Method, device and computer storage medium for processing range data

Abstract

The present disclosure relates to a kind of endurance mileage data processing method, device, equipment and computer storage medium, the method includes: obtaining multiple frames of vehicle driving state data and endurance mileage prediction data;The multiple frames of vehicle driving state data are processed, and the change trend data of the multiple frames of vehicle driving state data is obtained;According to the multiple frames of vehicle driving state data and the change trend data, the endurance mileage prediction data is processed, and the smoothed endurance mileage prediction data is obtained.The present disclosure constrains endurance mileage prediction data by vehicle driving state data and change trend data, so that endurance mileage prediction data is smoothed according to the change trend reflected by vehicle driving state data, and finally the smoothed endurance mileage prediction data is obtained, which avoids the jump of endurance mileage data, facilitates the user to determine the real endurance mileage, and improves the driving experience of user.

Description

Technical Field

[0001] This disclosure relates to the field of vehicle technology, and in particular to a method, apparatus, device, and computer storage medium for processing driving range data. Background Technology

[0002] The driving range is the distance a vehicle can travel in its current state. It is used to remind users to charge the vehicle before the battery runs out, or to remind users whether they can reach their destination, thus providing them with travel assurance.

[0003] The vehicle's driving range changes in real time depending on the vehicle's driving status and the external environment. When the vehicle's driving status data fluctuates significantly or the external environment is complex, the driving range displayed on the center console will change, making it impossible for users to determine the true driving range and resulting in a poor driving experience. Summary of the Invention

[0004] To address the aforementioned technical issues, this disclosure provides a method, apparatus, device, and computer storage medium for processing driving range data, thereby preventing jumps in driving range data, facilitating users in determining the actual driving range, and improving the user's driving experience.

[0005] In a first aspect, embodiments of this disclosure provide a method for processing driving range data, including:

[0006] Acquire multi-frame vehicle driving status data and range prediction data;

[0007] The multi-frame vehicle driving status data is processed to obtain the change trend data of the multi-frame vehicle driving status data.

[0008] Based on the multi-frame vehicle driving status data and the trend data, the range prediction data is processed to obtain smoothed range prediction data.

[0009] In some embodiments, obtaining the predicted driving range data includes:

[0010] The vehicle's driving status is input into the range prediction model, and the range prediction data is obtained through the range prediction model.

[0011] In some embodiments, the multi-frame vehicle driving state data is processed to obtain trend data of the changes in the multi-frame vehicle driving state data, including:

[0012] The difference between every two adjacent frames of vehicle driving status data in the multi-frame vehicle driving status data is calculated to obtain multiple first difference values;

[0013] The plurality of first differences are used as the trend data.

[0014] In some embodiments, the range prediction data is processed based on the multi-frame vehicle driving state data and the trend data to obtain smoothed range prediction data, including:

[0015] The multi-frame vehicle driving status data, the trend data, and the range prediction data are segmented according to time windows.

[0016] Multiple frames of vehicle driving status data, change trend data, and range prediction data within the same time window are input into the range smoothing model, and smoothed range prediction data are obtained through the range smoothing model.

[0017] In some embodiments, after obtaining the smoothed range prediction data through the range smoothing model, the method further includes:

[0018] The vehicle driving status data and the smoothed driving range prediction data are input into the driving range smoothing model, and the driving range smoothing stability data is obtained through the driving range smoothing model.

[0019] The parameters of the range smoothing model are updated based on the range smoothing stability data.

[0020] Secondly, embodiments of this disclosure provide a method for training a range smoothing model, including:

[0021] The system acquires range prediction data, smoothed range prediction data, and range smoothing stability data, wherein the range prediction data is obtained through a range prediction model, and the smoothed range prediction data and range smoothing stability data are obtained through a range smoothing model.

[0022] The difference between the predicted driving range data and the smoothed driving range prediction data is calculated to obtain a second difference value. The second difference value and the driving range smoothing stability data are processed to determine the loss value of the driving range smoothing model.

[0023] The parameters of the range smoothing model are updated based on the loss value of the range smoothing model, so that the loss value of the range smoothing model is less than or equal to a preset threshold.

[0024] Thirdly, embodiments of this disclosure provide a range data processing device, including:

[0025] The first acquisition module is used to acquire multiple frames of vehicle driving status data and driving range prediction data.

[0026] The first processing module is used to process the multi-frame vehicle driving state data to obtain the change trend data of the multi-frame vehicle driving state data.

[0027] The second processing module is used to process the range prediction data based on the multi-frame vehicle driving status data and the change trend data to obtain smoothed range prediction data.

[0028] Fourthly, embodiments of this disclosure provide a range smoothing model training device, comprising:

[0029] The second acquisition module is used to acquire range prediction data, smoothed range prediction data, and range smoothing stability data, wherein the range prediction data is obtained through a range prediction model, and the smoothed range prediction data and range smoothing stability data are obtained through a range smoothing model.

[0030] The third processing module is used to calculate the difference between the predicted driving range data and the smoothed predicted driving range data to obtain a second difference value, and to process the second difference value and the driving range smoothing stability data to determine the loss value of the driving range smoothing model.

[0031] The second update module is used to update the parameters of the range smoothing model based on the loss value of the range smoothing model, so that the loss value of the range smoothing model is less than or equal to a preset threshold.

[0032] Fifthly, embodiments of this disclosure provide a vehicle including the range data processing device as described in the third aspect above.

[0033] Sixthly, embodiments of this disclosure provide an electronic device, including:

[0034] Memory;

[0035] Processor; and

[0036] Computer programs;

[0037] The computer program is stored in the memory and configured to be executed by the processor to implement the method as described in the first aspect.

[0038] In a seventh aspect, embodiments of this disclosure provide a computer-readable storage medium having a computer program stored thereon, the computer program being executed by a processor to implement the method described in the first or second aspect.

[0039] Eighthly, this disclosure also provides a computer program product, which includes a computer program or instructions that, when executed by a processor, implement the range data processing method or range smoothing model training method as described above.

[0040] The driving range data processing method, apparatus, device, and computer storage medium provided in this disclosure constrain the driving range prediction data through vehicle driving status data and change trend data, so that the driving range prediction data is smoothly corrected according to the change trend reflected by the vehicle driving status data, and finally smoothed driving range prediction data is obtained. This avoids jumps in driving range data, makes it easier for users to determine the actual driving range, and improves the user's driving experience. Attached Figure Description

[0041] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0042] To more clearly illustrate the technical solutions in the embodiments of this disclosure or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 A flowchart of a range data processing method provided in this embodiment of the disclosure;

[0044] Figure 2 A flowchart of a range data processing method provided in another embodiment of this disclosure;

[0045] Figure 3 This is a schematic diagram of the structure of a vehicle range smoothing model provided in an embodiment of the present disclosure;

[0046] Figure 4 A flowchart of the range smoothing model training method provided in this embodiment of the disclosure;

[0047] Figure 5 A schematic diagram of the structure of the range data processing device provided in the embodiments of this disclosure;

[0048] Figure 6 A schematic diagram of the structure of the range smoothing model training device provided in this embodiment of the disclosure;

[0049] Figure 7 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this disclosure. Detailed Implementation

[0050] To better understand the above-mentioned objectives, features, and advantages of this disclosure, the solutions disclosed herein will be further described below. It should be noted that, unless otherwise specified, the embodiments and features described herein can be combined with each other.

[0051] Numerous specific details are set forth in the following description in order to provide a full understanding of this disclosure, but this disclosure may also be implemented in other ways different from those described herein; obviously, the embodiments in the specification are only some, and not all, of the embodiments of this disclosure.

[0052] This disclosure provides a method for processing driving range data, which will be described below with reference to specific embodiments.

[0053] Figure 1 This is a flowchart illustrating a range data processing method provided in an embodiment of this disclosure. This method can be executed by an in-vehicle device, which may be a vehicle-mounted system, smartphone, PDA, tablet computer, laptop computer, all-in-one device, intelligent driving device, etc. It is understood that the range data processing method provided in this disclosure can also be applied to other scenarios.

[0054] like Figure 1 As shown, the specific steps included in this method are as follows:

[0055] S101. Acquire multiple frames of vehicle driving status data and range prediction data.

[0056] Vehicle driving status data reflects the current operating status of the vehicle and the status of its various components. Onboard equipment continuously acquires and caches vehicle driving status data through various sensors, with each acquired frame constituting one unit of vehicle driving status data. For example, vehicle driving status data may include the vehicle's remaining battery percentage, vehicle speed, etc.

[0057] The predicted driving range data is the remaining driving range of a vehicle calculated by onboard equipment based on the vehicle's driving status data at the current moment or over a period of time. This value changes with the vehicle's driving status. Specifically, the predicted driving range data can be the vehicle's driving range calculated in real-time by onboard equipment based on the vehicle's driving status at the current moment; it can also be the vehicle's driving range calculated using fitting methods such as averaging or linear regression based on the vehicle's driving status data over a period of time; or it can be the vehicle's driving range calculated using fitting methods such as averaging or linear regression based on the vehicle's driving status data during a predetermined distance traveled. This disclosure does not limit the specific calculation method.

[0058] S102. Process the multi-frame vehicle driving status data to obtain the change trend data of the multi-frame vehicle driving status data.

[0059] This involves analyzing and processing multiple frames of vehicle driving status data collected over a period of time to extract the trend data of vehicle driving status over that period. For example, for multiple frames of vehicle remaining battery percentage data in the multi-frame vehicle driving status data, the difference between two adjacent frames of vehicle remaining battery percentage data within a period of time can be used as the trend data. It is understood that vehicle driving status data includes, but is not limited to, vehicle remaining battery percentage data; therefore, the trend data is not limited to the trend data of vehicle remaining battery percentage data, but can also include the trend data of other vehicle driving status data.

[0060] S103. Based on the multi-frame vehicle driving status data and the change trend data, the range prediction data is processed to obtain smoothed range prediction data.

[0061] Vehicle driving status data, combined with its corresponding trend data, reflects the changing trend of the vehicle's driving status over a period of time. The trend of the vehicle's driving range should correspond to this trend. Therefore, using vehicle driving status data and trend data to constrain the driving range prediction data can remove anomalous data with large jumps in the predicted driving range without affecting other normal driving range prediction data. For example, if the vehicle's remaining battery percentage collected by the onboard equipment over a period of time is 95%, 94%, 93%, 92%, and 91%, with a corresponding trend of 1, then the vehicle's driving status is stable during this period, and the driving range prediction data should also be in a stable state. If multiple driving range prediction data show significant differences from adjacent prediction data, then this prediction data is considered anomalous. The anomalous prediction data can be removed or corrected to driving range prediction data that conforms to the changing trend of the vehicle's driving status.

[0062] It is understood that in practice, multiple vehicle driving status data and their corresponding trend data can be combined for comprehensive analysis to achieve a more accurate result on the trend of vehicle driving status changes. The above embodiment of judging the vehicle driving status by the remaining battery percentage is only an easy-to-understand example and should not be regarded as a limitation of the embodiments of this disclosure.

[0063] This embodiment of the disclosure acquires multiple frames of vehicle driving status data and range prediction data; processes the multiple frames of vehicle driving status data to obtain trend data of the changes in the multiple frames of vehicle driving status data; and processes the range prediction data based on the multiple frames of vehicle driving status data and the trend data of changes, constraining the range prediction data by means of the vehicle driving status data and the trend data of changes, so that the range prediction data is smoothly corrected according to the trend of changes reflected by the vehicle driving status data, and finally obtaining smoothed range prediction data, avoiding jumps in range data, making it easier for users to determine the actual range, and improving the user's driving experience.

[0064] Figure 2 This is a flowchart illustrating a range data processing method according to another embodiment of this disclosure. Figure 2 As shown, the method includes the following steps:

[0065] S201. Obtain multiple frames of vehicle driving status data.

[0066] Specifically, vehicle driving status data should include at least one or more of the following: vehicle remaining battery percentage data, time data, current data, and driving speed data.

[0067] The vehicle-mounted equipment continuously acquires multiple frames of vehicle driving status data through various sensors in the vehicle. This vehicle driving status data includes, but is not limited to, the vehicle's remaining battery percentage, time data, current data, and driving speed data.

[0068] S202. Input the vehicle driving status of multiple frames into the driving range prediction model, and obtain the driving range prediction data through the driving range prediction model.

[0069] A pre-trained range prediction model is used to predict the vehicle's range under its current driving conditions, thus obtaining range prediction data. Specifically, various vehicle driving state data can be used as input to the range prediction model, and the corresponding vehicle range can be used as the output to train the range prediction model.

[0070] S203. Calculate the difference between every two adjacent frames of vehicle driving status data in the multi-frame vehicle driving status data to obtain multiple first difference values, and use the multiple first difference values ​​as the change trend data.

[0071] For each type of vehicle driving state data in the multi-frame vehicle driving state data, the difference between each two adjacent frames of the vehicle driving state data is calculated, and the obtained multiple differences are used as trend data.

[0072] S204. The multi-frame vehicle driving status data, the change trend data, and the range prediction data are segmented according to time windows.

[0073] S205. Input the vehicle driving status data, change trend data and range prediction data of multiple frames within the same time window into the range smoothing model, and obtain the smoothed range prediction data through the range smoothing model.

[0074] Figure 3 This is a schematic diagram of a vehicle range smoothing model provided in an embodiment of this disclosure. Figure 3 As shown, vehicle driving status data and corresponding trend data are input into the vehicle driving status input layer of the model. Multiple vehicle driving status features are obtained by extracting features from different vehicle driving status data through an N-layer vehicle driving status data feature extraction layer. Similarly, the range prediction data output from the range prediction model is input into the range prediction data input layer. M-layer range prediction data feature extraction layers are then used to extract features from the range prediction data, resulting in range prediction data features.

[0075] Furthermore, the extracted features are input into the self-attention mechanism feature fusion layer to fuse the various features, allowing each feature to adaptively incorporate relevant parts of other features, thus enriching and supplementing the current features. After feature fusion, the vehicle driving state features are input into the weight calculation layer. Based on pre-set weights, stability weight features are extracted from the vehicle driving state features. The stability weight features can be a weight matrix of size W*W. The stability weights reflect the stability of various types of vehicle driving state data. For example, the vehicle's remaining battery percentage data is more stable than vehicle speed and vehicle current speed compared to time data, and correspondingly, more weights are assigned to stable data.

[0076] The stability weight feature and the range prediction data feature after feature fusion are input into the weighted summation layer. The range prediction data feature after feature fusion is weighted by the stability weight feature to adjust the characteristics of the range prediction data, so that the range prediction data can be smoothed, and finally the smoothed range prediction data is obtained.

[0077] S206. Input the vehicle driving status data and the smoothed range prediction data into the range smoothing model, and obtain the range smoothing stability data through the range smoothing model.

[0078] Vehicle driving status data is input into the feature extraction and transformation layer of the range smoothing model. Data from different dimensions is transformed into the same feature space. For example, vehicle driving status data from different dimensions can be transformed into the same feature space as the smoothed range prediction data, serving as the basis for calculating range smoothing stability. Stability data of the input vehicle driving status data is then extracted. Optionally, relatively stable data can be selected from multiple vehicle driving status data sets and input into the feature extraction and transformation layer for calculating range smoothing stability data. For example, vehicle remaining battery percentage data and time data can be input into the feature extraction and transformation layer. The stability features of the vehicle driving status data in the same feature space and the smoothed range prediction data are input into the range smoothing stability calculation layer. The stability feature data is used to measure the stability of the smoothed range prediction data, and the final output is the range smoothing stability data. Optionally, the similarity between multiple differences between multiple frames of vehicle driving status data and multiple differences between multiple frames of smoothed range prediction data is calculated, and the calculated similarity is used as the range smoothing stability data. The higher the similarity, the better the stability of the smoothed range prediction data is considered.

[0079] S207. Update the parameters of the range smoothing model based on the range smoothing stability data.

[0080] Based on the smoothed driving range stability data, the parameters of the driving range smoothing model are updated to improve the stability of the smoothed driving range prediction data output by the model. Optionally, the parameters of the driving range smoothing model are updated to increase the similarity between multiple differences between multiple frames of vehicle driving state data and multiple differences between multiple frames of smoothed driving range prediction data.

[0081] This embodiment of the invention uses a vehicle range smoothing model to filter and denoise the range prediction data based on the characteristics of the vehicle driving status data and its changing trend data, thereby achieving the effect of smoothing the vehicle range prediction data. Compared with the prior art, which uses data processing methods such as moving average, weighted summation, and linear regression fitting to process the range prediction data, this method avoids the impact of abnormal range prediction data with large fluctuations on the final result and ensures that the accuracy of the range prediction data is not affected.

[0082] In addition, the embodiments of this disclosure can also obtain range smoothing stability data through the vehicle range smoothing model, and further update the model parameters, so as to achieve a better smoothing effect for the vehicle range prediction data.

[0083] Figure 4This is a flowchart illustrating the range smoothing model training method provided in this embodiment. This method can be executed by a terminal device capable of data processing, such as an in-vehicle infotainment system, smartphone, PDA, tablet, laptop, all-in-one computer, or autonomous driving device. It is understood that the range smoothing model training method provided in this embodiment can also be applied to other scenarios. For example... Figure 4 As shown, the method includes the following steps:

[0084] S401, Obtain driving range prediction data, smoothed driving range prediction data, and driving range smoothing stability data.

[0085] The predicted driving range data is obtained through a driving range prediction model, and the smoothed driving range prediction data and driving range smoothing stability data are obtained through a driving range smoothing model.

[0086] S402. Calculate the difference between the predicted driving range data and the smoothed predicted driving range data to obtain a second difference value.

[0087] Specifically, the predicted driving range data is subtracted from the smoothed driving range data pairwise to obtain the difference between the predicted driving range data and the smoothed driving range data. The absolute value of the difference between the predicted driving range data and the smoothed driving range data is used as the second difference value.

[0088] S403. Process the second difference and the range smoothing stability data to determine the loss value of the range smoothing model.

[0089] Specifically, the loss value of the range smoothing model is obtained by adding the first loss value corresponding to the second difference to the second loss value corresponding to the range smoothing stability data. The first loss value is obtained by multiplying the second difference by the first weight value, and the second loss value is obtained by multiplying the range smoothing stability data by the second weight value.

[0090] Based on pre-set weights, the second difference and the range smoothing stability data are multiplied separately, and the two products are added together to obtain the loss value of the range smoothing model. A first weight value is pre-set for the second difference, and a second weight value is set for the range smoothing stability data. The product of the second difference and the first weight value, and the product of the range smoothing stability data and the second weight value, are added together to obtain the loss value of the range smoothing model.

[0091] S404. Update the parameters of the range smoothing model according to the loss value of the range smoothing model, so that the loss value of the range smoothing model is less than or equal to a preset threshold.

[0092] Using gradient backpropagation, the parameters of the range smoothing model are adjusted based on the output loss value. The above steps are repeated to calculate the loss value of the range smoothing model until the loss value is minimized. When the loss value is less than or equal to a preset threshold, the training of the range smoothing model is considered complete.

[0093] This embodiment calculates the loss value of the range smoothing model by integrating the range prediction data, the smoothed range prediction data, and the range smoothing stability data. Based on the loss value, the range smoothing model is trained with constraints to obtain a range smoothing model with better data smoothing effect, thereby further improving the stability of range smoothing.

[0094] Figure 5 This is a schematic diagram of the structure of the range data processing device provided in this embodiment. The range data processing device can be an in-vehicle device as described in the above embodiment, or it can be a component or assembly within the in-vehicle device. The range data processing device provided in this embodiment can execute the processing flow provided in the range data processing method embodiment, such as... Figure 5 As shown, the driving range data processing device 50 includes: a first acquisition module 51, a first processing module 52, and a second processing module 53; wherein the first acquisition module 51 is used to acquire multiple frames of vehicle driving status data and driving range prediction data; the first processing module 52 is used to process the multiple frames of vehicle driving status data to obtain the changing trend data of the multiple frames of vehicle driving status data; the second processing module 53 is used to process the driving range prediction data according to the multiple frames of vehicle driving status data and the changing trend data to obtain smoothed driving range prediction data.

[0095] Optionally, the first acquisition module 51 is further configured to input the vehicle driving status into the range prediction model, and obtain the range prediction data through the range prediction model.

[0096] Optionally, the first processing module 52 is further configured to calculate the difference between every two adjacent frames of vehicle driving state data in the multi-frame vehicle driving state data to obtain a plurality of first difference values; and use the plurality of first difference values ​​as the change trend data.

[0097] Optionally, the second processing module 53 is further configured to segment the multi-frame vehicle driving status data, the change trend data, and the range prediction data according to a time window; input the multi-frame vehicle driving status data, change trend data, and range prediction data within the same time window into the range smoothing model, and obtain the smoothed range prediction data through the range smoothing model.

[0098] Optionally, the range data processing device 50 further includes a first update module 54, which is used to input the vehicle driving status data and the smoothed range prediction data into the range smoothing model, obtain range smoothing stability data through the range smoothing model, and update the parameters of the range smoothing model according to the range smoothing stability data.

[0099] Figure 5 The range data processing device shown in the embodiment can be used to execute the technical solution of the above method embodiment. Its implementation principle and technical effect are similar, and will not be described again here.

[0100] Figure 6 This is a schematic diagram of the structure of the range smoothing model training device provided in this embodiment. The range data processing device can be the terminal device described in the above embodiment, or the range data processing device can be a component or assembly within the terminal device. The range smoothing model training device provided in this embodiment can execute the processing flow provided in the range smoothing model training method embodiment, such as... Figure 6 As shown, the range smoothing model training device 60 includes: a second acquisition module 61, a third processing module 62, and a second update module 63; wherein the second acquisition module 61 is used to acquire range prediction data, smoothed range prediction data, and range smoothing stability data, wherein the range prediction data is obtained through a range prediction model, and the smoothed range prediction data and range smoothing stability data are obtained through a range smoothing model; the third processing module 62 is used to calculate the difference between the range prediction data and the smoothed range prediction data to obtain a second difference value, and to process the second difference value and the range smoothing stability data to determine the loss value of the range smoothing model; the second update module 63 is used to update the parameters of the range smoothing model according to the loss value of the range smoothing model, so that the loss value of the range smoothing model is less than or equal to a preset threshold.

[0101] Optionally, the third processing module 62 is further configured to add the first loss value corresponding to the second difference to the second loss value corresponding to the range smoothing stability data to obtain the loss value of the range smoothing model, wherein the first loss value is obtained by multiplying the second difference by the first weight value, and the second loss value is obtained by multiplying the range smoothing stability data by the second weight value.

[0102] Figure 6The range smoothing model training device shown in the embodiment can be used to execute the technical solution of the above method embodiment. Its implementation principle and technical effect are similar, and will not be described again here.

[0103] Additionally, this disclosure also provides a vehicle that includes the range data processing device described above.

[0104] Figure 7 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this disclosure. The electronic device can be a vehicle-mounted device or a terminal device as described in the above embodiments. The electronic device provided in this disclosure can execute the processing flow provided in the embodiments of the range data processing method or the processing flow provided in the embodiments of the range smoothing model training method. Figure 7 As shown, the electronic device 70 includes: a memory 71, a processor 72, a computer program, and a communication interface 73; wherein, the computer program is stored in the memory 71 and is configured to be executed by the processor 82 as described above, either the range data processing method or the range smoothing model training method.

[0105] In addition, this disclosure also provides a computer-readable storage medium storing a computer program thereon, which is executed by a processor to implement the range data processing method or the range smoothing model training method described in the above embodiments.

[0106] Furthermore, this disclosure also provides a computer program product, which includes a computer program or instructions. When the computer program or instructions are executed by a processor, they implement the range data processing method or the range smoothing model training method described above.

[0107] Computer program code for performing the operations of this disclosure can be written in one or more programming languages ​​or a combination thereof, including but not limited to object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0108] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0109] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0110] The above description is merely a specific embodiment of this disclosure, enabling those skilled in the art to understand or implement it. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this disclosure. Therefore, this disclosure is not to be limited to the embodiments described herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for processing driving range data, characterized in that, The method includes: Acquire multi-frame vehicle driving status data and range prediction data; The multi-frame vehicle driving status data is processed to obtain the change trend data of the multi-frame vehicle driving status data. Based on the multi-frame vehicle driving status data and the change trend data, the range prediction data is processed to obtain smoothed range prediction data. The multi-frame vehicle driving state data is processed to obtain the changing trend data of the multi-frame vehicle driving state data, including: The difference between every two adjacent frames of vehicle driving status data in the multi-frame vehicle driving status data is calculated to obtain multiple first difference values; the multiple first difference values ​​are used as the change trend data. Based on the multi-frame vehicle driving status data and the trend data, the range prediction data is processed to obtain smoothed range prediction data, including: The multi-frame vehicle driving status data, the trend data, and the range prediction data are segmented according to time windows. The vehicle driving status data, the trend data, and the range prediction data within the same time window are input into the range smoothing model, and the smoothed range prediction data are obtained through the range smoothing model.

2. The method according to claim 1, characterized in that, The acquisition of the predicted driving range data includes: The multi-frame vehicle driving status data is input into the range prediction model, and the range prediction data is obtained through the range prediction model.

3. The method according to claim 1, characterized in that, After obtaining the smoothed range prediction data through the range smoothing model, the method further includes: The vehicle driving status data and the smoothed driving range prediction data are input into the driving range smoothing model, and the driving range smoothing stability data is obtained through the driving range smoothing model. The parameters of the range smoothing model are updated based on the range smoothing stability data.

4. The method according to any one of claims 1-3, characterized in that, The training method for the range smoothing model is as follows: The system acquires range prediction data, smoothed range prediction data, and range smoothing stability data, wherein the range prediction data is obtained through a range prediction model, and the smoothed range prediction data and range smoothing stability data are obtained through a range smoothing model. The difference between the predicted driving range data and the smoothed driving range prediction data is calculated to obtain a second difference value. The second difference value and the driving range smoothing stability data are processed to determine the loss value of the driving range smoothing model. The parameters of the range smoothing model are updated based on the loss value of the range smoothing model, so that the loss value of the range smoothing model is less than or equal to a preset threshold.

5. The method according to claim 4, characterized in that, The second difference and the range smoothing stability data are processed to determine the loss value of the range smoothing model, including: The loss value of the range smoothing model is obtained by adding the first loss value corresponding to the second difference to the second loss value corresponding to the range smoothing stability data. The first loss value is obtained by multiplying the second difference by the first weight value, and the second loss value is obtained by multiplying the range smoothing stability data by the second weight value.

6. A driving range data processing device, characterized in that, The device includes: The first acquisition module is used to acquire multiple frames of vehicle driving status data and driving range prediction data. The first processing module is used to process the multi-frame vehicle driving state data to obtain the change trend data of the multi-frame vehicle driving state data. The second processing module is used to process the range prediction data based on the multi-frame vehicle driving status data and the change trend data to obtain smoothed range prediction data. The first processing module is further configured to calculate the difference between every two adjacent frames of vehicle driving status data in the multi-frame vehicle driving status data to obtain multiple first difference values; and use the multiple first difference values ​​as the change trend data. The second processing module is also used to segment the multi-frame vehicle driving status data, the change trend data, and the range prediction data according to a time window; input the multi-frame vehicle driving status data, change trend data, and range prediction data within the same time window into the range smoothing model, and obtain the smoothed range prediction data through the range smoothing model.

7. The apparatus according to claim 6, characterized in that, The range smoothing model is obtained by training a range smoothing model training device, which includes: The second acquisition module is used to acquire range prediction data, smoothed range prediction data, and range smoothing stability data, wherein the range prediction data is obtained through a range prediction model, and the smoothed range prediction data and range smoothing stability data are obtained through a range smoothing model. The third processing module is used to calculate the difference between the predicted driving range data and the smoothed predicted driving range data to obtain a second difference value, and to process the second difference value and the driving range smoothing stability data to determine the loss value of the driving range smoothing model. The second update module is used to update the parameters of the range smoothing model based on the loss value of the range smoothing model, so that the loss value of the range smoothing model is less than or equal to a preset threshold.

8. A vehicle comprising: The range data processing device as described in claim 6.

9. An electronic device, characterized in that, include: Memory; processor; as well as Computer programs; The computer program is stored in the memory and configured to be executed by the processor to implement the method as described in any one of claims 1-5.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1-5.

Images