein y ij represents the standardized data of the jth parameter of the ith first vehicle driving data sample, x ij represents the original data of the jth parameter of the ith first vehicle driving data sample, represents the average value of the original data of the jth parameter of all first vehicle driving data samples, n represents the number of first vehicle driving data samples, and m represents the number of parameters contained in the feature vector of each first vehicle driving data sample. In some embodiments of the present disclosure, m is exemplified as 13.

[0114] wherein σ represents the standard deviation of the original data, the average value and the standard deviation σ are calculated as follows:

[0115] By standardizing each parameter in the feature vector in the above manner, the clustering analysis can be more accurately performed. Then, based on the standardized feature vector, clustering analysis is performed to obtain a plurality of sample sets.

[0116] In some embodiments, an unsupervised clustering analysis method can be used, for example, a k-means algorithm is used for clustering analysis, data classification is performed by calculating the closeness between first vehicle driving data samples, and the clustering effect will realize that the first vehicle driving data samples in the same class have great feature similarity, while the features between different classes have obvious difference and interface. For example, the process of clustering analysis is as follows:

[0117] 1) Select k feature vectors of first vehicle driving data samples as initial clustering centers from the vehicle driving data sample set, k can be selected according to actual application, for example, k=5, the value of k determines how many classes the samples are divided into, the greater the value of k, the more the samples are classified, the more detailed the classification, the smaller the value of k, the fewer the samples are classified, the more rough the classification, the value of k can be determined based on actual application scenarios. The k feature vectors of first vehicle driving data samples here can be standardized feature vectors.

[0118] 2) Find the clustering center closest to the feature vector y i by calculating the Euclidean distance, the calculation method of the Euclidean distance between two vectors is as follows:

[0119] wherein y i represents the standardized data of the jth parameter of the ith first vehicle driving data sample, xi the feature vector of the i-th first vehicle driving data sample after standardization of the parameters, y iz denotes the standardized vector y i corresponding to the z-th cluster center, y z the Euclidean distance between the standardized vector y ij denotes the standardized data of the j-th parameter of the i-th first vehicle driving data sample, y zj denotes the standardized data of the j-th parameter of the z-th cluster center.

[0120] Through this step, the Euclidean distances of y i to each cluster center can be calculated, and y i is divided into the classification of the cluster center with the smallest Euclidean distance.

[0121] 3) Repeat the iterative calculation and update the cluster center of each class until the cluster center of each class no longer changes, and obtain a plurality of sample sets.

[0122] The working condition label is determined based on the relevant parameter information of the first vehicle driving data sample in each sample set, and the relevant parameter information includes at least one of the following: speed-related feature parameters and acceleration-related feature parameters. The S323 can be used to achieve:

[0123] S323: Based on the relevant parameter information of the first vehicle driving data sample in each sample set, obtain the working condition label of the sample set.

[0124] The relevant parameter information includes at least one of the following: speed-related feature parameters and acceleration-related feature parameters. The speed-related feature parameters include: low speed, medium speed and high speed, and the acceleration-related feature parameters include: low acceleration, medium acceleration and high acceleration.

[0125] After obtaining the relevant parameter information of the first vehicle driving data sample in each sample set, the working condition label of the sample set is obtained based on the experience value. For example, the average speed of the first vehicle driving data sample in the sample set is 120km / h, which is high speed, and the average acceleration is close to 0m / s 2 , that is, the speed is at high speed and the acceleration is at low acceleration, it can be considered that the driving working condition is high-speed smooth, and the working condition label corresponding to the sample set is high-speed smooth.

[0126] Further, the parameter threshold of the driving condition can be obtained based on the related parameter information of the first vehicle driving data sample in each sample set. In combination with the foregoing example, the minimum speed of the first vehicle driving data sample in the sample set is 90 km / h, the maximum speed is 130 km / h, the maximum acceleration is 2 m / s 2 to 4 m / s 2 , and the minimum acceleration is -1 m / s 2 to -3 m / s 2 .

[0127] In some embodiments of the present disclosure, by obtaining the feature vectors corresponding to the plurality of first vehicle driving data samples, performing standardization processing on each parameter in the feature vectors, performing clustering analysis based on the standardized feature vectors, obtaining a plurality of sample sets, and obtaining the condition label of each sample set based on the related parameter information of the first vehicle driving data sample in each sample set, the sample set is obtained, and the sample set is labeled, the condition label of each sample set is determined, and a data basis is provided for online vehicle driving condition prediction.

[0128] FIG. 6 is a flowchart of another vehicle driving condition prediction method according to some embodiments of the present disclosure. FIG. 6 is based on any of the embodiments shown in FIGS. 1-5, and the method further includes outputting target driving condition speed data, or outputting a set of motion parameters predicted based on the target driving condition speed data, to facilitate vehicle energy management strategy optimization, as follows:

[0129] S14: determining target driving condition speed data corresponding to the target driving condition based on the target driving condition.

[0130] The correspondence between the driving condition and the driving condition speed data can be established in advance, so that after the target driving condition is determined, the target driving condition speed data corresponding to the target driving condition can be obtained.

[0131] In some embodiments, the correspondence between the driving condition and the driving condition speed data can be achieved by the following implementation manners. One implementation manner is as follows:

[0132] The condition speed data corresponding to the driving condition is constructed based on historical vehicle driving data samples. Based on the historical vehicle driving data samples, the condition speed data corresponding to the driving condition is constructed. The sample set can be the sample set obtained after clustering analysis by the above embodiments.

[0133] The driving condition speed data corresponding to the driving condition is obtained by splicing the second vehicle driving data samples, which are obtained by dividing the historical vehicle driving data samples according to vehicle speed. For example, as shown in FIG. 7, FIG. 7 is a flowchart of another vehicle driving condition prediction method according to some embodiments of the present disclosure.

[0134] S71: dividing the historical vehicle driving data samples according to vehicle speed to obtain second vehicle driving data samples.

[0135] The principle of dividing according to vehicle speed is that the vehicle speed Vte at the travel end point te of the second vehicle driving data samples obtained by dividing is 0, and the vehicle speed Vte-1 at the last second te-1 of the travel end point is greater than 0.

[0136] S72: splicing the second vehicle driving data samples to obtain driving condition speed data corresponding to the driving condition.

[0137] In some embodiments, the second vehicle driving data samples can be spliced in a random splicing manner to obtain driving condition speed data corresponding to the driving condition.

[0138] In some embodiments, in order to improve the quality of the constructed driving condition speed data, the driving condition speed data corresponding to the driving condition can also be filtered based on a preset condition, and the second vehicle driving data samples obtained by splicing the filtered second vehicle driving data samples can also be filtered based on a preset condition. The preset condition includes at least one of the following:

[0139] If the second vehicle driving data sample has data frame loss, the second vehicle driving data sample is deleted.

[0140] If the running time of the second vehicle driving data sample is less than a preset time length, the second vehicle driving data sample is deleted, and the filtered second vehicle driving data samples are spliced to obtain driving condition speed data corresponding to the driving condition.

[0141] For example, the second vehicle driving data samples are filtered based on preset conditions to obtain filtered second vehicle driving data samples; and a driving condition corresponding to the second vehicle driving data samples is determined based on a parameter threshold of the driving condition. The parameter threshold of the driving condition can be the parameter threshold of the driving condition in some embodiments described above. After the condition label of the sample set is determined, the condition speed data corresponding to the driving condition can also be obtained by splicing the second vehicle driving data samples belonging to the same driving condition. The driving condition corresponding to the second vehicle driving data samples is determined based on the parameter threshold of the driving condition. The second vehicle driving data samples are obtained by filtering the second vehicle driving data samples based on preset conditions. The parameter threshold of the driving condition is determined based on the related parameter information of the first vehicle driving data samples in each sample set. For details, see the description in some embodiments described above. The condition speed data corresponding to the driving condition is obtained by splicing the second vehicle driving data samples belonging to the same driving condition. As shown in FIGS. 8A to 8E, FIGS. 8A to 8E are a plurality of schematic diagrams of condition speed data according to some embodiments of the present disclosure.

[0142] By dividing the historical vehicle driving data based on the vehicle speed as described above, the second vehicle driving data samples are obtained. The driving condition of the second vehicle driving data samples is determined based on the parameter threshold of the driving condition, which can improve the accuracy of the driving condition classification. Therefore, the second vehicle driving data samples belonging to the same driving condition are spliced based on this to obtain the condition speed data corresponding to the driving condition. The condition speed data can more accurately reflect the motion parameter characteristics of the corresponding driving condition.

[0143] S15: output the target driving condition speed data, or a group of motion parameters predicted based on the target driving condition speed data.

[0144] For example, the target driving condition speed data can be output, or a group of motion parameters can be predicted based on the target driving condition speed data. The motion parameters include speed information and acceleration information, etc. The group of motion parameters predicted is the motion parameter at a future time.

[0145] In some embodiments, by determining the target driving condition speed data corresponding to the target driving condition based on the target driving condition, outputting the target driving condition speed data, or predicting a group of motion parameters based on the target driving condition speed data, the energy management strategy of the vehicle is optimized.

[0146] In some embodiments described above, the examples shown in FIG. 1-3 and the embodiments shown in FIG. 6, FIG. 8A-8E can be executed by an online identification module running an online identification model, which can be disposed in the vehicle, in the cloud server, or in both the vehicle and the cloud server, to predict the driving condition of the vehicle and obtain the driving condition speed data, so as to optimize the energy management strategy of the vehicle. The embodiments shown in FIG. 4 and FIG. 5 can be executed by a condition construction module, which is used to construct the online identification model. The condition construction module can execute the steps of some embodiments described above offline. The condition construction module can be disposed in the vehicle or not. The condition construction module disposed in the vehicle can obtain new sample data, update the online identification model based on the new sample data, and further improve the accuracy of the online identification model.

[0147] FIG. 9 is a block diagram of a vehicle driving condition prediction device according to some embodiments of the present disclosure. As shown in FIG. 9, the vehicle driving condition prediction device 900 includes a processing module 901 and a prediction module 902. The processing module 901 is configured to obtain a target vehicle driving data sample matching real-time driving data of a target vehicle based on the real-time driving data and a vehicle driving data sample set, and the vehicle driving data sample set includes a plurality of first vehicle driving data samples. The prediction module 902 is configured to predict a target driving condition of the target vehicle based on the target vehicle driving data sample.

[0148] In some embodiments, the prediction module 902 is further configured to determine target driving condition speed data corresponding to the target driving condition based on the target driving condition; and output the target driving condition speed data, or output a set of motion parameters predicted based on the target driving condition speed data.

[0149] In some embodiments, the vehicle driving data sample set includes a plurality of sample sets, each sample set including a plurality of first vehicle driving data samples, and each sample set has a corresponding condition label identifying the driving condition corresponding to the first vehicle driving data samples in the sample set. The number of target vehicle driving data samples includes at least two.

[0150] In some embodiments, the prediction module 902 is further configured to determine, for each target vehicle driving data sample, a sample set including the target vehicle driving data sample; and determine the condition label corresponding to the target sample set as the target driving condition of the target vehicle, wherein the target sample set is the sample set including the largest number of target vehicle driving data samples.

[0151] In some embodiments, the processing module 901 is further configured to perform feature extraction on the real-time driving data, obtain a feature vector corresponding to the real-time driving data; and determine, based on similarities between the feature vector and feature vectors of each first vehicle driving data sample in the vehicle driving data sample set, the first vehicle driving data samples corresponding to the top N similarities as the target vehicle driving data samples in descending order of the similarities, where N is an integer greater than or equal to 2.

[0152] In some embodiments, the similarity is determined based on Euclidean distance.

[0153] In some embodiments, the working condition label is determined based on relevant parameter information of the first vehicle driving data sample in each sample set, and the relevant parameter information includes at least one of a speed-related feature parameter and an acceleration-related feature parameter.

[0154] In some embodiments, the plurality of first vehicle driving data samples are obtained by dividing historical vehicle driving data samples according to a preset time length; and the plurality of sample sets are obtained by performing clustering analysis on the plurality of first vehicle driving data samples based on feature vectors corresponding to the plurality of first vehicle driving data samples.

[0155] In some embodiments, the plurality of sample sets are obtained by performing clustering analysis based on standardized feature vectors, by performing standardization processing on each parameter of the feature vectors corresponding to the plurality of first vehicle driving data samples.

[0156] In some embodiments, the working condition speed data corresponding to the driving working condition is constructed based on historical vehicle driving data samples.

[0157] In some embodiments, the working condition speed data corresponding to the driving working condition is obtained by splicing second vehicle driving data samples, which are obtained by dividing the historical vehicle driving data samples according to vehicle speed.

[0158] In some embodiments, the working condition speed data corresponding to the driving working condition is obtained by splicing second vehicle driving data samples after filtering the second vehicle driving data samples based on a preset condition, and the preset condition includes at least one of:

[0159] If the second vehicle driving data sample has data frame loss, the second vehicle driving data sample is deleted.

[0160] If the running time of the second vehicle driving data sample is less than a preset time length, the second vehicle driving data sample is deleted.

[0161] In some embodiments, the driving condition corresponding to the driving condition speed data is obtained by splicing second vehicle driving data samples belonging to the same driving condition, the driving condition corresponding to the second vehicle driving data samples is determined based on a parameter threshold of the driving condition, and the second vehicle driving data samples are obtained by screening the second vehicle driving data samples based on a preset condition.

[0162] In some embodiments, the parameter threshold of the driving condition is determined based on the related parameter information of the first vehicle driving data samples in each sample set.

[0163] The device corresponding to some embodiments of the present disclosure can be used to execute the technical solutions of the above-mentioned method embodiments, and the implementation principles and technical effects are similar, which will not be described here.

[0164] As shown in FIG. 10, some embodiments of the present disclosure also provide a controller 1000, which includes a processor and a memory, the memory stores programs or instructions executable on the processor, and the programs or instructions are executed by the processor to implement the vehicle driving condition prediction method described above.

[0165] As shown in FIG. 11, some embodiments of the present disclosure also provide a vehicle 2000, which includes a processor and a memory, the memory stores programs or instructions executable on the processor, and the programs or instructions are executed by the processor to implement the vehicle driving condition prediction method described above.

[0166] As shown in FIG. 12, some embodiments of the present disclosure also provide a cloud server 3000, which includes a processor and a memory, the memory stores programs or instructions executable on the processor, and the programs or instructions are executed by the processor to implement the vehicle driving condition prediction method described above.

[0167] Some embodiments of the present disclosure also provide a computer readable storage medium, the readable storage medium stores programs or instructions, and the programs or instructions are executed by the processor to implement the vehicle driving condition prediction method described above.

[0168] Some embodiments of the present disclosure also provide a computer program product, which is executed by the processor of the vehicle or the cloud server to implement the vehicle driving condition prediction method described above.

[0169] It should be noted that, in the present document, the terms "comprising", "including", or any other variant thereof are intended to cover a non-exclusive inclusion, such that processes, methods, articles, or apparatuses that comprise a list of elements are not limited to those elements, but can also include other elements not expressly listed, or can also include elements inherent in such processes, methods, articles, or apparatuses. Without further limitation, an element defined by the phrase "comprising a" does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes the element. In addition, it should be noted that the scope of the methods and apparatuses in the present disclosure is not limited to performing functions in the order shown or discussed, but can also include performing functions in a substantially simultaneous manner or in a reverse order, for example, the described methods can be performed in an order different from that described, and various steps can also be added, omitted, or combined. In addition, features described with reference to certain examples can be combined in other examples.

[0170] From the above description of the embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be realized by means of computer software products and general hardware platforms as necessary, and of course, can also be realized by hardware. The computer software product is stored in a storage medium (such as ROM, RAM, magnetic disc, optical disc, etc.), and includes a plurality of instructions for making the terminal or network side device execute the method described in various embodiments of the present disclosure.

[0171] The embodiments of the present disclosure are described above in combination with the drawings, but the present disclosure is not limited to the above-mentioned specific embodiments, and the above-mentioned specific embodiments are only illustrative, but not restrictive. Those skilled in the art can make many forms of embodiments under the inspiration of the present disclosure without departing from the scope of the present disclosure and the scope protected by the claims, and these embodiments all belong to the protection scope of the present disclosure.

Claims

1. A vehicle driving condition prediction method, comprising: obtaining at least one target vehicle driving data sample matching real-time driving data of a target vehicle based on the real-time driving data and a vehicle driving data sample set, wherein the vehicle driving data sample set comprises a plurality of first vehicle driving data samples; predicting a target driving condition of the target vehicle based on the at least one target vehicle driving data sample.

2. The method of claim 1, further comprising: determining target driving condition speed data corresponding to the target driving condition based on the target driving condition; outputting the target driving condition speed data, or outputting a set of motion parameters predicted based on the target driving condition speed data.

3. The method of claim 1, wherein, The vehicle driving data sample set comprises a plurality of sample sets, each sample set of the plurality of sample sets comprises at least two first vehicle driving data samples, each sample set has a corresponding condition label configured to identify a driving condition corresponding to the at least two first vehicle driving data samples in the each sample set; the at least one target vehicle driving data sample comprises at least two target vehicle driving data samples; The prediction of the target driving condition of the target vehicle based on the at least one target vehicle driving data sample comprises: determining, for each target vehicle driving data sample of the at least two target vehicle driving data samples, a sample set comprising the target vehicle driving data sample; determining that a condition label corresponding to a target sample set is the target driving condition of the target vehicle, wherein the target sample set is a sample set comprising the most target vehicle driving data samples.

4. The method of claim 1, wherein, The obtaining of the at least one target vehicle driving data sample matching the real-time driving data based on the real-time driving data and the vehicle driving data sample set comprises: performing feature extraction on the real-time driving data to obtain a feature vector corresponding to the real-time driving data; determining, based on similarities between the feature vector and feature vectors of each first vehicle driving data sample in the vehicle driving data sample set, first vehicle driving data samples corresponding to top N similarities in descending order of similarity as the target vehicle driving data samples, wherein N is an integer greater than or equal to 2.

5. The method of claim 4, wherein, The similarity is determined based on Euclidean distance.

6. The method of claim 3, wherein, The condition label is determined based on relevant parameter information of the first vehicle driving data samples in the each sample set, wherein the relevant parameter information comprises at least one of a speed-related feature parameter and an acceleration-related feature parameter.

7. The method of claim 3, wherein, The plurality of first vehicle driving data samples are obtained by dividing historical vehicle driving data samples according to a preset time length; and the plurality of sample sets are obtained by performing clustering analysis on the plurality of first vehicle driving data samples based on feature vectors corresponding to the plurality of first vehicle driving data samples.

8. The method of claim 7, wherein, The plurality of sample sets are obtained by performing standardization processing on each parameter of the feature vectors corresponding to the plurality of first vehicle driving data samples, and performing clustering analysis based on the feature vectors after the standardization processing.

9. The method according to any one of claims 2-8, wherein, The driving condition speed data corresponding to the driving condition is constructed based on historical vehicle driving data samples.

10. The method of claim 9, wherein, The driving condition speed data corresponding to the driving condition is obtained by splicing second vehicle driving data samples, which are obtained by dividing the historical vehicle driving data samples according to vehicle speed.

11. The method of claim 9, wherein, The driving condition speed data corresponding to the driving condition is obtained by splicing second vehicle driving data samples after screening the second vehicle driving data samples based on a preset condition, wherein the preset condition includes at least one of the following: If the second vehicle driving data sample has data frame loss, the second vehicle driving data sample is deleted. If the running time of the second vehicle driving data sample is less than a preset time length, the second vehicle driving data sample is deleted.

12. The method of claim 9, wherein, The driving condition speed data corresponding to the driving condition is obtained by splicing second vehicle driving data samples belonging to the same driving condition, wherein the driving condition corresponding to the second vehicle driving data sample is determined based on a parameter threshold of the driving condition, and the second vehicle driving data sample is obtained by screening the second vehicle driving data sample based on a preset condition.

13. The method of claim 9, wherein, The parameter threshold of the driving condition is determined based on related parameter information of the first vehicle driving data sample in each sample set.

14. A controller comprising: A processor and a memory, the memory stores a program or instructions executable on the processor, and the program or the instructions are executed by the processor to implement the vehicle driving condition prediction method according to any one of claims 1 to 13.

15. A vehicle comprising: A processor and a memory, the memory stores a program or instructions executable on the processor, and the program or the instructions are executed by the processor to implement the vehicle driving condition prediction method according to any one of claims 1 to 13.

16. A cloud server comprising: A processor and a memory, the memory stores a program or instructions executable on the processor, and the program or the instructions are executed by the processor to implement the vehicle driving condition prediction method according to any one of claims 1 to 13.

17. A computer readable storage medium, wherein, The readable storage medium stores a program or instructions, and the program or the instructions are executed by the processor to implement the vehicle driving condition prediction method according to any one of claims 1 to 13.

18. A computer program product, wherein, The computer program product is executed by the processor of the vehicle or the cloud server to implement the vehicle driving condition prediction method according to any one of claims 1 to 13.

Images