Hybrid driving mode classification method, vehicle and storage medium

Abstract

The invention belongs to the technical field of hybrid vehicle driving modes, and particularly relates to a hybrid driving mode classification method, a vehicle and a storage medium, and the method comprises the steps: obtaining the personal driving data of a driver; performing normalization operation on the individual driving data of the driver; and outputting the normalized data through a big data calculation model to obtain the driving preference of the target user. According to the method, the parameters of personal information such as age, gender and consumption tendency of the user are added in the model as model training features, so that the driving behavior of the driver and the suitable mode can be more accurately judged; according to the method, on the basis of the initial driving mode, the user preference can be predicted according to the actual driving data of the user, so that a personalized vehicle driving mode scheme conforming to the user preference is given, a user driving preference suggestion is given, and appropriate vehicle HCU calibration parameters are provided.

Description

Technical Field

[0001] The present invention relates to the technical field of hybrid vehicle driving modes, and in particular to a hybrid driving mode classification method, a vehicle and a storage medium. Background Art

[0002] The energy sources of hybrid vehicles (including PHEV / REM extended-range vehicles) mainly include batteries and fuel. The power of fuel vehicles passes through the engine, and the transmission torque is amplified and then transmitted to the wheel end.

[0003] The current HCU controller usually divides the hybrid vehicle driving mode into sport / eco / normal modes for the driver to choose, and provides different calibration parameters so that the vehicle can cover the driving needs of most drivers.

[0004] Although these modes can meet the needs of drivers to a certain extent, in the current historical environment that emphasizes personalization, for a large group of drivers, the three driving modes cannot fully meet the needs of all drivers.

[0005] Some drivers are unable to distinguish the driving experience and reasonably select the driving mode according to the actual driving conditions, and have difficulty in selecting the driving mode.

[0006] To this end, the present invention proposes a hybrid driving mode classification method, a vehicle and a storage medium. Summary of the invention

[0007] In view of the problems existing in the prior art, the present invention provides a hybrid driving mode classification method, device, vehicle and storage medium.

[0008] In order to achieve the above object, the present invention adopts the following technical solutions:

[0009] A first aspect of the present invention provides a hybrid driving mode classification method, the method comprising:

[0010] Obtain driver's personal driving data;

[0011] Normalizing the driver's personal driving data;

[0012] The normalized data is output through a big data computing model to obtain the target user's driving preference.

[0013] Furthermore, the driver's personal driving data includes user portrait indicators and driver's personal driving data indicators, the user portrait indicators include age, gender and consumption tendency, and the driver's personal driving data indicators include comfort indicators, power indicators, economy indicators and road condition indicators.

[0014] Furthermore, the driver's personal driving data indicators are recorded by a user driving data recording module, and the user driving data recording module is used to record the driver's vehicle jerk, vehicle acceleration time, fuel consumption, equivalent fuel consumption of electricity, and vehicle driving slope every 5 km, and the data of vehicle jerk, vehicle acceleration time, fuel consumption, equivalent fuel consumption of electricity, and vehicle driving slope are normalized, and the formula for the normalization operation is:

[0015]

[0016] Among them, Xmax and Xmin are the maximum and minimum values ​​that the corresponding indicators may reach during driving, which are determined by vehicle characteristics and obtained through calibration.

[0017] Furthermore, the big data computing model is composed of three groups of parallel generative adversarial networks, and the generative adversarial network includes a generator and a discriminator. The generator receives a random noise as input and generates a simulated sample. The discriminator receives a sample as input and outputs the probability that the sample is a real sample. The goal of the discriminator is to accurately distinguish between real samples and generated samples.

[0018] Furthermore, the three groups of discriminators are respectively used to discriminate whether the driving type is a power mode, an economic mode or a comfort mode.

[0019] A second aspect of the present invention provides a vehicle, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the hybrid driving mode classification method.

[0020] A third aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, the program being executed by a processor to implement the hybrid driving mode classification method.

[0021] A fourth aspect of the present invention provides a computer program product, comprising a computer program, which, when executed, is used to implement the hybrid driving mode classification method.

[0022] The above technical solution of the present invention has the following beneficial technical effects:

[0023] Since the present invention adds parameters of personal information such as user age, gender, and consumption tendency as model training features in the model, it can more accurately judge the driver's driving behavior and suitable mode. Based on the initial driving mode, the present invention can predict the user's preferences according to the actual user's driving data, thereby providing a personalized vehicle driving mode solution that meets the user's preferences, giving the user driving preference suggestions, and providing appropriate vehicle HCU calibration parameters. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 A system block diagram of a hybrid driving mode classification method proposed by the present invention; Figure 2 This is a principle flow chart of the user driving data recording module in the present invention; Figure 3 This is a principle flow chart of the big data calculation model in the present invention; Figure 4 is a system block diagram of the discriminator in the present invention; Figure 5 A schematic diagram of the model training steps of the big data computing model in the present invention; Figure 6 This is a schematic diagram of the initial model training steps before the vehicle leaves the factory in the present invention; Figure 7 Schematic diagram of the model recommendation result output and personalized training method after the vehicle leaves the factory according to the present invention. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical scheme and advantages of the present invention clearer, the present invention is further described in detail below in conjunction with specific embodiments and with reference to the accompanying drawings. It should be understood that these descriptions are only exemplary and are not intended to limit the scope of the present invention. In addition, in the following description, the description of well-known structures and technologies is omitted to avoid unnecessary confusion of the concept of the present invention.

[0026] like Figure 1-Figure 7 As shown, the present invention provides a hybrid driving mode classification method according to a first aspect of the present invention, the method comprising:

[0027] Obtain driver's personal driving data;

[0028] Normalizing the driver's personal driving data;

[0029] The normalized data is output through a big data computing model to obtain the target user's driving preference.

[0030] In a specific embodiment of the present invention, the driver's personal driving data includes user portrait indicators and the driver's personal driving data indicators. The user portrait indicators include age, gender and consumption tendency, consumption tendency parameters, and the driver's personal Engel coefficient / product type statistics for the past three years can be counted. The driver's personal driving data indicators include comfort indicators, power indicators, economy indicators and road condition indicators.

[0031] In a specific implementation of the present invention, the driver's personal driving data index is recorded by a user driving data recording module, and the user driving data recording module is used to record the driver's vehicle jerk, vehicle acceleration time, fuel consumption, equivalent fuel consumption of electricity, and vehicle driving slope every 5 km, and the data of vehicle jerk, vehicle acceleration time, fuel consumption, equivalent fuel consumption of electricity, and vehicle driving slope are normalized, and the formula for the normalization operation is:

[0032]

[0033] Among them, Xmax and Xmin are the maximum and minimum values ​​that the corresponding indicators may reach during driving, which are determined by vehicle characteristics and obtained through calibration.

[0034] In a specific embodiment of the present invention, the big data computing model is composed of three groups of parallel generative adversarial networks, and the generative adversarial network includes a generator and a discriminator. The generator receives a random noise as input and generates a simulated sample. The discriminator receives a sample as input and outputs the probability that the sample is a real sample. The goal of the discriminator is to accurately distinguish between real samples and generated samples.

[0035] In a specific embodiment of the present invention, three groups of discriminators are respectively used to discriminate whether the driving type is a power mode, an economic mode or a comfort mode.

[0036] It is worth noting that if Figure 5-Figure 7 As shown, the model training steps of the present application are: initial model training before the vehicle leaves the factory, model recommendation result output and personalized training after the vehicle leaves the factory, wherein, in the initial model training step before the vehicle leaves the factory, the initial discriminators of three driving modes are trained to generate adversarial network training data.

[0037] 1. Initial discriminator training method (existing technology):

[0038] 1) Data Labeling

[0039] Select multiple representative drivers to set the initial model data labels. Calculate the above required statistics, and ask multiple drivers to choose the driving mode preference for each road section according to their own situation.

[0040] 2) Data Normalization

[0041] The above mentioned parameters are normalized and used as the input of the neural network, and the network is trained using the steepest descent method.

[0042] The neural error function looks like this:

[0043]

[0044] 3) The inter-layer activation function uses the si gmo id function

[0045] 4) Obtain the basic discriminator neural network parameters.

[0046] After the vehicle leaves the factory, the model recommendation result output and personalized training method steps are as follows Figure 7 As shown, the arbitrator rules are: Arbitrator rule 1. If there is only one true, output the corresponding mode;

[0047] 2. If there are two or more true values, the output is selected according to the priority, comfort > economy > power;

[0048] 3. If true does not exist, the default mode when the vehicle is powered on is output.

[0049] A second aspect of the present invention provides a vehicle, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the hybrid driving mode classification method.

[0050] A third aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, the program being executed by a processor to implement the hybrid driving mode classification method.

[0051] A fourth aspect of the present invention provides a computer program product, comprising a computer program, which, when executed, is used to implement the hybrid driving mode classification method.

[0052] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention is described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A hybrid driving mode classification method, characterized in that: The method comprises: Obtain driver's personal driving data; Normalizing the driver's personal driving data; The normalized data is output through a big data computing model to obtain the target user's driving preference.

2. A hybrid driving mode classification method according to claim 1, characterized in that: The driver's personal driving data includes user portrait indicators and the driver's personal driving data indicators. The user portrait indicators include age, gender and consumption tendency. The driver's personal driving data indicators include comfort indicators, power indicators, economy indicators and road condition indicators.

3. The hybrid driving mode classification method according to claim 2, characterized in that: The driver's personal driving data index is recorded by a user driving data recording module, and the user driving data recording module is used to record the driver's vehicle jerk, vehicle acceleration time, fuel consumption, equivalent fuel consumption of electricity, and vehicle driving slope every 5 km, and normalize the data of vehicle jerk, vehicle acceleration time, fuel consumption, equivalent fuel consumption of electricity, and vehicle driving slope. The formula for the normalization operation is: Among them, Xmax and Xmin are the maximum and minimum values ​​that the corresponding indicators may reach during driving, which are determined by vehicle characteristics and obtained through calibration.

4. The hybrid driving mode classification method according to claim 3, characterized in that: The big data computing model consists of three groups of parallel generative adversarial networks, which include a generator and a discriminator. The generator receives a random noise as input and generates a simulated sample. The discriminator receives a sample as input and outputs the probability that the sample is a real sample. The goal of the discriminator is to accurately distinguish between real samples and generated samples.

5. The hybrid driving mode classification method according to claim 4, characterized in that: The three sets of discriminators are used to determine whether the driving type is power mode, economic mode or comfort mode.

6. A vehicle, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the hybrid driving mode classification method as described in any one of claims 1 to 5.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: The program is executed by a processor to implement the hybrid driving mode classification method as described in any one of claims 1 to 5.

8. A computer program product, comprising a computer program, characterized in that When the computer program is executed, it is used to implement the hybrid driving mode classification method as described in any one of claims 1-5.

Images