Method and device for predicting energy consumption of vehicle, computer equipment and storage medium

Abstract

The invention relates to the technical field of computers, and discloses a method and device for predicting the energy consumption of a vehicle, computer equipment and a storage medium, and the method comprises the steps: dividing a target navigation route for a target vehicle into a plurality of sub-target navigation routes; for each sub-target navigation route, utilizing an energy consumption prediction network to obtain predicted energy consumption of the target vehicle for the sub-target navigation route according to the characteristics used for predicting energy consumption and corresponding to the sub-target navigation route; and determining the predicted energy consumption of the target vehicle for the target navigation route according to the predicted energy consumption of the target vehicle for each sub-target navigation route.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and in particular to a method, an apparatus, a computer device and a storage medium for predicting the energy consumption of a vehicle. Background Art

[0002] The energy consumption of a vehicle, such as the power consumption of an electric vehicle, is the indicator that the driver in the vehicle is most concerned about. At present, when the driver in the vehicle uses the navigation function, the driver in the vehicle needs to predict the energy consumption of the vehicle according to the navigation route to the destination based on experience. The accuracy of the energy consumption predicted by the driver in the vehicle is low. How to improve the accuracy of the predicted energy consumption has become a problem that needs to be solved. Summary of the invention

[0003] In view of this, embodiments of the present disclosure provide a method, an apparatus, a computer device, and a storage medium for predicting the energy consumption of a vehicle.

[0004] In a first aspect, an embodiment of the present disclosure provides a method for predicting energy consumption of a vehicle, the method comprising:

[0005] Dividing a target navigation route for a target vehicle into a plurality of sub-target navigation routes, wherein the road sections corresponding to each of any two adjacent sub-target navigation routes have different road section type characteristics, and the road section type characteristics of the road sections corresponding to the sub-target navigation routes include: a type related to the slope of the road sections corresponding to the sub-target navigation routes, and a traffic type of the road sections corresponding to the sub-target navigation routes;

[0006] For each sub-target navigation route, using an energy consumption prediction network, according to the features for predicting energy consumption corresponding to the sub-target navigation route, the predicted energy consumption of the target vehicle for the sub-target navigation route is obtained, wherein the features for predicting energy consumption include: a section type feature of a section corresponding to the sub-target navigation route, and a route feature of the sub-target navigation route, wherein the energy consumption prediction network is a trained neural network, and the training data of the energy consumption prediction network is generated according to the historical navigation route for the target vehicle;

[0007] The predicted energy consumption of the target vehicle for the target navigation route is determined according to the predicted energy consumption of the target vehicle for each sub-target navigation route.

[0008] In a possible implementation, the using the energy consumption prediction network to obtain the predicted energy consumption of the target vehicle for the sub-target navigation route according to the features for predicting energy consumption corresponding to the sub-target navigation route includes:

[0009] For each first energy consumption prediction network, using the first energy consumption prediction network, according to the features for predicting energy consumption corresponding to the sub-target navigation route, obtain the candidate energy consumption for the sub-target navigation route predicted by the first energy consumption prediction network, wherein the first energy consumption prediction network is a trained neural network, and when there are multiple first energy consumption prediction networks, each first energy consumption prediction network has a different training data set;

[0010] The predicted energy consumption of the target vehicle for the sub-target navigation route is determined according to the candidate energy consumption for the sub-target navigation route predicted by each first energy consumption prediction network.

[0011] In a possible implementation, the method further includes:

[0012] Comparing the candidate energy consumption for the sub-target navigation route predicted by each first energy consumption prediction network with the actual energy consumption of the target vehicle for the sub-target navigation route, to obtain an accuracy parameter for the sub-target navigation route of each first energy consumption prediction network, wherein the accuracy parameter indicates whether the candidate energy consumption for the sub-target navigation route predicted by the first energy consumption prediction network is accurate or inaccurate;

[0013] When the accuracy parameters predicted by each of the first energy consumption prediction networks for the sub-target navigation route indicate that the candidate energy consumption is inaccurate, difference data for the first energy consumption prediction network is generated, and the difference data includes: route characteristics of the sub-target navigation route, segment type characteristics of the segment corresponding to the sub-target navigation route, and actual energy consumption of the target vehicle for the sub-target navigation route.

[0014] In a possible implementation, the method further includes:

[0015] When the amount of difference data generated for the first energy consumption prediction network reaches a difference data amount threshold, generating a second energy consumption prediction network;

[0016] Using each difference data for the first energy consumption prediction network among the difference data quantity threshold number of difference data for the first energy consumption prediction network as training data for the second energy consumption prediction network, to obtain a training data set for the second energy consumption prediction network;

[0017] The second energy consumption prediction network is trained using its training data set.

[0018] In a possible implementation manner, each of the first energy consumption prediction networks has a confidence level; and

[0019] The method further includes:

[0020] The confidence of each first energy consumption prediction network is updated according to the accuracy parameter of each first energy consumption prediction network for the sub-target navigation route, wherein the confidence of the first energy consumption prediction network indicates the accuracy of the energy consumption inferred by the first energy consumption prediction network.

[0021] In a possible implementation, the method further includes:

[0022] When the number of times the confidence of the target first energy consumption prediction network is updated is greater than or equal to the number threshold and the confidence of the target first energy consumption prediction network is less than the confidence threshold, the target first energy consumption prediction network is deleted, wherein the target first energy consumption prediction network is any first energy consumption prediction network.

[0023] In a second aspect, an embodiment of the present disclosure provides a device for predicting energy consumption of a vehicle, comprising:

[0024] A division unit is used to divide a target navigation route for a target vehicle into a plurality of sub-target navigation routes, wherein the road sections corresponding to each of any two adjacent sub-target navigation routes have different road section type characteristics, and the road section type characteristics of the road sections corresponding to the sub-target navigation routes include: a type related to the slope of the road sections corresponding to the sub-target navigation routes, and a traffic type of the road sections corresponding to the sub-target navigation routes;

[0025] A first prediction unit is used for obtaining, for each sub-target navigation route, a predicted energy consumption of a target vehicle for the sub-target navigation route using an energy consumption prediction network according to features for predicting energy consumption corresponding to the sub-target navigation route, wherein the features for predicting energy consumption include: a segment type feature of a segment corresponding to the sub-target navigation route and a route feature of the sub-target navigation route, wherein the energy consumption prediction network is a trained neural network, and training data of the energy consumption prediction network is generated according to historical navigation routes for the target vehicle;

[0026] The second prediction unit is used to determine the predicted energy consumption of the target vehicle for the target navigation route according to the predicted energy consumption of the target vehicle for each sub-target navigation route.

[0027] In one possible implementation, the first prediction unit is further used to, for each first energy consumption prediction network, utilize the first energy consumption prediction network to obtain the candidate energy consumption for the sub-target navigation route predicted by the first energy consumption prediction network according to the features for predicting energy consumption corresponding to the sub-target navigation route, wherein the first energy consumption prediction network is a trained neural network, and when the number of first energy consumption prediction networks is multiple, each first energy consumption prediction network has a different training data set; and determine the predicted energy consumption of the target vehicle for the sub-target navigation route according to the candidate energy consumption for the sub-target navigation route predicted by each first energy consumption prediction network.

[0028] In a possible implementation, the device for predicting the energy consumption of a vehicle further includes:

[0029] A difference data generating unit is used to compare the candidate energy consumption for the sub-target navigation route predicted by each first energy consumption prediction network with the actual energy consumption of the target vehicle for the sub-target navigation route, and obtain the accuracy parameter of each first energy consumption prediction network for the sub-target navigation route, wherein the accuracy parameter indicates whether the candidate energy consumption for the sub-target navigation route predicted by the first energy consumption prediction network is accurate or inaccurate; when the accuracy parameters for the sub-target navigation route predicted by each first energy consumption prediction network indicate that the candidate energy consumption is inaccurate, generating difference data for the first energy consumption prediction network, the difference data including: route characteristics of the sub-target navigation route, section type characteristics of the section corresponding to the sub-target navigation route, and the actual energy consumption of the target vehicle for the sub-target navigation route.

[0030] In a possible implementation, the device for predicting the energy consumption of a vehicle further includes:

[0031] The second energy consumption prediction network training unit is used to generate a second energy consumption prediction network when the number of difference data generated for the first energy consumption prediction network reaches a difference data number threshold; use each difference data for the first energy consumption prediction network among the difference data number threshold number of difference data for the first energy consumption prediction network as training data for the second energy consumption prediction network to obtain a training data set for the second energy consumption prediction network; and use the training data set of the second energy consumption prediction network to train the second energy consumption prediction network.

[0032] In a possible implementation, each of the first energy consumption prediction networks has a confidence level; and the device for predicting the energy consumption of a vehicle further includes:

[0033] A confidence updating unit is used to update the confidence of each first energy consumption prediction network according to the accuracy parameter of each first energy consumption prediction network for the sub-target navigation route, wherein the confidence of the first energy consumption prediction network indicates the accuracy of the energy consumption inferred by the first energy consumption prediction network.

[0034] In a possible implementation, the device for predicting the energy consumption of a vehicle further includes:

[0035] A deleting unit is used to delete the target first energy consumption prediction network when the number of times the confidence of the target first energy consumption prediction network is updated is greater than or equal to the number threshold and the confidence of the target first energy consumption prediction network is less than the confidence threshold, wherein the target first energy consumption prediction network is any first energy consumption prediction network.

[0036] In a third aspect, an embodiment of the present disclosure provides a computer device, comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, computer instructions being stored in the memory, and the processor executing the method of the first aspect or any corresponding embodiment thereof by executing the computer instructions.

[0037] In a fourth aspect, an embodiment of the present disclosure provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the method of the first aspect or any corresponding implementation manner thereof.

[0038] In a fifth aspect, the present invention provides a computer program product, comprising computer instructions for causing a computer to execute the method of the first aspect or any corresponding embodiment thereof.

[0039] The method for predicting the energy consumption of a vehicle provided by the embodiment of the present disclosure uses an energy consumption prediction network for each sub-target navigation route, and obtains the predicted energy consumption of the target vehicle for the sub-target navigation route based on the features with a high correlation with energy consumption, i.e., the features corresponding to the sub-target navigation route for predicting energy consumption. Thus, the accuracy of the predicted energy consumption of the target vehicle for each sub-target navigation route is high. The predicted energy consumption of the target vehicle for the target navigation route is determined based on the predicted energy consumption of the target vehicle for each sub-target navigation route. Thus, a relatively accurate predicted energy consumption of the target vehicle for the target navigation route is obtained.

[0040] At the same time, considering that the relationship between different road section type features and the energy consumption of the vehicle is different, the target navigation route of the target vehicle is finely divided, and the target navigation route for the target vehicle is divided into multiple sub-target navigation routes. The road section corresponding to each sub-target navigation route in any two adjacent sub-target navigation routes has different road section type features. For each sub-target navigation route, a single road section type feature is used to predict the predicted energy consumption for the sub-target navigation route. Finely dividing the target navigation route of the target vehicle can avoid the following situation that leads to low accuracy of the predicted energy consumption: the road section corresponding to the divided sub-navigation road section includes multiple different types of road sections. Since the feature complexity of the road section type features of multiple different types of road sections is much higher than the feature complexity of a single road section type feature, it is difficult for the neural network used to predict energy consumption to learn the relationship between the road section type features of multiple different types of road sections and the energy consumption of the vehicle more accurately when the road section type features of multiple different types of road sections are taken as a whole. The predicted energy consumption for the sub-target navigation route is predicted using the road section type features of the multiple different types of road sections, and the predicted energy consumption for the sub-target navigation route is predicted with low accuracy.

[0041] In addition, the training data of the energy consumption prediction network is generated according to the historical navigation route of the target vehicle, and the training data of the energy consumption prediction network has a high correlation with the target vehicle, so that after the energy consumption prediction network is trained, the energy consumption prediction network is more suitable for predicting the predicted energy consumption of the target vehicle for the target navigation route. Thus, a more accurate predicted energy consumption of the target vehicle for the target navigation route is obtained. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the specific embodiments of the present disclosure or the technical solutions in the prior art, the drawings required for use in the specific embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0043] Figure 1 It is a schematic diagram of multiple slope types and multiple traffic types;

[0044] Figure 2 It is a schematic diagram of an example of the relationship between the sub-target navigation route and the road section of the target navigation route;

[0045] Figure 3 is a flow chart of a method for predicting energy consumption of a vehicle provided by an embodiment of the present disclosure;

[0046] Figure 4 is a flow chart of another method for predicting energy consumption of a vehicle provided by an embodiment of the present disclosure;

[0047] Figure 5 is a flowchart of an example of generating difference data for a first energy consumption prediction network;

[0048] Figure 6 It is a schematic diagram of the structure of a computer device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solution and advantages of the embodiments of the present disclosure clearer, the technical solution in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the 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 the embodiments in the present disclosure, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present disclosure.

[0050] refer to Figure 1 , which shows a schematic diagram of an example of the relationship between the sub-target navigation route and the road segment of the target navigation route.

[0051] Figure 1 Sub-target navigation routes such as sub-target navigation route 1, sub-target navigation route 2, sub-target navigation route 3, and sub-target navigation route 4 are shown.

[0052] Sub-target navigation route 1 and sub-target navigation route 2 are two adjacent sub-target navigation routes, sub-target navigation route 2 and sub-target navigation route 3 are two adjacent sub-target navigation routes, and sub-target navigation route 3 and sub-target navigation route 4 are two adjacent sub-target navigation routes.

[0053] Figure 1 Sub-target navigation routes such as section 1 corresponding to sub-target navigation route 1, section 2 corresponding to sub-target navigation route 2, section 3 corresponding to sub-target navigation route 3, and section 4 corresponding to sub-target navigation route 4 are shown.

[0054] Section 1 and section 2 are two adjacent sections, section 2 and section 3 are two adjacent sections, and section 3 and section 4 are two adjacent sections.

[0055] When traveling according to the target navigation route for the target vehicle from the departure point of the target navigation route to the destination of the target navigation route, the target vehicle passes through section 1, section 2, section 3, and section 4 in sequence.

[0056] The type associated with the slope of road segment 1 is negative slope type, and the traffic type of road segment 1 is straight road type. The type associated with the slope of road segment 2 is flat road type, and the traffic type of road segment 2 is straight road type. The type associated with the slope of road segment 3 is positive slope type, and the traffic type of road segment 3 is turning intersection type. The type associated with the slope of road segment 4 is positive slope type, and the traffic type of road segment 4 is straight road type.

[0057] refer to Figure 2 , which shows a flow chart of a method for predicting the energy consumption of a vehicle provided by an embodiment of the present disclosure. The method can be executed by a target vehicle, such as a vehicle-mounted system of the target vehicle.

[0058] It should be noted that once the navigation application on the target vehicle generates a target navigation route, the method for predicting the energy consumption of the vehicle provided in the embodiment of the present disclosure can be executed to quickly obtain the predicted energy consumption of the target vehicle for the target navigation route.

[0059] In step S201, a target navigation route for a target vehicle is divided into a plurality of sub-target navigation routes.

[0060] The target vehicle may be any vehicle to which the method provided by the embodiments of the present disclosure may be applied.

[0061] The target navigation route is calculated by a navigation application installed on the target vehicle. The target navigation route can be any navigation route for the target vehicle.

[0062] When the target vehicle travels according to the target navigation route for the target vehicle and reaches the destination of the target navigation route from the departure point of the target navigation route, it passes through the road section corresponding to each sub-target navigation route in sequence.

[0063] For a sub-target navigation route, the sub-target navigation route may indicate how to pass through the road section corresponding to the sub-target navigation route.

[0064] Each of the multiple sub-target navigation routes of the target navigation route corresponds to a different road section.

[0065] For a sub-target navigation route, when traveling according to the target navigation route for the target vehicle from the starting point of the target navigation route to the destination of the target navigation route, the target vehicle travels according to the sub-target navigation route, passing through the road section corresponding to the sub-target navigation route, that is, from the starting point of the sub-target navigation route to the end point of the sub-target navigation route.

[0066] When the target vehicle travels according to the target navigation route for the target vehicle and reaches the destination of the target navigation route from the departure point of the target navigation route, it passes through the road section corresponding to each sub-target navigation route in sequence.

[0067] For each sub-target navigation route, the road section type feature of the sub-target navigation route includes: a type related to the slope of the road section corresponding to the sub-target navigation route, and a traffic type of the road section corresponding to the sub-target navigation route.

[0068] For a sub-target navigation route, the road segment type feature of the sub-target navigation route may indicate: the type of the road segment corresponding to the sub-target navigation route.

[0069] In the embodiment of the present disclosure, a plurality of types related to slope and a plurality of traffic types are preset.

[0070] As an example, the pre-set multiple slope-related types include but are not limited to: positive slope road type, negative slope road type, flat road type. The pre-set multiple traffic types include but are not limited to: straight road type, straight intersection type, and turning intersection type. For a sub-target navigation route, if the type of slope-related to the road section corresponding to the sub-target navigation route is a positive slope road type, then the road section corresponding to the sub-target navigation route is a positive slope road. If the type of slope-related to the road section corresponding to the sub-target navigation route is a negative slope road type, then the road section corresponding to the sub-target navigation route is a negative slope road. If the type of slope-related to the road section corresponding to the sub-target navigation route is a flat road type, then the road section corresponding to the sub-target navigation route is a flat road. If the traffic type of the road section corresponding to the sub-target navigation route is a straight road type, then the road section corresponding to the sub-target navigation route is a straight section of road. If the traffic type of the road section corresponding to the sub-target navigation route is a straight intersection type, then the road section corresponding to the sub-target navigation route is a section of road passing through a straight intersection. If the traffic type of the road section corresponding to the sub-target navigation route is a turning intersection type, then the road section corresponding to the sub-target navigation route is a section of road passing through a turning intersection.

[0071] For a sub-target navigation route, the type of slope-related road section corresponding to the sub-target navigation route can be determined based on the comparison result of the angle of the road section corresponding to the sub-target navigation route and the angle threshold. As an example, the angle threshold is 3°. If the angle of the road section corresponding to the sub-target navigation route is less than -3°, the type of slope-related road section corresponding to the sub-target navigation route is a negative slope type. If the angle of the road section corresponding to the sub-target navigation route is greater than or equal to -3° and the angle of the road section corresponding to the sub-target navigation route is less than or equal to 3°, the type of slope-related road section corresponding to the sub-target navigation route is a flat road type. If the angle of the road section corresponding to the sub-target navigation route is greater than 3°, the type of slope-related road section corresponding to the sub-target navigation route is a positive slope type.

[0072] In the embodiment of the present disclosure, the road sections corresponding to each sub-target navigation route in any two adjacent sub-target navigation routes have different road section type characteristics.

[0073] For sub-target navigation route A and sub-target navigation route B, at least one of the road segment type features of sub-target navigation route A does not appear in the road segment type features of sub-target navigation route B.

[0074] Among them, sub-target navigation route A and sub-target navigation route B are any two sub-target navigation routes.

[0075] The road section corresponding to the sub-target navigation route A and the road section corresponding to the sub-target navigation route B are two adjacent road sections.

[0076] In step S202, for each sub-target navigation route among the multiple sub-target navigation routes, an energy consumption prediction network is used to obtain the predicted energy consumption of the target vehicle for the sub-target navigation route according to the features for predicting energy consumption corresponding to the sub-target navigation route.

[0077] In step S202, for sub-target navigation route i, the features for predicting energy consumption corresponding to sub-target navigation route i are input into the energy consumption prediction network to obtain the predicted energy consumption of the target vehicle for the sub-target navigation route output by the energy consumption prediction network.

[0078] As an example, the target navigation route is divided into target navigation route 1, sub-target navigation route 2, sub-target navigation route 3, sub-target navigation route 4...sub-target navigation route n. Using the energy consumption prediction network, according to the features for predicting energy consumption corresponding to the sub-target navigation route 1, the predicted energy consumption of the target vehicle for sub-target navigation route 1 is obtained. Using the energy consumption prediction network, according to the features for predicting energy consumption corresponding to the sub-target navigation route 2, the predicted energy consumption of the target vehicle for sub-target navigation route 2 is obtained. Using the energy consumption prediction network, according to the features for predicting energy consumption corresponding to the sub-target navigation route 3, the predicted energy consumption of the target vehicle for sub-target navigation route 3 is obtained. Using the energy consumption prediction network, according to the features for predicting energy consumption corresponding to the sub-target navigation route 4, the predicted energy consumption of the target vehicle for sub-target navigation route 4 is obtained. Using the energy consumption prediction network, according to the features for predicting energy consumption corresponding to the sub-target navigation route n, the predicted energy consumption of the target vehicle for sub-target navigation route n is obtained.

[0079] For a sub-target navigation route, the features for predicting energy consumption corresponding to the sub-target navigation route include: the segment type features of the segment corresponding to the sub-target navigation route, and the route features of the sub-target navigation route. In addition, the features for predicting energy consumption corresponding to the sub-target navigation route may also include the type of weather in the area where the segment corresponding to the sub-target navigation route is located.

[0080] refer to Figure 3 , which shows a flowchart of an example of predicting the energy consumption of a target vehicle for a sub-target navigation route.

[0081] In this example, the predicted energy consumption of the target vehicle for the sub-target navigation route 1 is predicted.

[0082] The features corresponding to the sub-target navigation route 1 and used for predicting energy consumption include: the section type features of the section corresponding to the sub-target navigation route 1, the route features of the sub-target navigation route 1, and the like.

[0083] The road section type characteristics of the road section corresponding to the sub-target navigation route 1 include: the type of the road section corresponding to the sub-target navigation route 1 related to the slope, and the traffic type of the road section corresponding to the sub-target navigation route 1.

[0084] In this example, the features for predicting energy consumption corresponding to sub-target navigation route 1 are input into the energy consumption prediction network, and the energy consumption prediction network outputs the predicted energy consumption of the target vehicle for sub-target navigation route 1.

[0085] In the disclosed embodiment, for a sub-target navigation route, the predicted energy consumption indication of the target vehicle for the sub-target navigation route is: how much energy the target vehicle will consume when passing through the road section corresponding to the sub-target navigation route along the sub-target navigation route.

[0086] For a sub-target navigation route, the type of predicted energy consumption of the target vehicle for the sub-target navigation route depends on the type of power used when the target vehicle passes through the road section corresponding to the sub-target navigation route.

[0087] In the embodiments of the present disclosure, the type of predicted energy consumption is not limited.

[0088] As an example, if the target vehicle is an electric vehicle, the predicted energy consumption is the predicted power consumption.

[0089] As an example, if the target vehicle is a fuel vehicle, the predicted energy consumption is the predicted fuel consumption.

[0090] In step S202, for a sub-target navigation route, the route characteristics of the sub-target navigation route are obtained from the route characteristics of the target navigation route.

[0091] The route features of the target navigation route are calculated by a navigation application installed on the target vehicle.

[0092] In a possible implementation, for a sub-target navigation route, the route features of the sub-target navigation route include but are not limited to: the length of the sub-target navigation route, and the navigation information of the sub-target navigation route. The navigation information of the sub-target navigation route may include one or more of the following items: the traffic information of the road section corresponding to the sub-target navigation route, the speed limit information of the road section corresponding to the sub-target navigation route, and the lane change information corresponding to the sub-target navigation route.

[0093] The energy consumption prediction network is a trained neural network. The trained network may specifically mean that: before step S201, the training of the energy consumption prediction network is completed.

[0094] As an example, the type of the energy consumption prediction network is one of Convolutional Neural Networks (CNN), Deep Neural Networks (DNN), and Multilayer Perceptrons (MLP).

[0095] In the embodiment of the present disclosure, during the training of the energy consumption prediction network, the energy consumption prediction network is trained using a training data set of the energy consumption prediction network.

[0096] In order to obtain a training data set for the energy consumption prediction network, the historical navigation route for the target vehicle is divided into a plurality of sub-historical navigation routes of the historical navigation route. The historical navigation route is generated before executing step S201. Before executing step S201, the target vehicle travels to the destination of the historical navigation route for the target vehicle according to the historical navigation route for the target vehicle.

[0097] The training data of the energy consumption prediction network includes: features for training corresponding to a sub-historical navigation route of a historical navigation route of a target vehicle, and actual energy consumption of the target vehicle for the sub-historical navigation route.

[0098] The actual energy consumption indication of the target vehicle for the sub-historical navigation route: how much energy the target vehicle actually consumed when passing through the road section corresponding to the sub-historical navigation route according to the sub-historical navigation route.

[0099] For a sub-historical navigation route of the historical navigation route of the target vehicle, the features for training corresponding to the sub-historical navigation route include: the segment type feature of the segment corresponding to the sub-historical navigation route, the route feature of the sub-historical navigation route, etc. In addition, the features for training corresponding to the sub-historical navigation route may also include: the type of weather in the area where the segment corresponding to the sub-historical navigation route is located.

[0100] During the training of the energy consumption prediction network, the segment type characteristics of the segment corresponding to the sub-historical navigation route of the historical navigation route of the target vehicle, the route characteristics of the sub-historical navigation route of the historical navigation route of the target vehicle, etc. are used as inputs of the energy consumption prediction network, and the actual energy consumption of the target vehicle for the sub-historical navigation route is used as the ground truth. The ground truth can be called the annotated energy consumption.

[0101] During the training of the energy consumption prediction network, the training operation for the energy consumption prediction network is repeatedly performed until the condition for completing the training of the energy consumption prediction network is met.

[0102] The training operation for the energy consumption prediction network includes: inputting data other than the labeled energy consumption in a training data set of the energy consumption prediction network into the energy consumption prediction network, obtaining the predicted energy consumption corresponding to the training data output by the energy consumption prediction network, calculating the loss between the predicted energy consumption corresponding to the training data and the labeled energy consumption in the training data, and the loss is used to update the parameters of the energy consumption prediction network.

[0103] In a possible implementation, the condition for completing the training of the energy consumption prediction network may be that the accuracy of the energy consumption prediction network reaches an accuracy threshold or the number of times the parameters of the energy consumption prediction network are updated reaches a number threshold.

[0104] In step S203, the predicted energy consumption of the target vehicle for the target navigation route is determined according to the predicted energy consumption of the target vehicle for each sub-target navigation route.

[0105] The predicted energy consumption of the target vehicle for the target navigation route may refer to: how much energy is expected to be consumed when the target vehicle travels along the target navigation route to the destination of the target navigation route.

[0106] In one possible implementation, the predicted energy consumption of the target vehicle for each sub-target navigation route is added together to obtain the sum of the predicted energy consumption of the target vehicle for each sub-target navigation route, and the sum of the predicted energy consumption of the target vehicle for each sub-target navigation route is determined as the predicted energy consumption of the target vehicle for the target navigation route.

[0107] After determining the predicted energy consumption of the target vehicle for the target navigation route, the predicted energy consumption of the target vehicle for the target navigation route can be displayed on the vehicle screen of the target vehicle, so that the driver of the target vehicle knows the predicted energy consumption of the target vehicle for the target navigation route.

[0108] refer to Figure 4 , which shows a flow chart of another method for predicting the energy consumption of a vehicle provided by an embodiment of the present disclosure.

[0109] In step S401, a target navigation route for a target vehicle is divided into a plurality of sub-target navigation routes.

[0110] Step S401 refers to step S201.

[0111] In step S402, for each sub-target navigation route, the candidate energy consumption for the sub-target navigation route predicted by each first energy consumption prediction network is obtained, and the predicted energy consumption of the target vehicle for the sub-target navigation route is determined based on the candidate energy consumption for the sub-target navigation route predicted by each first energy consumption prediction network.

[0112] Among them, each first energy consumption prediction network is a trained neural network.

[0113] As an example, the type of the first energy consumption prediction network is one of a convolutional neural network, a deep neural network, and a multi-layer perceptron.

[0114] It should be noted that the first energy consumption prediction network can be understood as: a neural network that can currently be used to infer the energy consumption of the target vehicle for the navigation route.

[0115] The number of the first energy consumption prediction network may be one or more.

[0116] When there are multiple first energy consumption prediction networks, each first energy consumption prediction network may have the same structure, and each first energy consumption prediction network has different network parameters.

[0117] When there are multiple first energy consumption prediction networks, each of the multiple first energy consumption prediction networks has a different training data set.

[0118] The method of obtaining the training data of the first energy consumption prediction network j refers to the method of obtaining the training data of the above energy consumption prediction network. The process of training the first energy consumption prediction network j using the training data set of the first energy consumption prediction network j refers to the process of training the above energy consumption prediction network.

[0119] The first energy consumption prediction network j may be any first energy consumption prediction network.

[0120] For sub-target navigation route i, determining the predicted energy consumption of the target vehicle for the sub-target navigation route according to the candidate energy consumption for the sub-target navigation route predicted by each first energy consumption prediction network includes: step S4021-step S4022.

[0121] Among them, the sub-target navigation route i can be any sub-target navigation route.

[0122] In step S4021, for each first energy consumption prediction network, the first energy consumption prediction network is used to obtain the candidate energy consumption for the sub-target navigation route i predicted by the first energy consumption prediction network according to the features for predicting energy consumption corresponding to the sub-target navigation route i.

[0123] In step S4021, for the first energy consumption prediction network j, the features for predicting energy consumption corresponding to the sub-target navigation route i are input into the first energy consumption prediction network j to obtain the candidate energy consumption for the sub-target navigation route i predicted by the first energy consumption prediction network.

[0124] In step S4022, the predicted energy consumption of the target vehicle for the sub-target navigation route i is determined based on the candidate energy consumption for the sub-target navigation route i predicted by each first energy consumption prediction network.

[0125] In a possible implementation, in step S4022, if the number of the first energy consumption prediction networks is one, the candidate energy consumption predicted by the first energy consumption prediction network for the sub-target navigation route i is determined as the predicted energy consumption of the target vehicle for the sub-target navigation route i. If the number of the first energy consumption prediction networks is multiple, the sum of the candidate energy consumption predicted by each first energy consumption prediction network for the sub-target navigation route i is divided by the number of the first energy consumption prediction networks to obtain the average value of the candidate energy consumption predicted by the first energy consumption prediction network for the sub-target navigation route i, and the average value of the candidate energy consumption predicted by the first energy consumption prediction network for the sub-target navigation route i is determined as the predicted energy consumption of the target vehicle for the sub-target navigation route i.

[0126] In the disclosed embodiment, multiple first energy consumption prediction networks can be used to predict the predicted energy consumption of the target vehicle for the sub-target navigation route. The predicted energy consumption of the target vehicle for the sub-target navigation route i can be determined based on the candidate energy consumption for the sub-target navigation route i predicted by each first energy consumption prediction network. Compared with using a single energy consumption prediction network to predict the predicted energy consumption of the target vehicle for the sub-target navigation route i, the energy consumption of the target vehicle for the sub-target navigation route i can be more comprehensively evaluated, thereby improving the accuracy of the predicted energy consumption of the target vehicle for the sub-target navigation route i.

[0127] In step S403, the predicted energy consumption of the target vehicle for the target navigation route is determined according to the predicted energy consumption of the target vehicle for each sub-target navigation route.

[0128] In one possible implementation, in step S403, the predicted energy consumption of the target vehicle for each sub-target navigation route can be added to obtain the sum of the predicted energy consumption of the target vehicle for each sub-target navigation route, and the sum of the predicted energy consumption of the target vehicle for each sub-target navigation route can be determined as the predicted energy consumption of the target vehicle for the target navigation route.

[0129] In a possible implementation, it also includes: step S4041-step S4042.

[0130] In step S4041, for sub-target navigation route i, the candidate energy consumption predicted by each first energy consumption prediction network for sub-target navigation route i is compared with the actual energy consumption of the target vehicle for sub-target navigation route i to obtain the accuracy parameter of each first energy consumption prediction network for sub-target navigation route i.

[0131] The accuracy parameter of the first energy consumption prediction network j for the sub-target navigation route i indicates whether the candidate energy consumption predicted by the first energy consumption prediction network j for the sub-target navigation route i is accurate or inaccurate.

[0132] The first energy consumption prediction network j may be any first energy consumption prediction network.

[0133] The actual energy consumption indication of the target vehicle for the sub-target navigation route i: how much energy the target vehicle actually consumes when passing through the road section corresponding to the sub-target navigation route i according to the sub-target navigation route i.

[0134] The actual energy consumption of the target vehicle for the sub-target navigation route i is obtained through the unit for recording energy consumption of the target vehicle after the target vehicle passes through the road section corresponding to the sub-target navigation route i according to the sub-target navigation route i.

[0135] In step S4041, for the first energy consumption prediction network j, the candidate energy consumption predicted by the first energy consumption prediction network j for the sub-target navigation route i is compared with the actual energy consumption of the target vehicle for the sub-target navigation route i, and the accuracy parameter of the first energy consumption prediction network j for the sub-target navigation route i is obtained, including: calculating the difference between the candidate energy consumption predicted by the first energy consumption prediction network j for the sub-target navigation route i and the actual energy consumption of the target vehicle for the sub-target navigation route i; if the absolute value of the difference is less than the energy consumption difference threshold, then the accuracy parameter of the first energy consumption prediction network j for the sub-target navigation route i indicates: the candidate energy consumption predicted by the first energy consumption prediction network j for the sub-target navigation route i is accurate, wherein the energy consumption difference threshold is greater than 0; if the absolute value of the difference is not less than the energy consumption difference threshold, then the accuracy parameter of the first energy consumption prediction network j for the sub-target navigation route i indicates: the candidate energy consumption predicted by the first energy consumption prediction network j for the sub-target navigation route i is inaccurate.

[0136] Step S4042, when the accuracy parameters predicted by each first energy consumption prediction network for the sub-target navigation route i indicate that the candidate energy consumption is inaccurate, difference data for the first energy consumption prediction network is generated, and the difference data for the first energy consumption prediction network includes: route characteristics of the sub-target navigation route i, section type characteristics of the section corresponding to the sub-target navigation route i, and the actual energy consumption of the target vehicle for the sub-target navigation route i.

[0137] The difference data for the first energy consumption prediction network can be used as training data for the second energy consumption prediction network. The second energy consumption prediction network is a neural network that has not been trained before the difference data for the first energy consumption prediction network is generated for the first time. The structure of the second energy consumption prediction network can be the same as the structure of the first energy consumption prediction network. Each time the difference data for the first energy consumption prediction network is generated, the difference data generated for the first energy consumption prediction network can be used as training data for the second energy consumption prediction network. Thus, the difference data for the first energy consumption prediction network can be used to train the second energy consumption prediction network. After completing the training of the second energy consumption prediction network, the second energy consumption prediction network can be used as the first energy consumption prediction network to predict energy consumption.

[0138] In the disclosed embodiment, difference data for the first energy consumption prediction network can be generated, taking into account the following situation: the first driving mode adopted by the driver of the target vehicle when the target vehicle passes through the road section corresponding to the corresponding sub-target navigation route is significantly different from the second driving mode adopted when the target vehicle passes through the road section corresponding to the sub-historical navigation route of the type of the road section corresponding to the corresponding sub-target navigation route. As a result, the candidate energy consumption predicted by each first neural network for the sub-target navigation route i is inaccurate.

[0139] Difference data for the first energy consumption prediction network is generated, and the difference data for the first energy consumption prediction network can be used to train the second energy consumption prediction network. After the training of the second energy consumption prediction network is completed, the second energy consumption prediction network can more accurately predict the energy consumption of the target vehicle for the corresponding sub-navigation route when the driver of the target vehicle drives the target vehicle in the first driving manner or a driving manner close to the first driving manner. Thereby, the accuracy of the predicted energy consumption can be further improved.

[0140] refer to Figure 5 , which shows a flowchart of an example of generating difference data for a first energy consumption prediction network.

[0141] The features for predicting energy consumption corresponding to the sub-target navigation route 1 are input into the first energy consumption prediction network 1, and the candidate energy consumption 1 predicted by the first energy consumption prediction network 1 for the sub-target navigation route 1 is obtained. The candidate energy consumption 1 is significantly different from the actual energy consumption of the target vehicle for the sub-navigation route 1. The features for predicting energy consumption corresponding to the sub-target navigation route 1 are input into the first energy consumption prediction network 2, and the candidate energy consumption 2 predicted by the first energy consumption prediction network 2 for the sub-target navigation route 1 is obtained. The candidate energy consumption 1 is significantly different from the actual energy consumption of the target vehicle for the sub-navigation route 1. The candidate energy consumption 2 is significantly different from the actual energy consumption of the target vehicle for the sub-navigation route 1. Difference data 1 for the first energy consumption prediction network is generated, and the difference data 1 includes: route features of the sub-target navigation route 1, segment type features of the segment corresponding to the sub-target navigation route 1, and the actual energy consumption of the target vehicle for the sub-target navigation route 1.

[0142] In a possible implementation, the method further includes: step S405.

[0143] In step S405, when the number of difference data generated for the first energy consumption prediction network reaches a difference data number threshold, a second energy consumption prediction network is generated; each difference data for the first energy consumption prediction network among the difference data number threshold number of difference data for the first energy consumption prediction network is used as training data for the second energy consumption prediction network to obtain a training data set for the second energy consumption prediction network; and the second energy consumption prediction network is trained using the training data set of the second energy consumption prediction network.

[0144] The structure of the second energy consumption prediction network may be the same as that of the first energy consumption prediction network.

[0145] During the training of the second energy consumption prediction network, data other than the actual energy consumption in the difference data for the first energy consumption prediction network is used as input to the second energy consumption prediction network.

[0146] During the training of the second energy consumption prediction network, the training operation for the second energy consumption prediction network is repeatedly performed until the condition for completing the training of the second energy consumption prediction network is met.

[0147] The training operation for the second energy consumption prediction network includes: inputting the features of a training data in the training data set of the second energy consumption prediction network except the marked energy consumption into the second energy consumption prediction network, obtaining the predicted energy consumption corresponding to the training data output by the second energy consumption prediction network, calculating the loss between the predicted energy consumption corresponding to the training data and the marked energy consumption in the training data, and the loss is used to update the parameters of the second energy consumption prediction network.

[0148] It should be noted that after the training of the second energy consumption prediction network is completed, that is, after the second energy consumption prediction network is trained, the second energy consumption prediction network can be used as the first energy consumption prediction network.

[0149] In step S405, when the number of difference data for the first energy consumption prediction network reaches the difference data number threshold, a second energy consumption prediction network is generated, taking into account the following situation: the above-mentioned first driving method may be a driving method occasionally adopted by the driver of the target vehicle. If a new energy consumption prediction network, i.e., the second energy consumption prediction network, is trained for the occasionally adopted driving method, it will cause a waste of resources of the target vehicle. When the number of difference data for the first energy consumption prediction network reaches the difference data number threshold, it means that the first driving method or a driving method close to the first driving method has become a driving method frequently adopted by the driver of the target vehicle. Only then will the second energy consumption prediction network be generated. Thus, the problem of training a new energy consumption prediction network, i.e., the second energy consumption prediction network, for the occasionally adopted driving method, which causes a waste of resources of the target vehicle, is avoided.

[0150] In a possible implementation, each first energy consumption prediction network has a confidence level; and further includes: step S406.

[0151] In step S406, for the sub-target navigation route i, the confidence of each first energy consumption prediction network is updated according to the accuracy parameter of each first energy consumption prediction network for the sub-target navigation route i.

[0152] For the first energy consumption prediction network j, the confidence of the first energy consumption prediction network j may indicate: the accuracy of the energy consumption inferred by the first energy consumption prediction network j.

[0153] The higher the confidence of the first energy consumption prediction network, the higher the performance of the first energy consumption prediction network.

[0154] In the disclosed embodiment, the confidence level of each first energy consumption prediction network can enable vehicle engineers to know the performance of each first energy consumption prediction network, thereby facilitating relevant analysis by vehicle engineers.

[0155] For the first energy consumption prediction network j, each time the accuracy parameter of the first energy consumption prediction network j for the corresponding sub-navigation route is obtained, the confidence of the first energy consumption prediction network j can be updated according to the accuracy parameter of the first energy consumption prediction network j for the corresponding sub-navigation route.

[0156] When the confidence of the first energy consumption prediction network j is updated once, the number of accuracy parameters indicating that the candidate energy consumption is accurate among all accuracy parameters of the first energy consumption prediction network j generated before the start time of the update is divided by the number of all accuracy parameters of the first energy consumption prediction network j generated before the start time of the update to obtain the updated confidence of the first energy consumption prediction network j.

[0157] In a possible implementation, it also includes: when the number of times the confidence of the target first energy consumption prediction network is updated is greater than or equal to the number threshold and the confidence of the target first energy consumption prediction network is less than the confidence threshold, deleting the target first energy consumption prediction network, wherein the target first energy consumption prediction network is any first energy consumption prediction network.

[0158] Deleting the target first energy consumption prediction network may include: stopping the operation of the target first energy consumption prediction network; and deleting the code of the target first energy consumption prediction network from a storage space storing the code of the energy consumption prediction network.

[0159] If the number of times the confidence of the target first energy consumption prediction network is updated is greater than or equal to the number threshold and the confidence of the target first energy consumption prediction network is less than the confidence threshold, it can be reflected that the target first energy consumption prediction network has been used to predict energy consumption a sufficient number of times, and the target first energy consumption prediction network has predicted energy consumption inaccurately for a number of times. If the target first energy consumption prediction network is deleted, the target first energy consumption prediction network no longer occupies the computing resources of the running first energy consumption prediction network, and the target first energy consumption prediction network no longer occupies the storage resources of the storage target first energy consumption prediction network, which can save the resources of the target vehicle.

[0160] The disclosed embodiments provide a device for predicting the energy consumption of a vehicle. The device for predicting the energy consumption of a vehicle is used to implement the above-mentioned embodiments and preferred implementation modes, and those that have been described will not be repeated. As used below, the term "unit" may be a combination of software and / or hardware that implements a predetermined function. Although the systems described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware, is also possible and contemplated.

[0161] The device for predicting the energy consumption of a vehicle comprises:

[0162] A division unit is used to divide a target navigation route for a target vehicle into a plurality of sub-target navigation routes, wherein the road sections corresponding to each of any two adjacent sub-target navigation routes have different road section type characteristics, and the road section type characteristics of the road sections corresponding to the sub-target navigation routes include: a type related to the slope of the road sections corresponding to the sub-target navigation routes, and a traffic type of the road sections corresponding to the sub-target navigation routes;

[0163] A first prediction unit is used for obtaining, for each sub-target navigation route, a predicted energy consumption of a target vehicle for the sub-target navigation route using an energy consumption prediction network according to features for predicting energy consumption corresponding to the sub-target navigation route, wherein the features for predicting energy consumption include: a segment type feature of a segment corresponding to the sub-target navigation route and a route feature of the sub-target navigation route, wherein the energy consumption prediction network is a trained neural network, and training data of the energy consumption prediction network is generated according to historical navigation routes for the target vehicle;

[0164] The second prediction unit is used to determine the predicted energy consumption of the target vehicle for the target navigation route according to the predicted energy consumption of the target vehicle for each sub-target navigation route.

[0165] In one possible implementation, the first prediction unit is further used to, for each first energy consumption prediction network, utilize the first energy consumption prediction network to obtain the candidate energy consumption for the sub-target navigation route predicted by the first energy consumption prediction network according to the features for predicting energy consumption corresponding to the sub-target navigation route, wherein the first energy consumption prediction network is a trained neural network, and when the number of first energy consumption prediction networks is multiple, each first energy consumption prediction network has a different training data set; and determine the predicted energy consumption of the target vehicle for the sub-target navigation route according to the candidate energy consumption for the sub-target navigation route predicted by each first energy consumption prediction network.

[0166] In a possible implementation, the device for predicting the energy consumption of a vehicle further includes:

[0167] A difference data generating unit is used to compare the candidate energy consumption for the sub-target navigation route predicted by each first energy consumption prediction network with the actual energy consumption of the target vehicle for the sub-target navigation route, and obtain the accuracy parameter of each first energy consumption prediction network for the sub-target navigation route, wherein the accuracy parameter indicates whether the candidate energy consumption for the sub-target navigation route predicted by the first energy consumption prediction network is accurate or inaccurate; when the accuracy parameters for the sub-target navigation route predicted by each first energy consumption prediction network indicate that the candidate energy consumption is inaccurate, generating difference data for the first energy consumption prediction network, the difference data including: route characteristics of the sub-target navigation route, section type characteristics of the section corresponding to the sub-target navigation route, and the actual energy consumption of the target vehicle for the sub-target navigation route.

[0168] In a possible implementation, the device for predicting the energy consumption of a vehicle further includes:

[0169] The second energy consumption prediction network training unit is used to generate a second energy consumption prediction network when the number of difference data generated for the first energy consumption prediction network reaches a difference data number threshold; use each difference data for the first energy consumption prediction network among the difference data number threshold number of difference data for the first energy consumption prediction network as training data for the second energy consumption prediction network to obtain a training data set for the second energy consumption prediction network; and use the training data set of the second energy consumption prediction network to train the second energy consumption prediction network.

[0170] In a possible implementation, each of the first energy consumption prediction networks has a confidence level; and the device for predicting the energy consumption of a vehicle further includes:

[0171] A confidence updating unit is used to update the confidence of each first energy consumption prediction network according to the accuracy parameter of each first energy consumption prediction network for the sub-target navigation route, wherein the confidence of the first energy consumption prediction network indicates the accuracy of the energy consumption inferred by the first energy consumption prediction network.

[0172] In a possible implementation, the device for predicting the energy consumption of a vehicle further includes:

[0173] A deleting unit is used to delete the target first energy consumption prediction network when the number of times the confidence of the target first energy consumption prediction network is updated is greater than or equal to the number threshold and the confidence of the target first energy consumption prediction network is less than the confidence threshold, wherein the target first energy consumption prediction network is any first energy consumption prediction network.

[0174] In this embodiment, the device is presented in the form of a functional unit, where the unit refers to an ASIC circuit, a processor and a memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0175] The further functional description of each of the above units is the same as that of the above corresponding embodiments and will not be repeated here.

[0176] refer to Figure 6, which shows a schematic diagram of the structure of a computer device provided by an embodiment of the present disclosure, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. The various components are connected to each other using different buses for communication, and can be installed on a common motherboard or installed in other ways as needed. The processor can process instructions executed in the computer device, including instructions stored in or on the memory to display graphical information of the GUI on an external input / output device (such as a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides part of the necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system).

[0177] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0178] The memory 20 stores instructions executable by at least one processor 10, so that the at least one processor 10 executes the method shown in the above embodiment.

[0179] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0180] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory.

[0181] The computer device further includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30 and the output device 40 may be connected via a bus or other means.

[0182] The input device 30 can receive input digital or character information, and generate key signal input related to the user settings and function control of the computer device, such as a touch screen, a keypad, a mouse, a track pad, a touch pad, an indicator bar, one or more mouse buttons, a trackball, a joystick, etc. The output device 40 may include a display device, an auxiliary lighting device (e.g., an LED) and a tactile feedback device (e.g., a vibration motor), etc. The above-mentioned display device includes but is not limited to a liquid crystal display, a light emitting diode, a display and a plasma display. In some optional embodiments, the display device can be a touch screen.

[0183] The embodiments of the present disclosure also provide a computer-readable storage medium. The above-mentioned method according to the embodiments of the present disclosure can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium and downloaded through a network, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state drive, etc.; further, the storage medium can also include a combination of the above-mentioned types of memory. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.

[0184] A portion of the embodiments of the present disclosure may be applied as a computer program product, such as a computer program instruction, which, when executed by a computer, can execute or provide the method and / or technical solution according to the present invention through the operation of the computer. Those skilled in the art should understand that the existence of computer program instructions in computer-readable media includes, but is not limited to, source files, executable files, installation package files, etc., and accordingly, the way in which computer program instructions are executed by a computer includes, but is not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium may be any available computer-readable storage medium or communication medium accessible to the computer.

[0185] Although the embodiments of the present disclosure have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present disclosure, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A method for predicting energy consumption of a vehicle, characterized in that: The method comprises: Dividing a target navigation route for a target vehicle into a plurality of sub-target navigation routes, wherein the road sections corresponding to each of any two adjacent sub-target navigation routes have different road section type characteristics, and the road section type characteristics of the road sections corresponding to the sub-target navigation routes include: a type related to the slope of the road sections corresponding to the sub-target navigation routes, and a traffic type of the road sections corresponding to the sub-target navigation routes; For each sub-target navigation route, using an energy consumption prediction network, according to the features for predicting energy consumption corresponding to the sub-target navigation route, the predicted energy consumption of the target vehicle for the sub-target navigation route is obtained, wherein the features for predicting energy consumption include: a section type feature of a section corresponding to the sub-target navigation route, and a route feature of the sub-target navigation route, wherein the energy consumption prediction network is a trained neural network, and the training data of the energy consumption prediction network is generated according to the historical navigation route for the target vehicle; The predicted energy consumption of the target vehicle for the target navigation route is determined according to the predicted energy consumption of the target vehicle for each sub-target navigation route.

2. The method according to claim 1, characterized in that The using the energy consumption prediction network to obtain the predicted energy consumption of the target vehicle for the sub-target navigation route according to the features for predicting energy consumption corresponding to the sub-target navigation route includes: For each first energy consumption prediction network, using the first energy consumption prediction network, according to the features for predicting energy consumption corresponding to the sub-target navigation route, obtain the candidate energy consumption for the sub-target navigation route predicted by the first energy consumption prediction network, wherein the first energy consumption prediction network is a trained neural network, and when there are multiple first energy consumption prediction networks, each first energy consumption prediction network has a different training data set; The predicted energy consumption of the target vehicle for the sub-target navigation route is determined according to the candidate energy consumption for the sub-target navigation route predicted by each first energy consumption prediction network.

3. The method according to claim 2, characterized in that The method further comprises: Comparing the candidate energy consumption for the sub-target navigation route predicted by each first energy consumption prediction network with the actual energy consumption of the target vehicle for the sub-target navigation route, to obtain an accuracy parameter for the sub-target navigation route of each first energy consumption prediction network, wherein the accuracy parameter indicates whether the candidate energy consumption for the sub-target navigation route predicted by the first energy consumption prediction network is accurate or inaccurate; When the accuracy parameters of each of the first energy consumption prediction networks for the sub-target navigation route indicate that the candidate energy consumption is inaccurate, difference data for the first energy consumption prediction network is generated, and the difference data includes: route characteristics of the sub-target navigation route, segment type characteristics of the segment corresponding to the sub-target navigation route, and actual energy consumption of the target vehicle for the sub-target navigation route.

4. The method according to claim 3, characterized in that The method further comprises: When the amount of difference data generated for the first energy consumption prediction network reaches a difference data amount threshold, generating a second energy consumption prediction network; Using each difference data for the first energy consumption prediction network among the difference data quantity threshold number of difference data for the first energy consumption prediction network as training data for the second energy consumption prediction network, to obtain a training data set for the second energy consumption prediction network; The second energy consumption prediction network is trained using its training data set.

5. The method according to claim 3, characterized in that: Each of the first energy consumption prediction networks has a confidence level; and The method further comprises: The confidence of each first energy consumption prediction network is updated according to the accuracy parameter of each first energy consumption prediction network for the sub-target navigation route, wherein the confidence of the first energy consumption prediction network indicates the accuracy of the energy consumption inferred by the first energy consumption prediction network.

6. The method according to claim 5, characterized in that The method further comprises: When the number of times the confidence of the target first energy consumption prediction network is updated is greater than or equal to the number threshold and the confidence of the target first energy consumption prediction network is less than the confidence threshold, the target first energy consumption prediction network is deleted, wherein the target first energy consumption prediction network is any first energy consumption prediction network.

7. A device for predicting energy consumption of a vehicle, characterized in that: include: A division unit is used to divide a target navigation route for a target vehicle into a plurality of sub-target navigation routes, wherein the road sections corresponding to each of any two adjacent sub-target navigation routes have different road section type characteristics, and the road section type characteristics of the road sections corresponding to the sub-target navigation routes include: a type related to the slope of the road sections corresponding to the sub-target navigation routes, and a traffic type of the road sections corresponding to the sub-target navigation routes; A first prediction unit is used for obtaining, for each sub-target navigation route, a predicted energy consumption of a target vehicle for the sub-target navigation route using an energy consumption prediction network according to features for predicting energy consumption corresponding to the sub-target navigation route, wherein the features for predicting energy consumption include: a segment type feature of a segment corresponding to the sub-target navigation route and a route feature of the sub-target navigation route, wherein the energy consumption prediction network is a trained neural network, and training data of the energy consumption prediction network is generated according to historical navigation routes for the target vehicle; The second prediction unit is used to determine the predicted energy consumption of the target vehicle for the target navigation route according to the predicted energy consumption of the target vehicle for each sub-target navigation route.

8. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the method according to any one of claims 1 to 6 by executing the computer instructions.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the method according to any one of claims 1 to 6.

10. A computer program product, characterized in that The method comprises computer instructions for causing a computer to execute the method according to any one of claims 1 to 6.

Images