On-vehicle device, energy consumption prediction system, energy consumption prediction method, and energy consumption prediction program

Abstract

A first domain controller (9) is capable of performing data communication with a server (4) that learns statistical quantities including data obtained while a vehicle is traveling and calculates feature quantities necessary for predicting energy consumption of the vehicle. The first domain controller (9) includes: a link information acquisition unit (12a) that acquires link information indicating links included in map information; a statistical quantity calculation unit (12b) that calculates statistical quantities for each link indicated by the link information; a statistical quantity transmission unit (12c) that causes the statistical quantities to be transmitted to the server; a feature quantity acquisition unit (12d) that acquires feature quantities from the server; and an energy consumption prediction unit (12g) that predicts energy consumption of the vehicle on the basis of the feature quantities.

Description

Vehicle-mounted device, energy consumption prediction system, energy consumption prediction method, and energy consumption prediction program Cross-reference to related applications

[0001] This application is based on Japanese Application No. 2024-203247 filed on November 21, 2024, the contents of which are incorporated herein by reference.

[0002] The present disclosure relates to a vehicle-mounted device, an energy consumption prediction system, an energy consumption prediction method, and an energy consumption prediction program.

[0003] In order to predict the cruising range of a vehicle, it is necessary to accurately predict the energy consumption of the vehicle. If the energy consumption of the vehicle can be accurately predicted, the cruising range of the vehicle can be predicted from the relationship between the predicted energy consumption and the remaining amount of fuel or battery at a reference point such as the current location or the departure point. For example, in Patent Document 1, a technique is disclosed in which a server learns data during traveling to calculate an individual correction coefficient as a feature amount, and uses the calculated individual correction coefficient to predict the energy consumption of the vehicle.

[0004] Japanese Patent Application Laid-Open No. 2014-163796

[0005] In the technique of Patent Document 1, it is necessary for the vehicle-mounted device to periodically calculate the data during traveling required for learning at a predetermined cycle. However, for example, when calculating the data during traveling required for learning at a short cycle, the amount of data becomes excessive for the purpose of energy consumption prediction. As a result, there are problems such as an increase in communication cost, an increase in the processing load on the communication network side, an increase in cost due to an increase in the processing load on the server side and the amount of stored data, and it becomes difficult for developers to handle data due to an increase in the data preprocessing load when considering improving the prediction model. On the other hand, when calculating the data during traveling required for learning at a long cycle, the amount of data becomes insufficient for the purpose of predicting energy consumption, and as a result, there are problems such as deterioration of the prediction accuracy of energy consumption.

[0006] An object of the present disclosure is to optimize the amount of data during traveling required for learning and appropriately predict the energy consumption of a vehicle.

[0007] One aspect of this disclosure enables data communication with an external device that learns statistical quantities including data from the vehicle while it is running and calculates feature quantities necessary for predicting the vehicle's energy consumption. The in-vehicle device includes a link information acquisition unit that acquires link information indicating links included in map information; a statistical quantity calculation unit that calculates the statistical quantities on a link-by-link basis indicated by the link information; a statistical quantity transmission unit that transmits the statistical quantities to the external device; a feature quantity acquisition unit that acquires the feature quantities from the external device; and an energy consumption prediction unit that predicts the vehicle's energy consumption based on the feature quantities.

[0008] One aspect of this disclosure comprises an in-vehicle device and an external device that learns statistical quantities including data while the vehicle is running and calculates feature quantities necessary for predicting the energy consumption of the vehicle, wherein the in-vehicle device and the external device are configured to communicate with each other. The energy consumption prediction system comprises a link information acquisition unit that acquires link information indicating links included in map information, a statistical quantity calculation unit that calculates the statistical quantities in units of links indicated by the link information, and an energy consumption prediction unit that predicts the energy consumption of the vehicle based on the feature quantities.

[0009] One aspect of the present disclosure is a method for predicting energy consumption in a system configured to enable data communication between an in-vehicle device and an external device that learns statistical quantities including data while the vehicle is in motion and calculates feature quantities necessary for predicting the energy consumption of the vehicle, the method comprising: a link information acquisition procedure for acquiring link information indicating links included in map information; a statistical quantity calculation procedure for calculating the statistical quantities in units of links indicated by the link information; and an energy consumption prediction procedure for predicting the energy consumption of the vehicle based on the feature quantities.

[0010] One aspect of the present disclosure involves causing a control unit of an in-vehicle device, which is configured to communicate data with an external device that learns statistical quantities including data from the vehicle while it is running and calculates feature quantities necessary for predicting the energy consumption of the vehicle, to execute a link information acquisition procedure for acquiring link information indicating links included in map information; a statistical quantity calculation procedure for calculating the statistical quantities in units of links indicated by the link information; a statistical quantity transmission procedure for transmitting the statistical quantities to the external device; a feature quantity acquisition procedure for acquiring the feature quantities from the external device; and an energy consumption prediction procedure for predicting the energy consumption of the vehicle based on the feature quantities.

[0011] According to one aspect of this disclosure, statistical quantities for calculating the features necessary for predicting vehicle energy consumption are calculated on a per-link basis. By calculating statistical quantities on a per-link basis, the amount of driving data required for training can be optimized, and vehicle energy consumption can be accurately predicted.

[0012] The above-mentioned objectives and other objectives, features and advantages of this disclosure will become clearer with reference to the attached drawings and the detailed description below. The drawings are as follows: Figure 1 is a diagram showing the overall configuration of one embodiment; Figure 2 is a functional block diagram showing the configuration of the vehicle system; Figure 3 is a functional block diagram showing the configuration of the first domain controller; Figure 4 is a diagram showing links; Figure 5 is a diagram showing statistics; Figure 6 is a diagram showing the calculation formula for determining energy consumption; Figure 7 is a functional block diagram showing the configuration of the server; Figure 8 is a flowchart showing the link ID determination process performed by the first domain controller; Figure 9 is a flowchart showing the statistics transmission determination process performed by the first domain controller; Figure 10 is a flowchart showing the energy consumption prediction process performed by the first domain controller; Figure 11 is a flowchart showing the statistics storage process performed by the server; Figure 12 is a flowchart showing the feature calculation process performed by the server; and Figure 13 is a flowchart showing the feature distribution determination process performed by the server.

[0013] Hereinafter, one embodiment will be described with reference to the drawings. As shown in Figure 1, the energy consumption prediction system 1 is configured so that a vehicle system 3, which is installed in an unspecified number of vehicles 2, and a server 4 (corresponding to an external device) can communicate data via a communication network 5 including the Internet. The vehicle 2 is, for example, an electric vehicle (EV) that runs using electrical energy stored in a battery as a power source.

[0014] The vehicle system 3 can also communicate data via the communication network 5 with the weather information distribution server 6, which distributes weather information, and the traffic congestion information distribution server 7, which distributes traffic congestion information. The weather information distributed from the weather information distribution server 6 includes information such as the area and time of rainfall and snowfall. The traffic congestion information distributed from the traffic congestion information distribution server 7 includes information such as the length and time of traffic congestion. Both the weather information and the traffic congestion information affect the energy consumption of the vehicle 2. The weather information distributed from the weather information distribution server 6 and the traffic congestion information distributed from the traffic congestion information distribution server 7 may be temporarily stored in server 4 and then distributed to the vehicle system 3 via server 4.

[0015] The configuration of the vehicle system 3 will now be described. As shown in Figure 2, the vehicle system 3 is configured such that a communication terminal 8, a first domain controller 9 (corresponding to an in-vehicle device), and a second domain controller 10 (corresponding to a driving control device) are connected via a communication bus 11 to enable data communication. The communication bus 11 is, for example, a CAN (Controller Area Network) bus.

[0016] The communication terminal 8 controls the connection of the communication line to the communication network 5 and controls data communication via the communication network 5 between the server 4, weather information distribution server 6, and traffic congestion information distribution server 7. The first domain controller 9 includes a function block that controls the navigation function and a function block that controls the energy consumption prediction of the vehicle 2. The second domain controller 10 includes a function block that controls the driving of the vehicle 2. In this embodiment, a configuration in which the function block that controls the navigation function and the function block that controls the energy consumption prediction of the vehicle 2 are integrated into one domain controller is illustrated, but they may also be located in separate domain controllers. Furthermore, in this embodiment, a configuration in which the function block that controls the energy consumption prediction of the vehicle 2 and the function block that controls the driving of the vehicle 2 are located in separate domain controllers is illustrated, but they may also be integrated into a single domain controller.

[0017] As shown in Figure 3, the first domain controller 9 comprises a first control unit 12, a GNSS (Global Navigation Satellite System) receiving unit 13, a vibration gyroscope 14, a vehicle speed sensor 15, a hard disk drive (HDD) 16, a display monitor 17, a speaker 18, and an input device 19. The first control unit 12 is mainly composed of a microcontroller having a CPU, ROM, RAM, and I / O, and controls the operation of the first domain controller 9 by performing software processing by executing computer programs stored in a non-transitional physical storage medium with the CPU, and by performing hardware processing control using dedicated electronic circuits.

[0018] When the GNSS receiving unit 13 receives a GNSS signal transmitted from a GNSS satellite orbiting the sky, it outputs various information contained in the received GNSS signal to the first control unit 12. The various information output from the GNSS receiving unit 13 to the first control unit 12 is information necessary for calculating the current position of the vehicle 2, and includes information such as the position of the GNSS satellite that transmitted the GNSS signal and the transmission time.

[0019] The vibration gyro 14 is a sensor that detects the angular velocity of the vehicle 2, and when it detects the angular velocity of the vehicle 2, it outputs the detection result to the first control unit 12. The vehicle speed sensor 15 is a sensor that detects the driving speed of the vehicle 2, and when it detects the driving speed of the vehicle 2, it outputs the detection result to the first control unit 12. The first control unit 12 calculates the current position of the vehicle 2 based on various information input from the GNSS receiver 13, the detection result of the angular velocity of the vehicle 2 input from the vibration gyro 14, and the detection result of the driving speed of the vehicle 2 input from the vehicle speed sensor 15.

[0020] The HDD 16 is a non-volatile recording medium on which various data, such as control programs and map data, are stored for execution of various processes in the first control unit 12. The control programs and various data stored in the HDD 16 are read out as needed by the control of the first control unit 12 and used for various processes and controls executed by the first control unit 12.

[0021] The map data recorded on HDD16 includes route calculation data, road data, and background data. Route calculation data is used, for example, when searching for a recommended route from the current location to a destination. Road data is data that shows road shape, road type, etc. Background data is data that shows the background of the map. The background of the map refers to various elements other than roads that exist on the map, such as rivers, railways, green belts, and various structures. In the above configuration, the example is that the map data is recorded on HDD16, but the map data may also be recorded on a recording medium other than HDD16, for example, on a CD-ROM, DVD-ROM, memory card, etc.

[0022] As shown in Figure 4, the smallest unit of each road in map data is called a link, and each link is assigned a unique link ID. Each road is composed of multiple links corresponding to a predetermined road section, and route calculation data and road data are represented at the link level. The points connecting links, i.e., the endpoints of each link, are called nodes, and each node is assigned a unique node ID. In Figure 4, for example, link L1 is defined as the link between nodes N1 and N3. The information of each node in the map data includes coordinate information as location information. In addition, points called shape interpolation points may be set between nodes within a link as needed. The information of each shape interpolation point in the map data includes coordinate information as location information, similar to nodes. The shape of each link is determined as the road shape based on the location information of the nodes and shape interpolation points.

[0023] The route calculation data includes a link cost for each link corresponding to each road section, based on the estimated travel time for vehicle 2 to travel through that section. Based on the link costs, a combination of links is calculated according to pre-set route search conditions, and a recommended route is searched. For example, if the route search condition prioritizes the shortest travel time, the combination of links that minimizes the travel time from the starting point to the destination is calculated as the recommended route.

[0024] The display monitor 17 is positioned in a location easily visible to the user, such as on the dashboard or within the instrument panel of the vehicle 2, and is composed of an LCD display or the like. When a display command is input to the display monitor 17 from the first control unit 12, it displays various screens, such as a map screen or a route guidance screen, based on the input display command. When an audio output command is input to the speaker 18 from the first control unit 12, it outputs various sounds, such as warning sounds or route guidance voices, based on the input audio output command.

[0025] The input device 19 is a device for the user to perform various input operations related to the navigation function, and is composed of a touch panel or remote control integrated with the display monitor 17, and has various input switches. By operating the input device 19, the user can, for example, input the name of a facility or location to be set as a destination, set search conditions for recommended routes, select a destination from pre-registered locations, or scroll the map in any direction.

[0026] When the user operates the input device 19 to set the destination and route search conditions, the first control unit 12 uses the vehicle 2's current position as the starting point and performs a route search based on route calculation data using a predetermined algorithm to find a recommended route from the starting point to the destination. Once the first control unit 12 has found a recommended route, it displays the recommended route on the map displayed on the display monitor 17 in a way that makes it distinguishable from other roads, for example by changing the color. The first control unit 12 guides the vehicle 2 to the destination by displaying various screens on the display monitor 17 and outputting various sounds from the speaker 18 according to the recommended route.

[0027] The first control unit 12 includes a link information acquisition unit 12a, a statistical calculation unit 12b, a statistical transmission unit 12c, a feature acquisition unit 12d, a weather information acquisition unit 12e, a traffic congestion information acquisition unit 12f, and an energy consumption prediction unit 12g. Each of these units 12a to 12g executes the procedures in the energy consumption prediction method and the energy consumption prediction program.

[0028] The link information acquisition unit 12a acquires link information indicating links included in the map information described above and determines whether or not vehicle 2 has traveled across a link. The link information acquisition unit 12a acquires the link ID of the vehicle 2's current position and compares the current link ID with the previous link ID to determine whether or not vehicle 2 has traveled across a link. If the link information acquisition unit 12a determines that the current link ID and the previous link ID match, it determines that vehicle 2's travel position does not cross a link. If the link information acquisition unit 12a determines that the current link ID and the previous link ID do not match, it determines that vehicle 2's travel position crossed a link and determines that vehicle 2 has traveled across a link.

[0029] When the link information acquisition unit 12a determines that vehicle 2 has traveled across a link, the statistics calculation unit 12b calculates statistics for the previous link, i.e., the link from which vehicle 2 exited. The statistics are data from when vehicle 2 was traveling, and are used by the server 4 to calculate the feature quantities necessary for predicting vehicle 2's energy consumption. In other words, the server 4 learns the statistics calculated on a link-by-link basis in the vehicle system 3 and calculates the feature quantities necessary for predicting vehicle 2's energy consumption.

[0030] As shown in Figure 5, statistical quantities are defined for each parameter. For example, for battery charge, the value at the start of the link and the value at the end of the link are defined. When calculating battery charge, the statistical quantity calculation unit 12b obtains power data contributing to battery charge from the second domain controller 10 (described later) as input data, and calculates the battery charge value at the start of the link and the battery charge value at the end of the link by performing calculations on the obtained power data. Similarly, for example, for vehicle speed, an average value within the link excluding the stopping period is defined. When calculating vehicle speed, the statistical quantity calculation unit 12b obtains power data contributing to vehicle speed from the second domain controller 10 as input data, and calculates the average value of vehicle speed within the link excluding the stopping period by performing calculations on the obtained power data. Similarly, the statistical quantity calculation unit 12b calculates the defined values ​​for acceleration, road gradient, vehicle weight, number of occupants, air conditioning settings (set temperature, airflow), outside temperature, inside temperature, wind speed, wind direction, and weather as statistical quantities. Furthermore, the statistical calculation unit 12 may acquire the power data used to calculate the statistical quantities from a controller, ECU, sensor, etc., other than the second domain controller 10.

[0031] In this case, the statistics calculation unit 12b selects the statistics to be calculated according to instructions from the server 4 and calculates the selected statistics. That is, the server 4 can collect desired statistics from the vehicle system 3 by instructing the vehicle system 3 to calculate the statistics to be calculated. The server 4 can arbitrarily update the statistics collected from the vehicle system 3, for example, by adding statistics that were not anticipated in the early stages, such as during the development of the vehicle 2.

[0032] Furthermore, the statistical calculation unit 12b sets the format of the statistical quantity according to instructions from the server 4 and calculates the statistical quantity according to the set format. That is, the server 4 can collect statistical quantities from the vehicle system 3 in a desired format by instructing the vehicle system 3 on the format of the statistical quantity. The server 4 can arbitrarily update the format of the statistical quantities collected from the vehicle system 3, for example, by adding a new format of statistical quantities that was not anticipated in the early stages, such as during the development of the vehicle 2. The format is, for example, the average value within a link, and the server 4 can set, for example, the median value within a link in addition to the average value within a link.

[0033] When the statistical quantity is calculated by the statistical quantity calculation unit 12b, the statistical quantity transmission unit 12c causes the calculated statistical quantity to be transmitted from the communication terminal 8 to the server 4.

[0034] The feature acquisition unit 12d acquires feature quantities from the server 4 when the feature quantities transmitted from the server 4 are received by the communication terminal 8. The weather information acquisition unit 12e acquires weather information from the weather information distribution server 6 when the weather information distributed from the weather information distribution server 6 is received by the communication terminal 8. The traffic congestion information acquisition unit 12f acquires traffic congestion information from the traffic congestion information distribution server 7 when the traffic congestion information transmitted from the traffic congestion information distribution server 7 is received by the communication terminal 8.

[0035] The energy consumption prediction unit 12g predicts energy consumption based on vehicle information, navigation driving plan information, and the acquired feature quantities, weather information, and traffic congestion information once feature quantities have been acquired by the feature quantity acquisition unit 12d, weather information has been acquired by the weather information acquisition unit 12e, and traffic congestion information has been acquired by the traffic congestion information acquisition unit 12f. The above is an example of predicting energy consumption based on feature quantities, weather information, and traffic congestion information, but at least one of the weather information and traffic congestion information may be omitted. That is, the energy consumption prediction unit 12g may predict energy consumption based only on feature quantities. Alternatively, the energy consumption prediction unit 12g may predict energy consumption based on feature quantities and weather information, or based on feature quantities and traffic congestion information.

[0036] The energy consumption prediction unit 12g calculates and predicts energy consumption using the calculation formula shown in Figure 6. The energy consumption prediction unit 12g calculates and predicts the energy consumed for driving between two points by adding the kinetic energy consumed for motion during driving between the two points and the air conditioning energy consumed for air conditioning during driving between the two points. The first control unit 12 displays the predicted energy consumption result predicted by the energy consumption prediction unit 12g, for example, on the navigation screen. The first control unit 12 also transmits the predicted energy consumption result predicted by the energy consumption prediction unit 12g to the second domain controller 10, and reflects the predicted energy consumption result in the driving control of the vehicle 2. For example, if the first control unit 12 determines that the battery level has fallen to a threshold while the vehicle 2 is driving, it performs control such as switching the driving mode from normal mode to a suppression mode that reduces energy consumption.

[0037] The second domain controller 10 includes a second control unit 20. The second control unit 20 is mainly composed of a microcontroller having a CPU, ROM, RAM, and I / O, and controls the operation of the second domain controller 10 by performing software processing by executing computer programs stored in a non-transitional physical storage medium using the CPU, and by performing hardware processing control using dedicated electronic circuits.

[0038] The second domain controller 10 is connected to the battery 22 and the electric motor 23 via the power converter 21. The second domain controller 10 manages various power data used in the calculation of statistical quantities in the first domain controller 9 and transmits these power data to the first domain controller 9. That is, in the first domain controller 9, the statistical quantity calculation unit 12b receives the various power data transmitted from the second domain controller 10 and calculates data on the vehicle 2 in motion as statistical quantities by, for example, performing calculations on the acquired power data.

[0039] The battery 22 supplies power to drive the electric motor 23. The vehicle 2 moves when the electric motor 23 is driven using the power supplied from the battery 22. When the vehicle 2 decelerates, the electric motor 23 acts as a generator, generating power through regenerative power generation. The power generated by regenerative power generation is stored in the battery 22. The power converter 21 converts the power exchanged between the battery 22 and the electric motor 23 into a mutually usable format. For example, it converts the DC power supplied from the battery 22 into AC power and outputs it to the electric motor 23, and also converts the AC power generated by regenerative power generation in the electric motor 23 into DC power and outputs it to the battery 22.

[0040] The second control unit 20 monitors the driving state of the vehicle 2, the state of the battery 22, the state of the electric motor 23, etc., and controls the operation of the power converter 21 based on the monitoring results, causing power to be exchanged between the battery 22 and the electric motor 23 according to the driving state of the vehicle 2. In other words, the second control unit 20 consumes the electrical energy stored in the battery 22 to generate kinetic energy for the vehicle 2 to move using the electric motor 23, or recovers at least a portion of the vehicle 2's kinetic energy using the electric motor 23 and stores the electrical energy in the battery 22 as reusable regenerative energy.

[0041] Next, the configuration of server 4 will be described. As shown in Figure 7, server 4 comprises a control unit 24 and a communication unit 25. The communication unit 25 controls data communication via the communication network 5 between server 4, weather information distribution server 6, and traffic congestion information distribution server 7. The control unit 24 is mainly composed of a microcontroller having a CPU, ROM, RAM, and I / O, and controls the operation of server 4 by performing software processing by executing computer programs stored in a non-transitional physical storage medium on the CPU, and by performing hardware processing control using dedicated electronic circuits.

[0042] The control unit 24 includes a statistic acquisition unit 24a, a statistic storage unit 24b, a feature quantity calculation unit 24c, a feature quantity storage unit 24d, and a feature quantity distribution unit 24e. When the statistic acquisition unit 24a receives the statistic transmitted from the vehicle system 3 at the communication unit 25, it acquires the statistic from the vehicle system 3. When the statistic is acquired by the statistic acquisition unit 24a, the statistic storage unit 24b stores the acquired statistic in a predetermined storage area. The feature quantity calculation unit 24c reads out the statistic stored in the statistic storage unit 24b, and calculates the feature quantity by learning the read statistic by, for example, an algorithm of a neural network. When the feature quantity is calculated by the feature quantity calculation unit 24c, the feature quantity storage unit 24d stores the calculated feature quantity in a predetermined storage area. The feature quantity distribution unit 24e reads out the feature quantity stored in the feature quantity storage unit 24d, and distributes the read feature quantity from the communication unit 25 to the vehicle system 3.

[0043] Next, the operations of the above-described configuration will be described with reference to FIGS. 8 to 13. Here, the processes performed by the first control unit 12 of the first domain controller 9 and the processes performed by the control unit 24 of the server 4 will be described.

[0044] (1) As the processes performed by the first control unit 12 of the first domain controller 9, a link ID determination process, a statistic transmission determination process, and an energy consumption prediction process will be described. The first control unit 12 periodically performs the link ID determination process, the statistic transmission determination process, and the energy consumption prediction process at a predetermined period. The first control unit 12 may perform the link ID determination process, the statistic transmission determination process, and the energy consumption prediction process at the same predetermined period or at different predetermined periods.

[0045] (1-1) Link ID determination process (see Fig. 8) When the start condition of the link ID determination process is satisfied, the first control unit 12 starts the link ID determination process. When starting the link ID determination process, the first control unit 12 acquires the current position (A1), and based on the acquired current position, acquires the current link ID indicating the current link (A2, corresponding to the link information acquisition procedure). The first control unit 12 collates the acquired current link ID with the previous link ID indicating the link acquired in the previous link ID determination process, and determines whether the current link ID and the previous link ID match (A3).

[0046] When the first control unit 12 determines that the current link ID and the previous link ID match (A3: YES), it determines that the traveling position of the vehicle 2 does not cross a link. The first control unit 12 overwrites the current link ID with the previous link ID to update the previous link ID (A6), ends the link ID determination process, and waits for the start condition of the next link ID determination process to be satisfied.

[0047] On the other hand, when the first control unit 12 determines that the current link ID and the previous link ID do not match (A3: NO), it determines that the traveling position of the vehicle 2 has crossed a link, and determines that the vehicle 2 has traveled across a link. The first control unit 12 calculates the statistic of the previous link (A4, corresponding to the statistic calculation procedure), and stores the calculated statistic of the previous link in a predetermined storage area (A5). That is, the first control unit 12 calculates and stores the statistic for each link. The first control unit 12 overwrites the current link ID with the previous link ID to update the previous link ID (A6), and waits for the start condition of the next link ID determination process to be satisfied.

[0048] (1-2) Statistic transmission determination process (see Fig. 9) When the start condition of the statistic transmission determination process is satisfied, the first control unit 12 starts the statistic transmission determination process. When starting the statistic transmission determination process, the first control unit 12 checks the storage area of the statistic (A11), and determines whether an untransmitted statistic is stored (A12). When the first control unit 12 determines that no untransmitted statistic is stored (A12: NO), it ends the statistic transmission determination process, and waits for the start condition of the next statistic transmission determination process to be satisfied.

[0049] On the other hand, if the first control unit 12 determines that unsent statistics are stored (A12: YES), it determines whether data communication is possible with the server 4 (A13). If the first control unit 12 determines that data communication is not possible with the server 4, i.e., that it is offline (A13: NO), it terminates the statistics transmission determination process and waits for the conditions to be met for the start of the next statistics transmission determination process. If the first control unit 12 determines that data communication is possible with the server 4, i.e., that it is online (A13: YES), it has the unsent statistics sent from the communication terminal 8 to the server 4 (A14, corresponding to the statistics transmission procedure), terminates the statistics transmission determination process, and waits for the conditions to be met for the start of the next statistics transmission determination process.

[0050] Furthermore, the above example illustrates the case where the statistics transmission determination process is terminated when it is determined that the system is offline. However, when it is determined that the system is offline, it may also be possible to perform the data communication line connection process with the server 4, and after transitioning from the offline state to the online state, have the unsent statistics transmitted from the communication terminal 8 to the server 4. Also, the example illustrates the case where the statistics transmission determination process is performed periodically at predetermined intervals. However, the statistics transmission determination process may also be started based on the condition that the statistics for the previous link have been calculated in the link ID determination process described above. In addition, for electric vehicles, the statistics transmission determination process may be started based on the condition that the motor is turned on or off. For example, for gasoline-powered vehicles, the statistics transmission determination process may be started based on the condition that the ignition is turned on or off.

[0051] (1-3) Energy Consumption Prediction Processing (See Figure 10) When the conditions for starting the energy consumption prediction processing are met, the first control unit 12 starts the energy consumption prediction processing. When the first control unit 12 starts the energy consumption prediction processing, it acquires vehicle information (S21). The first control unit 12 acquires navigation driving plan information (S22). The first control unit 12 acquires feature quantities from the server 4 when the feature quantities distributed from the server 4 are received by the communication terminal 8 (S23, corresponding to the feature quantity acquisition procedure). The first control unit 12 acquires weather information from the weather information distribution server 6 when the weather information distributed from the weather information distribution server 6 is received by the communication terminal 8 (S24). The first control unit 12 acquires traffic congestion information from the traffic congestion information distribution server 7 when the traffic congestion information distributed from the traffic congestion information distribution server 7 is received by the communication terminal 8 (S25). The first control unit 12 predicts energy consumption based on vehicle information, navigation driving plan information, feature quantities, weather information, and traffic congestion information (corresponding to S26, energy consumption prediction procedure), terminates the energy consumption prediction process, and waits for the conditions for starting the next energy consumption prediction process to be met. The first control unit 12 transmits the energy consumption predicted in this way to the second domain controller 10, and reflects it in the driving control of the vehicle 2.

[0052] (2) The processes performed by the control unit 24 of the server 4 will be described as statistical data storage processing, feature calculation processing, and feature distribution determination processing. The control unit 24 performs statistical data storage processing, feature calculation processing, and feature distribution determination processing periodically at predetermined intervals. The control unit 24 may perform statistical data storage processing, feature calculation processing, and feature distribution determination processing at the same predetermined interval or at different predetermined intervals.

[0053] (2-1) Statistical data storage process (see Figure 11) When the start condition for statistical data storage is met, the control unit 24 starts the statistical data storage process. The control unit 24 determines whether or not statistical data has been acquired from the vehicle system 3 (B1). If the control unit 24 determines that statistical data has not been acquired (B1: NO), it terminates the statistical data storage process and waits for the start condition for the next statistical data storage process to be met. On the other hand, if the control unit 24 determines that statistical data has been acquired because the statistical data transmitted from the vehicle system 3 has been received by the receiving unit 25 (B1: YES), it stores the acquired statistical data in a predetermined storage area (B2), terminates the statistical data storage process, and waits for the start condition for the next statistical data storage process to be met.

[0054] (2-2) Feature calculation process (see Figure 12) When the start condition for the feature calculation process is met, the control unit 24 starts the feature calculation process. When the feature calculation process starts, the control unit 24 reads out the stored statistics (B11), learns the read statistics, and calculates the features (B12). The control unit 24 saves the calculated features (B13), ends the feature calculation process, and waits for the start condition for the next feature calculation process to be met.

[0055] (2-3) Feature distribution determination process (see Figure 13) When the start condition for the feature distribution determination process is met, the control unit 24 starts the feature distribution determination process. When the control unit 24 starts the feature distribution determination process, it checks the feature storage area (B21) and determines whether or not undistributed features are stored (B22). When the control unit 24 determines that no undistributed features are stored (B22: NO), it terminates the feature distribution determination process and waits for the start condition for the next feature distribution determination process to be met.

[0056] On the other hand, if the control unit 24 determines that undistributed feature quantities are stored (B22: YES), it determines whether data communication is possible with the in-vehicle system 3 (B23). If the control unit 24 determines that data communication is not possible with the in-vehicle system 3, i.e., that it is offline (B23: NO), it terminates the feature quantity distribution determination process and waits for the conditions for starting the next feature quantity distribution determination process to be met. If the control unit 24 determines that data communication is possible with the in-vehicle system 3, i.e., that it is online (B23: YES), it has the communication unit 25 transmit the undistributed feature quantities to the in-vehicle system 3 (B24), terminates the feature quantity distribution determination process, and waits for the conditions for starting the next feature quantity distribution determination process to be met.

[0057] The above is an example of terminating the feature distribution determination process when it is determined that the system is offline. However, when it is determined that the system is offline, it may also be necessary to perform the data communication line connection process with the in-vehicle system 3, and after transitioning from the offline state to the online state, transmit the undistributed features from the communication unit 25 to the in-vehicle system 3.

[0058] The above is an example of a case where the vehicle system 3 performs the energy consumption forecast, but the server 4 may also perform the energy consumption forecast and distribute the forecast results to the vehicle system 3.

[0059] As described above, the embodiment provides the following advantages and benefits. In the vehicle system 3, statistical quantities for calculating the feature quantities necessary for predicting the energy consumption of vehicle 2 are calculated on a link-by-link basis, and the server 4 learns these statistical quantities and predicts the energy consumption of vehicle 2 based on the calculated feature quantities. By calculating the statistical quantities on a link-by-link basis, the amount of driving data required for learning can be optimized, and the energy consumption of vehicle 2 can be appropriately predicted.

[0060] The system now receives feature data distributed from server 4, allowing it to retrieve features from server 4. This increases the flexibility in retrieving features from server 4, enabling accurate prediction of vehicle 2's energy consumption without having to visit a dealership or other location.

[0061] The system selects the statistical quantities to be calculated according to instructions from Server 4 and then calculates those selected statistical quantities. Server 4 can arbitrarily update the statistical quantities collected from the vehicle system 3, for example, by adding new statistical quantities that were not anticipated in the early stages, such as during the development of Vehicle 2.

[0062] Furthermore, the statistical calculation unit 12b sets the format of the statistical quantity according to instructions from the server 4 and calculates the statistical quantity according to the set format. The server 4 can arbitrarily update the format of the statistical quantity collected from the vehicle system 3, for example, by adding a new statistical quantity format that was not anticipated in the early stages, such as during the development of the vehicle 2.

[0063] When calculating statistics related to vehicle speed, the system now calculates a statistic that shows the average vehicle speed within the link, excluding the period when vehicle 2 is stationary. By calculating the average vehicle speed within the link, excluding the period when vehicle 2 is stationary, the accuracy of energy consumption prediction can be improved.

[0064] This disclosure includes, in addition to the claims, the following disclosures: [1] An in-vehicle device (9) capable of data communication with an external device (4) that learns statistical quantities including data while the vehicle is running and calculates feature quantities necessary for predicting the energy consumption of the vehicle, comprising: a link information acquisition unit (12a) that acquires link information indicating links included in map information; a statistical quantity calculation unit (12b) that calculates the statistical quantities in units of links indicated by the link information; a statistical quantity transmission unit (12c) that transmits the statistical quantities to the external device; a feature quantity acquisition unit (12d) that acquires the feature quantities from the external device; and an energy consumption prediction unit (12g) that predicts the energy consumption of the vehicle based on the feature quantities.

[0065] [2] The in-vehicle device described in [1], wherein the feature acquisition unit receives the feature distributed from the external device and acquires the feature from the external device.

[0066] [3] The in-vehicle device described in [1] or [2], wherein the statistical calculation unit selects the statistical quantity to be calculated according to instructions from the external device and calculates the selected statistical quantity.

[0067] [4] An in-vehicle device as described in any one of [1] to [3], wherein the statistical calculation unit sets the format of the statistical quantity according to instructions from the external device and calculates the statistical quantity according to the set format of the statistical quantity.

[0068] [5] The in-vehicle device described in any one of paragraphs [1] to [4], wherein the statistical calculation unit calculates a statistical amount that shows the average vehicle speed within the link, excluding the stopping period when the vehicle is stopped, when calculating a statistical amount related to vehicle speed.

[0069] [6] An in-vehicle device as described in any one of [1] to [5], which is capable of communicating with a driving control device (10) that controls the driving of the vehicle using electric power, and the statistical calculation unit acquires at least a portion of the power data used for calculating the statistical quantity from the driving control device, and calculates the acquired statistical quantity in units of links indicated by the link information.

[0070] [7] An energy consumption prediction system (1) comprising an in-vehicle device (9) and an external device (4) that learns statistical quantities including data while the vehicle is running and calculates feature quantities necessary for predicting the energy consumption of the vehicle, wherein the in-vehicle device and the external device are configured to communicate with each other, the system comprising: a link information acquisition unit (12a) that acquires link information indicating links included in map information; a statistical quantity calculation unit (12b) that calculates the statistical quantities in units of links indicated by the link information; and an energy consumption prediction unit (12g) that predicts the energy consumption of the vehicle based on the feature quantities.

[0071] [8] The external device acquires environmental information which includes at least one of weather information and traffic congestion information for each link targeted by the statistical quantity, and calculates the feature quantity by associating the statistical quantity with the environmental information, as described in [7].

[0072] [9] The energy consumption prediction system described in [7] or [8], wherein the statistical calculation unit selects the statistical quantity to be calculated and calculates the selected statistical quantity.

[0073]

[10] The energy consumption prediction system described in any one of the items [7] to [9], wherein the statistical calculation unit sets the format of the statistical quantity and calculates the statistical quantity according to the set format of the statistical quantity.

[0074]

[11] The energy consumption prediction system described in any one of [7] to

[10] , wherein the statistical calculation unit calculates a statistical amount that indicates the average vehicle speed within the link, excluding the stopping period when the vehicle is stopped, when calculating a statistical amount related to vehicle speed.

[0075]

[12] The on-board device is capable of communicating with a driving control device (10) that controls the driving of the vehicle using electric power, and the statistical calculation unit obtains at least a portion of the power data used to calculate the statistical quantity from the driving control device, and calculates the obtained statistical quantity in units of links indicated by the link information, as described in any one of [7] to

[11] .

[0076]

[13] A method for predicting energy consumption in a system (1) comprising an in-vehicle device (9) and an external device (4) that learns statistical quantities including data while the vehicle is running and calculates feature quantities necessary for predicting the energy consumption of the vehicle, wherein the in-vehicle device and the external device are configured to communicate with each other, the method comprising: a link information acquisition procedure for acquiring link information indicating links included in map information; a statistical quantity calculation procedure for calculating the statistical quantities in units of links indicated by the link information; and an energy consumption prediction procedure for predicting the energy consumption of the vehicle based on the feature quantities.

[0077]

[14] The procedure for calculating the statistic is the method for predicting energy consumption described in

[13] , which involves selecting the statistic to be calculated and calculating the selected statistic.

[0078]

[15] The procedure for calculating the statistic is the method for predicting energy consumption described in

[13] or

[14] , which involves setting the format of the statistic and calculating the statistic according to the set format of the statistic.

[0079]

[16] The method for predicting energy consumption described in any one of the paragraphs

[13] to

[15] , wherein the procedure for calculating the statistic is to calculate a statistic that shows the average vehicle speed within the link, excluding the stopping period when the vehicle is stopped, when calculating a statistic related to vehicle speed.

[0080]

[17] An energy consumption prediction program that causes the control unit (12) of an in-vehicle device (9), which is configured to communicate data with an external device (4) that learns statistical quantities including data on the vehicle while it is running and calculates feature quantities necessary for predicting the energy consumption of the vehicle, to execute: a link information acquisition procedure for acquiring link information indicating links included in map information; a statistical quantity calculation procedure for calculating the statistical quantities in units of links indicated by the link information; a statistical quantity transmission procedure for transmitting the statistical quantities to the external device; a feature quantity acquisition procedure for acquiring the feature quantities from the external device; and an energy consumption prediction procedure for predicting the energy consumption of the vehicle based on the feature quantities.

[0081]

[18] The feature acquisition procedure is a prediction program for energy consumption described in

[17] , which acquires the feature from the external device when the feature distributed from the external device is received.

[0082]

[19] The statistical calculation procedure is a predictive energy consumption program described in

[17] or

[18] that selects the statistical quantity to be calculated according to instructions from the external device and calculates the selected statistical quantity to be calculated.

[0083]

[20] The statistic calculation procedure is a prediction program for energy consumption as described in any one of

[17] to

[19] , which sets the format of the statistic according to instructions from the external device and calculates the statistic according to the set format of the statistic.

[0084]

[21] The statistical calculation procedure is an energy consumption prediction program described in any one of

[17] to

[20] , which calculates a statistical amount that shows the average vehicle speed within a link, excluding the stopping period when the vehicle is stopped, when calculating a statistical amount related to vehicle speed.

[0085] This disclosure is described in accordance with the embodiments, but it is understood that this disclosure is not limited to such embodiments or structures. This disclosure also includes various modifications and variations within the equivalence. In addition, various combinations and forms, as well as other combinations and forms that include only one, more, or fewer of those elements, fall within the scope and concept of this disclosure.

[0086] The control unit and its method described herein may be implemented by a dedicated computer provided by configuring a processor and memory programmed to perform one or more functions embodied by a computer program. Alternatively, the control unit and its method described herein may be implemented by a dedicated computer provided by configuring a processor by one or more dedicated hardware logic circuits. Alternatively, the control unit and its method described herein may be implemented by one or more dedicated computers configured by a combination of a processor and memory programmed to perform one or more functions and a processor configured by one or more hardware logic circuits. Furthermore, the computer program may be stored as instructions executed by the computer on a computer-readable non-transitional tangible recording medium.

Claims

1. An in-vehicle device (9) capable of data communication with an external device (4) that learns statistical quantities including data while the vehicle is running and calculates feature quantities necessary for predicting the energy consumption of the vehicle, comprising: a link information acquisition unit (12a) that acquires link information indicating links included in map information; a statistical quantity calculation unit (12b) that calculates the statistical quantities in units of links indicated by the link information; a statistical quantity transmission unit (12c) that transmits the statistical quantities to the external device; a feature quantity acquisition unit (12d) that acquires the feature quantities from the external device; and an energy consumption prediction unit (12g) that predicts the energy consumption of the vehicle based on the feature quantities.

2. The in-vehicle device according to claim 1, wherein the feature acquisition unit acquires the feature from the external device by receiving the feature distributed from the external device.

3. The in-vehicle device according to claim 1, wherein the statistical calculation unit selects the statistical quantity to be calculated according to instructions from the external device and calculates the selected statistical quantity.

4. The in-vehicle device according to claim 1, wherein the statistical calculation unit sets the format of the statistical quantity according to instructions from the external device and calculates the statistical quantity according to the set format of the statistical quantity.

5. The in-vehicle device according to claim 1, wherein, when calculating a statistical quantity relating to vehicle speed, the statistical quantity calculation unit calculates a statistical quantity indicating the average vehicle speed within the link, excluding the stopping period when the vehicle is stopped.

6. An in-vehicle device according to any one of claims 1 to 5, wherein the device is capable of communicating with a driving control device (10) that controls the driving of a vehicle using electric power, and the statistical calculation unit acquires at least a portion of the power data used for calculating the statistical quantity from the driving control device, and calculates the acquired statistical quantity in units of links indicated by the link information.

7. An energy consumption prediction system (1) comprising an in-vehicle device (9) and an external device (4) that learns statistical quantities including data while the vehicle is running and calculates feature quantities necessary for predicting the energy consumption of the vehicle, wherein the in-vehicle device and the external device are configured to communicate with each other, the energy consumption prediction system comprising: a link information acquisition unit (12a) that acquires link information indicating links included in map information; a statistical quantity calculation unit (12b) that calculates the statistical quantities in units of links indicated by the link information; and an energy consumption prediction unit (12g) that predicts the energy consumption of the vehicle based on the feature quantities.

8. The energy consumption prediction system according to claim 7, wherein the external device acquires environmental information including at least one of weather information and traffic congestion information for each link targeted by the statistical quantity, and calculates the feature quantity by associating the statistical quantity with the environmental information.

9. The energy consumption prediction system according to claim 7, wherein the statistical calculation unit selects the statistical quantity to be calculated and calculates the selected statistical quantity.

10. The energy consumption prediction system according to claim 7, wherein the statistical calculation unit sets the format of the statistical quantity and calculates the statistical quantity according to the set format of the statistical quantity.

11. The energy consumption prediction system according to claim 7, wherein the statistical calculation unit calculates a statistical quantity that indicates the average vehicle speed within the link, excluding the stopping period when the vehicle is stopped, when calculating a statistical quantity related to vehicle speed.

12. The on-board device is capable of communicating with a driving control device (10) that controls the vehicle's operation using electric power, and the statistical calculation unit obtains at least a portion of the power data used to calculate the statistical quantity from the driving control device, and calculates the obtained statistical quantity in units of links indicated by the link information, as described in any one of claims 7 to 11.

13. A method for predicting energy consumption in a system (1) comprising an in-vehicle device (9) and an external device (4) that learns statistical quantities including data while the vehicle is running and calculates feature quantities necessary for predicting the energy consumption of the vehicle, wherein the in-vehicle device and the external device are configured to communicate with each other, the method comprising: a link information acquisition procedure for acquiring link information indicating links included in map information; a statistical quantity calculation procedure for calculating the statistical quantities in units of links indicated by the link information; and an energy consumption prediction procedure for predicting the energy consumption of the vehicle based on the feature quantities.

14. The method for predicting energy consumption according to claim 13, wherein the procedure for calculating the statistic comprises selecting the statistic to be calculated and calculating the selected statistic.

15. The method for predicting energy consumption according to claim 13, wherein the procedure for calculating the statistic involves setting the format of the statistic and calculating the statistic according to the set format of the statistic.

16. The method for predicting energy consumption according to claim 13, wherein the statistical calculation procedure calculates a statistical quantity that indicates the average vehicle speed within the link, excluding the stopping period when the vehicle is stopped, when calculating a statistical quantity related to vehicle speed.

17. An energy consumption prediction program that causes the control unit (12) of an in-vehicle device (9), which is configured to communicate data with an external device (4) that learns statistical quantities including data on the vehicle while it is running and calculates feature quantities necessary for predicting the energy consumption of the vehicle, to execute: a link information acquisition procedure for acquiring link information indicating links included in map information; a statistical quantity calculation procedure for calculating the statistical quantities in units of links indicated by the link information; a statistical quantity transmission procedure for transmitting the statistical quantities to the external device; a feature quantity acquisition procedure for acquiring the feature quantities from the external device; and an energy consumption prediction procedure for predicting the energy consumption of the vehicle based on the feature quantities.

18. The feature acquisition procedure is a program for predicting energy consumption according to claim 17, wherein the feature is acquired from the external device by receiving the feature distributed from the external device.

19. The energy consumption prediction program according to claim 17, wherein the statistical calculation procedure involves selecting the statistical quantity to be calculated according to instructions from the external device, and calculating the selected statistical quantity to be calculated.

20. The energy consumption prediction program according to claim 17, wherein the statistical calculation procedure sets the format of the statistical quantity according to instructions from the external device and calculates the statistical quantity according to the set format of the statistical quantity.

21. The energy consumption prediction program according to claim 17, wherein the statistical calculation procedure calculates a statistical quantity that indicates the average vehicle speed within the link, excluding the stopping period when the vehicle is stopped, when calculating a statistical quantity related to vehicle speed.

Images