Charging route planning method and route processing apparatus

Abstract

A charging route planning method and a route processing apparatus are provided. A charging request message is detected. The charging request message corresponds to task period information, and the task period information indicates the initiation time of the charging request message. Power information is detected. The power information indicates the current power. Reference route information is obtained. The reference route information indicates the route from the current location that issued the charging request message to a plurality of charging stations. Target route information is determined based on the charging request message, the power information, and the reference route information. The target route information indicates the route from the current location to the target station among multiple charging stations. Therefore, charging efficiency could be improved.

Description

[Technical Field]

[0001] This invention relates to a route-related information processing technology, and more particularly to a charging route planning method and route processing device. [Previous Technology]

[0002] In recent years, with rising environmental awareness and government policy support, the number of electric vehicles has continued to grow. However, the construction of charging infrastructure has not kept pace, leading to frequent charging difficulties for electric vehicle users. For example, public transport vehicles can only charge at designated charging stations after their shifts are completed. This arrangement lacks flexibility and may cause peak usage times and congestion at charging stations. [Summary of the Invention]

[0003] The present invention provides a method for planning charging routes and a route processing device, which can provide suitable charging route suggestions by referring to various factors.

[0004] The charging route planning method of this embodiment of the invention is configured for a processor. The planning method includes (but is not limited to) the following steps: detecting a charging request message, wherein the charging request message corresponds to task period information, and the task period information indicates the initiation time of the charging request message; detecting power information, wherein the power information indicates the current power level; obtaining reference route information, wherein the reference route information indicates a route from the current location where the charging request message was issued to a plurality of charging stations; and determining target route information based on the charging request message, the power information, and the reference route information, wherein the target route information indicates a target station among the plurality of charging stations from the current location.

[0005] The route processing apparatus of this embodiment includes (but is not limited to) a memory and a processor. The memory stores program code. The processor is coupled to the memory. The processor loads the program code and executes: detecting a charging request message; detecting power information, wherein the power information indicates the current power level; obtaining reference route information; and determining target route information based on the charging request message, the power information, and the reference route information. The charging request message corresponds to task period information, and the task period information indicates the initiation time of the charging request message. The reference route information indicates a route from the current location where the charging request message was issued to a plurality of charging stations. The target route information indicates a route from the current location to a target station among the plurality of charging stations.

[0006] Based on the above, the charging route planning method and route processing device of this embodiment determine the target route information from the current location to the target (charging) station based on the initiation time of the charging request message, the current battery level, and route information to multiple charging stations. This allows for suitable charging station selection and improves passenger scheduling efficiency.

[0007] In order to make the above features and advantages of the present invention more apparent and understandable, specific embodiments are described below, and detailed descriptions are given in conjunction with the accompanying drawings.

Implementation Method

[0008] FIG1 is a component diagram of a system 1 according to an embodiment of the present invention. Referring to FIG1, the system 1 includes (but is not limited to) one or more mobile devices 10, one or more charging station systems 30, one or more route information supply devices 50, and route processing devices 100.

[0009] The mobile device 10 may be a mobile phone, tablet computer, laptop computer, voice assistant device, wearable device, in-vehicle system, or other electronic device. In one embodiment, the mobile device 10 is located in a vehicle. In another embodiment, the mobile device 10 is an in-vehicle system on a vehicle. The vehicle may be an electric vehicle, such as a locomotive, automobile, bus, or other type of mobility vehicle, or a plug-in hybrid vehicle.

[0010] The charging station system 30 includes one or more power supply devices (e.g., charging piles), battery swapping devices, or combinations thereof. The power supply devices can be used to provide electrical energy to electric vehicles or plug-in hybrid vehicles. The battery swapping devices provide charging and placement of swappable batteries. The charging station system 30 may further include a computer system to monitor the operating status of the power supply devices, such as whether charging is in progress, charging has stopped, or the devices are unoccupied.

[0011] The route information supply device 50 is a server or other computer system. In one embodiment, the route information supply device 50 provides at least one of the following functions: calculation of route distance and route travel time, route planning, traffic information, or traffic condition prediction. For example, when transmitting origin and destination information to an electronic map service server, the server can calculate and return the route distance and route travel time. It should be noted that route planning may take into account traffic conditions, constraints (e.g., avoiding highways or prioritizing energy conservation) or other factors, and this embodiment of the invention is not limited thereto.

[0012] The route processing device 100 includes (but is not limited to) a communication transceiver 110, a memory 120, and a processor 130.

[0013] The communication transceiver 110 may be a communication transceiver supporting technologies such as fourth-generation (4G) or other generations of mobile communications, Wi-Fi, Bluetooth, infrared, radio frequency identification (RFID), Ethernet, fiber optic networks, and serial communication interfaces (e.g., RS-232). It may also be a Universal Serial Bus (USB), Thunderbolt, or other communication transmission interfaces. In one embodiment, the communication transceiver 110 is used to transmit or receive data with other electronic devices (e.g., mobile device 10, charging station system 30, or route information supply device 50).

[0014] The memory 120 can be any type of fixed or removable random access memory (RAM), read-only memory (ROM), flash memory, hard disk drive (HDD), solid-state drive (SSD), or similar element. In one embodiment, the memory 120 is used to store code, software modules, configuration settings, data (e.g., messages, power / route information, or model parameters), or files, and embodiments thereof will be described in detail later.

[0015] The processor 130 may be a central processing unit (CPU), a graphics processing unit (GPU), or other programmable general-purpose or special-purpose microprocessors, digital signal processors (DSPs), programmable controllers, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), neural network accelerators, or other similar components or combinations thereof. In one embodiment, the processor 130 is used to perform all or part of the operations of the circuit processing device 100, and can load and execute various code, software modules, files, and data stored in the memory 120. In some embodiments, the functionality of the processor 130 may be implemented through software or a chip.

[0016] Figure 2 is a schematic diagram of a passenger transport application scenario according to an embodiment of the present invention. Referring to Figure 2, the passenger transport is an electric vehicle EVn (n is a positive integer greater than zero), and the driver inside the electric vehicle EVn carries a mobile device 10 or the electric vehicle EVn is equipped with a mobile device 10. The mobile device 10 is connected to the electric bus charging navigation system provided by the route processing device 100 and shares electric bus information. On the other hand, the charging station system 30 (distinguished as charging station CSn) is also connected to this charging navigation system to share information on the availability of charging equipment. The charging station CSn can be set up near the bus terminal, and there may be multiple charging stations near the bus terminal. Depending on actual needs, the route processing device 100 can additionally connect to a database 90, and the database 90 can store, for example, route information or power supply information.

[0017] However, the scenario shown in Figure 2 is only for illustrative purposes, and its quantity and function can still be adjusted according to actual needs.

[0018] The methods described in the embodiments of the present invention will be described below in conjunction with the various systems, devices, components and modules in System 1. The processes of this method may be adjusted according to the implementation situation.

[0019] Figure 3 is a flowchart of a charging route planning method according to an embodiment of the present invention. Referring to Figure 3, the processor 130 detects a charging request message (step S310). Specifically, the mobile device 10 may issue a charging request message in response to user operation or triggering conditions. The mobile device 10 may provide a message sending option, and the user operation is the operation of the user selecting the message sending function (e.g., pressing or tapping the message sending option on the user interface). The triggering conditions are, for example, that the current battery level of the vehicle is less than a corresponding threshold value, entering a specified time interval, or entering a specified geographical area, and are not limited thereto. The processor 130 may receive the charging request message from the mobile device 10 via a network through the transceiver 110.

[0020] In one embodiment, the charging request message corresponds to task duration information, and the task duration information indicates the initiation time of the charging request message. The initiation time can be the time of receiving a user operation, meeting a triggering condition, or sending a message, and is not limited thereto. Assume the task is to complete the charging operation at charging station CSn. Alternatively, assume the task is to arrive at the destination (which could be any designated location) after completing the charging operation at charging station CSn. However, the content of the task can still be adjusted according to actual needs.

[0021] In one embodiment, the task duration information corresponds to the time interval from the initiation time of the charging request message to the completion time of charging. The completion time of charging also represents the task completion time (point).

[0022] In one embodiment, the charging request message further indicates driving information. Driving information indicates the driver's vehicle usage behavior. For example, …(1)…(2)…(3), where is the set of driver information, is the information of the nth driver, DR is the set of driving information, is the driving information corresponding to the nth driver information, and is the occupancy information corresponding to the nth driver information. Occupancy information is the charging occupancy behavior previously recorded. For example, the number of times the charging device has been occupied.

[0023] The processor 130 detects battery information (step S320). Specifically, the battery information indicates the current battery level of the vehicle. This vehicle is the vehicle of the mobile device 10 that issued the charging request message. In one embodiment, the charging request message further indicates the battery information. In another embodiment, the mobile device 10 transmits the battery information to the route processing device 100 periodically or intermittently.

[0024] In one embodiment, the power information further indicates the target power corresponding to the charging request message. The target power may be a value selected by the user or a preset value.

[0025] In one embodiment, the power information further indicates charging specification information. The charging specification information indicates at least one of the following: battery capacity or socket type. For example, …(4)…(5), where is a set of charging specification information for electric vehicles, is the charging specification information for the nth electric vehicle, is the battery capacity corresponding to the charging specification information for the nth electric vehicle, and is the socket type corresponding to the charging specification information for the nth electric vehicle.

[0026] In one embodiment, the processor 130 obtains charging station information through the communication transceiver 110. The charging station system 30 transmits the charging station information. For example, …(6)…(7), where is the nth charging station, is the longitude corresponding to the nth charging station, is the latitude corresponding to the nth charging station, is the socket type provided by the nth charging station, is the number of power supply devices provided by the nth charging station, and is an indication of whether the nth charging station is in an available state.

[0027] The processor 130 obtains reference route information (step S330). Specifically, the reference route information indicates the route from the current location of the vehicle that issued the charging request message to multiple charging stations. The processor 130 can transmit the vehicle's current location (and the locations of multiple charging stations) to the route information supply device 50 via the transceiver 110. The route information supply device 50 can output the corresponding reference route information. For example, the reference route information indicates the route distance and travel time from the current location to multiple charging stations. For example, the route information supply device 50 provides an application program interface (API) and is used to provide reference route information based on the starting location and the ending location. The reference route information is, for example, …(8), where is the reference route information, is the longitude and latitude of the starting location (e.g., the longitude and latitude of the vehicle's current location), is the longitude and latitude of the ending location (e.g., the longitude and latitude of the charging station), is the route distance, and is the travel time from the starting location to the ending location.

[0028] In one embodiment, the reference route information further indicates the route from multiple charging stations to the destination corresponding to the charging request message. The destination may be, for example, a bus terminal, the driver's residence, or other locations. The processor 130 may transmit the locations of the multiple charging stations and the destination to the route information supply device 50 via the transceiver 110. The route information supply device 50 may output the corresponding route information. For example, the reference route information indicates the route distance and travel time from the multiple charging stations to the destination.

[0029] In one embodiment, the processor 130 can aggregate various information into charging request information. For example, …(9)…(10), where is a set of charging request information, is the charging request information of the nth electric vehicle, is the initiation time of the charging request message, is the completion time (representing the latest time point when charging needs to be completed), is the battery capacity corresponding to the nth charging specification information (m is a positive integer greater than zero), is the socket type corresponding to the nth charging specification information, is the current power (percentage) corresponding to the nth charging request information, is the target power (percentage) corresponding to the nth charging request information, is the longitude of the starting position (i.e., the current position) corresponding to the nth charging request information, is the latitude of the starting position (i.e., the current position) corresponding to the nth charging request information, is the longitude of the ending position (i.e., the destination position) corresponding to the nth charging request information, and is the latitude of the ending position (i.e., the destination position) corresponding to the nth charging request information.

[0030] In addition, …(11), where is the route distance from the starting position corresponding to the nth charging request information to the mth charging station, is the travel time from the starting position corresponding to the nth charging request information to the mth charging station, and is a function or rule used to calculate the route distance and travel time. …(12), where is the route distance from the mth charging station to the ending position corresponding to the nth charging request information, is the travel time from the mth charging station to the ending position corresponding to the nth charging request information. …(13), where is the route distance from the starting position corresponding to the nth charging request information through the mth charging station to the ending position corresponding to the nth charging request information.

[0031] In one embodiment, the processor 130 can predict task period information. The task period information corresponds to the time interval from the initiation time of the charging request message to the completion time of charging, and the completion time of charging is determined based on the initiation time of the charging request message, the power information, and the reference route information. For example, FIG4 is a schematic diagram illustrating various situations according to an embodiment of the present invention. Referring to FIG4, the processor 130 can determine the charging period based on the power information, and the power information further indicates the target power corresponding to the charging request message: …(14), where is the charging period, is the total battery capacity corresponding to the nth charging request message, is the current power (percentage) corresponding to the nth charging request message, is the target power (percentage) corresponding to the nth charging request message, is the route distance from the starting position to the mth charging station corresponding to the nth charging request message, is the energy consumption rate required for driving, and is the charging speed rate.

[0032] The processor 130 can determine the travel time based on the reference route information. For example, …(15), where is the vehicle's travel time, is the route distance from the starting position corresponding to the nth charging request information through the mth charging station to the ending position corresponding to the nth charging request information, and is the vehicle's average speed.

[0033] Next, the processor 130 can determine the charging completion time based on the charging period and the driving period. For example, …(16), where is the time when the vehicle arrives at the charging station, is the driving time of the vehicle, and is the distance traveled from the starting position corresponding to the nth charging request information to the mth charging station. …(17), where is the charging completion time, is the time when the charging request message is initiated, and is the time required for the task (e.g., the time interval from the time when the charging request message is initiated to the time when charging is completed). …(18), where is the time required for the task, is the driving time of the vehicle, and is the charging period. …(19), where is the early completion time or idle time, is the target completion time, and is the charging completion time. That is, the early completion time or idle time is the time interval from the charging completion time to the target completion time.

[0034] Furthermore, the waiting time is the interval between the vehicle's arrival at the charging station and the time it waits and is late (as shown in Figure 4, "late arrival"). The occupancy time is the interval between the vehicle's occupancy of the charging station (as shown in Figure 4, "occupancy"). In one embodiment, the processor 130 further obtains occupancy information. The occupancy information is the charging occupancy behavior previously recorded, and the charging completion time can also be determined based on the occupancy information. For example, the length of the occupancy time in the charging completion time is determined based on the number of charging occupancy behaviors.

[0035] In one embodiment, the charging station information indicates the number of charging stations available for charging. To allow vehicles to select a charging station, the number of available power supply devices (i.e., the number of charging stations) in the charging station may also be considered. The number of available charging stations may be updated in real time. For example, …(20), where is the number of available charging stations, is the number of power supply devices provided by the nth charging station, and is the number of vehicles charging at the nth charging station (possibly including reserved capacity). Different time periods are considered: …(21), where is the set of time periods for the number of charging stations available for charging, and is the number of charging stations available for charging at time period t.

[0036] Referring to Figure 3, the processor 130 determines the target route information based on the charging request message, power information, and reference route information (step S340). Specifically, the target route information indicates the route from the current location to a target station among multiple charging stations. That is, the processor 130 analyzes the charging request message, power information, and reference route information, determines the corresponding target route information, and uses it to select the target station.

[0037] In one embodiment, the processor 130 may input charging request messages, battery level information, and reference route information into a machine learning model. The machine learning model may be based on Deep Reinforcement Learning (DRL) (e.g., Deep Q Network (DQN), Convolutional Neural Network (CNN), or Generative Adversarial Network (GAN), and is not limited thereto). The machine learning model is built by learning the relationships between task-period information, current battery level, multiple reference routes, and labeled routes within these reference routes. For example, training samples include known task-period information, known battery level information, reference route information, and known selected charging stations. During the learning process, the machine learning model understands the relationships between known task-period information, known battery level information, reference route information, and known target stations. The machine learning algorithm can analyze the relationship between training samples (e.g., a dataset) and corresponding labels (e.g., a charging station) or real results to... By identifying patterns, the machine learning model can predict unknown data. This model can be used to infer from the data being evaluated (e.g., charging request messages, battery level information, and reference route information) to generate predictions. These predictions consist of one or more charging stations and their confidence scores. The confidence score reflects the machine learning model's confidence in the predicted charging station and / or the accuracy of the prediction. A higher confidence score for a charging station indicates greater confidence that the prediction is correct. The charging station with the highest confidence score can be selected as the target station. Conversely, a lower confidence score for a charging station indicates less confidence that the prediction is correct.

[0038] In one embodiment, the processor 130 can determine the target route information based on the charging request message, the power information, the route information, and the charging station information. For example, …(22)…(23), where is the input data to the machine learning model, is the initiation time of the charging request message, is the target completion time, is the current power of the vehicle, is the target power corresponding to the charging request message, is the charging station information, is the route distance from the starting position through the j-th charging station to the destination position, is the travel time from the starting position corresponding to the n-th charging request message through the j-th charging station to the destination position corresponding to the n-th charging request message, and is the time period set of the number of charging stations that can be charged at the j-th charging station.

[0039] In one embodiment, the processor 130 may perform data transformation on the input data through standardization and / or normalization. The processor 130 may select the transformation method according to the data characteristics of the input data to improve the convergence speed of the model. For the time data portion, the processor 130 may perform data transformation through normalization. For example, converting time-related data into values ​​between 0 and 1 to improve the convergence speed. For other data portions, the processor 130 may perform data transformation through standardization. For example, using Z-Score as the standardization method. This method narrows the range of data features, making it easier for the model to converge during training.

[0040] Taking DQN as an example, the machine learning model uses state, action, and reward function as three important inputs. State refers to the state of the environment; it can be any information that describes the environment, such as input data.

[0041] In one embodiment, the processor 130 may set multiple action information. Each action information indicates the selection behavior of one of the multiple charging stations. An action is an operation that an agent can take in the environment. For example, ... (24), where is the action information and is the action of selecting the j-th charging station.

[0042] In one embodiment, the processor 130 can determine reward information based on the charging request message, power information, and reference route information. The reward information indicates the reward obtained by selecting multiple action information, and the reward information is used to retrain the machine learning model. During model training, the collected data is divided into a training group and a testing group. The training group is used for model training, and different reward values ​​are obtained by selecting different actions. The machine learning model records what action is best in different states. After the training group is trained, the testing group is used to verify whether the machine learning model has been trained as required (i.e., whether the most suitable action has been selected). If the requirements are met, it is determined that the machine learning model has been trained successfully.

[0043] Reward information indicates the reward obtained by the agent after taking an action in the environment. In one embodiment, the processor 130 assigns a higher reward value to a shorter task period and a lower reward value to a longer task period. For example, the reward function is shown in formula (25), which is designed to achieve the earliest completion time of the task and the earliest arrival at the charging station. …(25), where is the reward function, is the input data or status, is the action information, refers to the early completion time or idle time of the nth charge, and refers to the travel time of the nth charge. The time from the completion time of the charging time to the target completion time is less than or equal to zero, which represents a longer task period (i.e., no early completion time or idle time), and therefore is given a reward value of -1. The time from the completion time of the charging time to the target completion time is greater than zero, which represents a shorter task period (i.e., a longer early completion time or idle time), and therefore is given a reward value greater than zero.

[0044] The reward function can be, for example, reward function (1): (subtracting travel time from task time, aiming for shorter travel time to complete the task) or reward function (2): (setting the reward value as longer lead time and shorter travel time). Table (1) shows the experimental data results, with a total of 8651 data points (including 8204 training data points and 447 test data points): Table (1) reward function Model learning rate Completion rate Reward Value 1 0.001 92.4% 3028 0.0005 92.2% 2820 2 0.001 93.0% 2368 0.0005 94.2% 3271

[0045] Figure 5 is a comparison chart of the completion rate between the embodiments of the present invention and the prior art. Referring to Figure 5, the comparison of task completion rates (e.g., defining task completion as charging to the target level within the scheduled time and counting the proportion of completed tasks to the total count) shows that a higher task completion rate indicates closer proximity to the target and higher scheduling and charging efficiency. The task completion rate of Embodiment 501 of the present invention is higher than that of the RLCharge model 502 and the Nearest method 503.

[0046] Figure 6 is a comparison chart of the time required for a task between the embodiment of the present invention and the prior art. Referring to Figure 6, the shorter the time required for a task, the closer it is to the target, and the higher the scheduling and charging efficiency. The task time required by embodiment 601 of the present invention is shorter than that using the RLCharge model 602 and the Nearest method 603.

[0047] In one embodiment, the processor 130 can transmit control commands via the transceiver 110 based on target route information. These control commands are used to drive the vehicle. For example, the vehicle system generates navigation information on the user interface based on the control commands. Alternatively, the vehicle system drives the motor and steering wheel based on the control commands to move the vehicle.

[0048] In summary, the charging route planning method and route processing apparatus of this invention comprehensively consider task-period information, power information, reference route information, and charging station information, and provide suitable charging stations and routes to the charging stations. This improves charging convenience and satisfaction, enhances scheduling and charging efficiency, and ultimately promotes the development of the electric vehicle market.

[0049] Although the present invention has been disclosed above by way of embodiments, it is not intended to limit the present invention. Anyone skilled in the art can make some modifications and refinements without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention shall be determined by the appended claims. [Simplified Explanation of the Diagram]

[0050] Figure 1 is a component diagram of a system according to an embodiment of the present invention. Figure 2 is a schematic diagram of a passenger transport application scenario according to an embodiment of the present invention. Figure 3 is a flowchart of a charging route planning method according to an embodiment of the present invention. Figure 4 is a schematic diagram illustrating various situations according to an embodiment of the present invention. Figure 5 is a comparison chart of the completion rate between the embodiment of the present invention and the prior art. Figure 6 is a comparison chart of the time required for a task between the embodiment of the present invention and the prior art.

Claims

1. A method for planning a charging route, configured for a processor, the method comprising: The steps of detecting a charging request message, wherein the charging request message corresponds to task period information and the task period information indicates the initiation time of the charging request message; detecting power information, wherein the power information indicates the current power level; obtaining reference route information, wherein the reference route information indicates a route from a current location that issued the charging request message to a plurality of charging stations; and determining a target route information based on the charging request message, the power information, and the reference route information, wherein the target route information indicates a route from the current location to a target station among the charging stations, and determining the target route information based on the charging request message, the power information, and the reference route information include: inputting the charging request message, the power information, and the reference route information into a machine learning model, wherein the machine learning model is established by learning the association between the task period information, the current power level, the plurality of reference routes, and the labeled routes among the reference routes; outputting the target route information using the machine learning model; and setting a plurality of action information, wherein each action information indicates a selection behavior of one of the charging stations; Based on the charging request message, the power information, and the reference route information, a reward information is determined, wherein the reward information indicates the reward value obtained by selecting these action information; and in model training, the machine learning model is retrained using the reward information.

2. The method for planning a charging route as described in claim 1, wherein the step of determining the reward information based on the charging request message, the power information, and the reference route information includes: A shorter task duration corresponds to a higher reward value; And a longer task duration corresponds to a lower reward value.

3. The method for planning charging routes as described in claim 1 further includes: The task period information is predicted, wherein the task period information corresponds to the time interval from the initiation time of the charging request message to a charging completion time, and the charging completion time is determined based on the initiation time of the charging request message, the power information, and the reference route information.

4. The method for planning charging routes as described in claim 3 further includes: The system determines a charging period based on the power information, wherein the power information further indicates the target power corresponding to the charging request message; determines a driving period based on the reference route information; and determines the charging completion time based on the charging period and the driving period.

5. The method for planning charging routes as described in claim 3 further includes: Obtain occupancy information, wherein the occupancy information is a previously recorded charging occupancy behavior, and the charging completion time is also determined based on the occupancy information.

6. The method for planning a charging route as described in claim 1, wherein the reference route information further indicates the route from the charging stations to the destination corresponding to the charging request message.

7. The method for planning a charging route as described in claim 1 further includes: The steps of obtaining charging station information, wherein the charging station information indicates the number of charging stations available for charging, and determining target route information based on the charging request message, the power information, and the reference route information include: determining a target route information based on the charging request message, the power information, the route information, and the charging station information.

8. The method for planning a charging route as described in claim 1, further comprising: The target route information is transmitted to a control command, which is used to drive a vehicle.

9. A route processing device, comprising: One memory, storing one piece of code; The processor, coupled to the memory, loads the code and executes: detecting a charging request message, wherein the charging request message corresponds to task duration information and the task duration information indicates the initiation time of the charging request message; detecting battery level information, wherein the battery level information indicates the current battery level; obtaining reference route information, wherein the reference route information indicates a route from a current location that issued the charging request message to a plurality of charging stations; and determining a target route information based on the charging request message, the battery level information, and the reference route information, wherein the target route information indicates a route from the current location to a target station among the charging stations, and the processor further executes: inputting the charging request message, the battery level information, and the reference route information into a machine learning model, wherein the machine learning model is established by learning the association between the task duration information, the current battery level, the plurality of reference routes, and the labeled routes among the reference routes; outputting the target route information using the machine learning model; setting a plurality of action information, wherein each action information indicates a selection behavior of one of the charging stations; A reward is determined based on the charging request message, the battery level information, and the reference route information, wherein the reward information indicates the reward value obtained by selecting these action information; during model training, the machine learning model is retrained using the reward information.

10. The route processing apparatus as claimed in claim 9, wherein the processor further performs: assigning a shorter task duration to a higher reward value; and assigning a longer task duration to a lower reward value.

11. The route processing apparatus of claim 9, wherein the processor further performs: predicting task period information, wherein the task period information corresponds to a time interval from the initiation time of the charging request message to a completion charging time, and the completion charging time is determined based on the initiation time of the charging request message, the power information, and the reference route information.

12. The route processing apparatus of claim 11, wherein the processor further performs: determining a charging period based on the power information, wherein the power information further indicates the target power corresponding to the charging request message; determining a travel period based on the reference route information; and determining the charging completion time based on the charging period and the travel period.

13. The route processing apparatus as claimed in claim 11, wherein the processor further performs: obtaining occupancy information, wherein the occupancy information is a previously recorded charging occupancy behavior, and the charging completion time is also determined based on the occupancy information.

14. The route processing apparatus as claimed in claim 9, wherein the reference route information further indicates the route from the charging stations to the destination corresponding to the charging request message.

15. The route processing apparatus of claim 9, wherein the processor further performs: obtaining charging station information, wherein the charging station information indicates the number of charging stations available for charging; and determining a target route information based on the charging request message, the power information, the route information and the charging station information.

16. The route processing apparatus of claim 9, wherein the processor further performs: transmitting a control command based on the target route information, wherein the control command is used to drive a vehicle.

Images