Information processing device, information processing method, program and mobile object

Abstract

To provide an information processing device that accurately predicts the amount of power consumption consumed by a moving object.SOLUTION: The information processing device includes an acquisition unit, a prediction value calculation unit, and an output unit. The acquisition unit acquires one or more factor values representing one or more factors that affect the amount of power consumption with respect to movement of a prediction object of a moving object. On the basis of a prediction model with one or more factor values as input variables and the amount of power consumption as an object variable, one or more factor values for the movement of the prediction object of the moving object, and an error between a predicted value by the prediction model and a measured value, the prediction value calculation unit calculates a prediction value of the amount of power consumption in the movement of the prediction object. The output unit outputs the prediction value.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] An embodiment of the present invention relates to an information processing device, an information processing method, a program, and a mobile object. [Background technology]

[0002] For example, in an electric vehicle, in order to prevent the battery from running out while the vehicle is running, it is necessary to calculate a predicted value of the amount of power consumed by the vehicle while running in advance, and charge the vehicle with an amount of power equal to or greater than the predicted value before the vehicle runs. The predicted value of the amount of power consumed by an electric vehicle is calculated by observing the distance traveled, the speed, the temperature, etc., and using these observed values ​​as input variables. However, there are factors that are not included in the input variables, and factors that are difficult to quantify, such as the driver's driving technique, and it has been difficult to reflect these factors in the mathematical model. Therefore, in a situation where factors that are difficult to reflect in the mathematical model have a large impact on the amount of power consumed, it has been difficult to accurately calculate the predicted value with the conventional mathematical model. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 6232258 Summary of the Invention [Problem to be solved by the invention]

[0004] An object of the present invention is to provide an information processing device, an information processing method, a program, and a mobile body that accurately predict the amount of power consumption consumed by the mobile body. [Means for solving the problem]

[0005] An information processing device according to an embodiment includes a predicted value calculation unit that calculates a predicted value of the power consumption of a moving object based on a prediction model that uses a power consumption of a moving object as a response variable, one or more factor values ​​that affect the power consumption of the moving object for the moving object, and an error between a predicted value by the prediction model and an actual measured value. The predicted value calculation unit calculates the predicted value for the movement of the prediction target by using the error for a movement immediately preceding the movement of the prediction target among a plurality of past movements of the moving object and the prediction model. [Brief description of the drawings]

[0006] [Figure 1] FIG. 1 is a diagram showing the configuration of a prediction device according to a first embodiment, together with a moving object. [Diagram 2] FIG. 2 is a diagram showing an example of a moving table according to the first embodiment. [Diagram 3] FIG. 13 is a diagram showing an example of a unit of movement for which movement data is created; [Figure 4] FIG. 13 is a diagram showing an example of a prediction model. [Diagram 5] FIG. 13 is a diagram showing an example of an error table. [Figure 6] 11 is a flowchart showing the flow of movement data creation processing. [Figure 7] 11 is a flowchart showing the flow of a process for creating evaluation data. [Figure 8] FIG. 11 is a diagram showing an example of evaluation data. [Figure 9] 11 is a flowchart showing the flow of a result data creation process. [Figure 10] FIG. 11 is a diagram showing an example of result data. [Figure 11] 11 is a flowchart showing the flow of a calculation process of a predicted value. [Figure 12] FIG. 13 is a diagram showing the configuration of a prediction device according to a second embodiment, together with a moving object. [Figure 13] 10 is a flowchart showing a processing flow of a prediction device according to a second embodiment. [Figure 14] FIG. 13 is a diagram showing the configuration of a prediction device according to a third embodiment, together with a moving object. [Figure 15] FIG. 13 is a diagram showing an example of a moving table according to the third embodiment. [Figure 16]10 is a flowchart showing the flow of a process for updating an error in a driver table. [Figure 17] FIG. 4 is a diagram showing an example of a pilot table. [Figure 18] FIG. 13 is a diagram showing the configuration of a prediction device according to a fourth embodiment, together with a moving object. [Figure 19] FIG. 13 is a diagram showing the configuration of a prediction device according to a fifth embodiment, together with a moving object. [Figure 20] FIG. 13 is a diagram showing the configuration of a prediction device according to a sixth embodiment, together with a moving object. [Figure 21] FIG. 23 is a diagram showing the configuration of a prediction device according to a seventh embodiment, together with a moving object. [Figure 22] FIG. 13 is a diagram showing an example of an approximation curve. [Figure 23] 11 is a flowchart showing the flow of a process for updating a category value. [Figure 24] FIG. 11 is a diagram showing an example of a setting file. [Diagram 25] FIG. 11 is a diagram showing an example of a threshold value. [Figure 26] FIG. 11 is a diagram showing a first example of a calculation process of a category value. [Figure 27] FIG. 11 is a diagram showing a second example of the process of calculating a category value. [Figure 28] FIG. 1 is a diagram showing the correspondence between category values ​​and model input values. [Figure 29] FIG. 13 is a diagram showing an example of an output image. [Diagram 30] FIG. 2 is a diagram illustrating an example of a hardware configuration of a prediction device. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0007] A prediction device 20 according to a number of embodiments will now be described with reference to the drawings.

[0008] (First embodiment) FIG. 1 is a diagram showing the configuration of a prediction device 20 according to the first embodiment, together with a moving object 22.

[0009] The prediction device 20 predicts the amount of power consumption consumed by a mobile object 22. The mobile object 22 includes a secondary battery and moves using power stored in the secondary battery. In this embodiment, the mobile object 22 is a route bus that moves along the same route repeatedly. Instead of a route bus, the mobile object 22 may be, for example, an electric vehicle, a train that moves on a railroad track, or another type of vehicle. The mobile object 22 may also be a ship, an airplane, a drone, or the like. The mobile object 22 may also be a vehicle that does not move along the same route repeatedly.

[0010] The prediction device 20 is realized by an information processing device such as a computer. The prediction device 20 may be an information processing device that operates independently, or may be a device in which one or more information processing devices are linked together and realized by a server or a cloud on a network or the like. The prediction device 20 may be mounted on a mobile object 22, or may be mounted on a charging system that charges the mobile object 22 with power.

[0011] The prediction device 20 calculates a predicted value of the amount of power consumed during movement of such a moving object 22. Then, the prediction device 20 outputs the calculated predicted value to, for example, the operator of the moving object 22.

[0012] The prediction device 20 includes a collection unit 32, a movement data creation unit 34, a movement history memory unit 36, an evaluation data creation unit 38, a model memory unit 40, an error calculation unit 42, an error memory unit 44, an acquisition unit 46, a prediction value calculation unit 48, and an output unit 50.

[0013] The collection unit 32 acquires, when the moving object 22 moves one unit, one or more factor values ​​that represent one or more factors that affect the amount of power consumption for the movement of the moving object 22. The one or more factor values ​​will be described in further detail below with reference to FIG.

[0014] One unit of movement of the mobile object 22 is, for example, a unit of movement for which an actual measured value of the amount of power consumed by the mobile object 22 can be obtained. Further details of one unit of movement of the mobile object 22 will be described later with reference to FIG.

[0015] The collection unit 32 may acquire a part or all of the one or more factor values ​​from an external system realized by a server or the like via a network. The collection unit 32 may also acquire a part or all of the one or more factor values ​​from a measuring device, a terminal device, or a storage device provided in the mobile object 22. The collection unit 32 sends the collected one or more factor values ​​to the movement data creation unit 34.

[0016] The movement data creation unit 34 generates movement identification information for identifying one unit of movement every time the moving object 22 moves one unit. The movement data creation unit 34 creates one piece of movement data including the generated movement identification information and one or more factor values ​​acquired from the collection unit 32. The movement data creation unit 34 stores the created movement data in a movement table stored in the movement history storage unit 36.

[0017] The movement history storage unit 36 ​​stores a movement table capable of storing multiple pieces of movement data. The movement table stores one piece of movement data in one record. Every time the movement history storage unit 36 ​​acquires movement data from the movement data creation unit 34, it creates a new record in the movement table and stores the movement data in the newly created record.

[0018] When the moving body 22 has completed one unit of movement, the evaluation data creation unit 38 acquires an actual measurement value of the amount of power actually consumed by the moving body 22 in the corresponding one unit of movement. The evaluation data creation unit 38 acquires, for example, the actual measurement value from the moving body 22. When the moving body 22 has completed one unit of movement, the evaluation data creation unit 38 creates evaluation data including the acquired actual measurement value and movement data for the corresponding one unit of movement. Then, the evaluation data creation unit 38 sends the created evaluation data to the error calculation unit 42.

[0019] The model storage unit 40 stores a prediction model in which one or more factor values ​​are a group of input variables and power consumption is an objective variable. In this embodiment, the model storage unit 40 stores a first prediction model and a second prediction model. The first prediction model is a model that predicts power consumption using one or more factor values ​​without using an error between a predicted value and an actual measurement value. The second prediction model is a model that predicts power consumption using an error between a predicted value and an actual measurement value and one or more factor values.

[0020] The first prediction model and the second prediction model are trained in advance. In this embodiment, the second prediction model is a model that obtains the same result as the first prediction model when the error is set to 0. For example, the second prediction model is a model that calculates a predicted value by adding an error to a value predicted using the first prediction model. The prediction models will be described in more detail below with reference to FIG. 4.

[0021] When the moving object 22 finishes one unit of movement, the error calculation unit 42 acquires evaluation data for the corresponding one unit of movement from the evaluation data creation unit 38. The error calculation unit 42 also acquires the first prediction model from the model storage unit 40. Furthermore, the error calculation unit 42 extracts one or more factor values ​​included in the movement data portion of the acquired evaluation data. Then, the error calculation unit 42 calculates a past prediction value of the power consumption for the one unit of movement that has finished, using the one or more factor values ​​extracted from the evaluation data and the first prediction model.

[0022] Furthermore, the error calculation unit 42 calculates the error between the calculated past predicted value of the power consumption and the actual measured value included in the evaluation data for the corresponding one unit of movement. The error calculation unit 42 creates one piece of result data including the calculated error and the evaluation data acquired from the evaluation data creation unit 38. The error calculation unit 42 stores the created result data in an error table stored in the error storage unit 44.

[0023] The error storage unit 44 stores an error table capable of storing multiple pieces of result data. The error table stores one piece of error data in one record. Every time the error storage unit 44 obtains result data from the error calculation unit 42, it creates a new record in the error table and stores the result data in the created new record.

[0024] When predicting the amount of power consumption for the movement of the prediction target of the mobile body 22, the acquisition unit 46 acquires one or more factor values ​​representing one or more factors that affect the amount of power consumption for the movement of the prediction target of the mobile body 22. The item or items of the one or more factor values ​​acquired by the acquisition unit 46 are the same as the item or items of the one or more factor values ​​collected by the collection unit 32. In the prediction device 20, the collection unit 32 and the acquisition unit 46 may be realized by a common module.

[0025] The acquisition unit 46 receives, for example, a prediction instruction input from a pilot or the like, and acquires one or more factor values ​​when the acquisition unit 46 receives the prediction instruction. Furthermore, when the moving body 22 starts moving, the acquisition unit 46 may acquire one or more factor values ​​for the movement as a prediction target prior to the start of the movement. In this case, the one or more factor values ​​collected by the collection unit 32 to create movement data are the same as the one or more factor values ​​acquired by the acquisition unit 46 for the movement of the prediction target. The acquisition unit 46 sends the one or more factor values ​​for the movement of the prediction target that have been acquired to the prediction value calculation unit 48.

[0026] The predicted value calculation unit 48 receives one or more factor values ​​from the acquisition unit 46, and acquires a prediction model from the model storage unit 40. The predicted value calculation unit 48 calculates a predicted value of the energy consumption in the movement of the prediction target based on the one or more factor values ​​acquired by the acquisition unit 46, using a prediction model in which the one or more factor values ​​are a group of input variables and the energy consumption is an objective variable.

[0027] Here, the predicted value calculation unit 48 calculates a predicted value for the movement of the prediction target using the error between the past predicted value of the power consumption using the prediction model for the past movement of the moving body 22 and the actual measured value of the power consumption actually consumed in the past movement, and the prediction model. In this embodiment, when an error for the past movement exists in the error table in the error storage unit 44, that is, when at least one result data is stored in the error table, the predicted value calculation unit 48 calculates a predicted value for the movement of the prediction target using the second prediction model stored in the model storage unit 40. Also, when an error for the past movement does not exist in the error table in the error storage unit 44, that is, when no result data is stored in the error table, the predicted value calculation unit 48 calculates a predicted value for the movement of the prediction target using the first prediction model stored in the model storage unit 40.

[0028] In addition, when there are multiple errors for multiple past movements in the error table in the error storage unit 44, that is, when multiple result data are stored in the error table, the prediction value calculation unit 48 selects one of the errors and calculates a prediction value for the movement of the prediction target using the selected error and the prediction model. In this embodiment, the prediction value calculation unit 48 calculates a prediction value for the movement of the prediction target using an error for a movement immediately preceding the movement of the prediction target among multiple past movements of the moving body 22 and the prediction model. Here, the most recent movement refers to a past movement that is closest in time to the movement of the prediction target. The prediction value calculation unit 48 can accurately calculate a prediction value by using an error for a movement that is close in time to the movement of the prediction target.

[0029] When the predicted value calculation unit 48 calculates the predicted value, the output unit 50 outputs the calculated predicted value. The output unit 50 may display the predicted value on a display device in the mobile object 22, for example, or may display the predicted value on a display device provided in a charging system that charges the mobile object 22 with power.

[0030] FIG. 2 is a diagram illustrating an example of a moving table according to the first embodiment.

[0031] The migration table can store a plurality of migration data. The migration data includes unique migration identification information. For example, the migration identification information is generated by incrementing a value by one in time series.

[0032] The one or more factor values ​​may include, by way of example, one or more of the start position, end position, travel distance, cumulative upward gradient, cumulative downward gradient, start time, end time, travel time, speed, weight, inside temperature, and outside temperature.

[0033] The start position is a position where the moving body 22 starts one unit of movement, and is represented by, for example, latitude and longitude, etc. The end position is a position where the moving body 22 ends one unit of movement, and is represented by, for example, latitude and longitude, etc. The movement distance represents the distance of the route along which the moving body 22 moves.

[0034] The cumulative upward gradient is the total upward gradient from the start position to the end position on the route. A map server or the like provides elevation values ​​on the route at equal intervals, for example, about 1 m. For example, the collection unit 32 collects a group of elevation values ​​on the route from the map server or the like, and calculates the cumulative upward gradient. For example, the collection unit 32 calculates the cumulative upward gradient by adding up all of the sets of values ​​that are positive numbers, which are the difference between the elevation value of the next point in the direction of travel of the route.

[0035] The cumulative downward gradient is the total downward gradient from the start position to the end position on the route. For example, the collection unit 32 collects a group of elevation values ​​on the route from a map server or the like, and calculates the cumulative downward gradient. For example, the collection unit 32 calculates the cumulative downward gradient by adding up all the absolute values ​​of the difference between the elevation value in the direction of travel of the route and the elevation value of the next point, which is a negative number.

[0036] The start time is the time when the moving body 22 starts one unit of movement and is expressed, for example, in a format such as "year / month / day hour:minute:second". The end time is the time when the moving body 22 ends one unit of movement and is expressed, for example, in a format such as "year / month / day hour:minute:second". The travel time is the time when the moving body 22 is actually traveling and is, for example, the time between the start time and the end time excluding stop times such as waiting at traffic lights or resting.

[0037] The speed is the speed of the moving object 22. The speed may be an average value of the speed not including stop times such as waiting at traffic lights or resting during the movement.

[0038] The weight is the weight of the moving object 22. For example, if the weight changes during movement, it may be the average value of the weight at each fixed time during the movement.

[0039] The inside air temperature is the air temperature inside the moving object 22. When the inside air temperature changes during movement, for example, it may be the average value of the inside air temperature at every fixed time during movement.

[0040] The outside air temperature is the air temperature around the moving object 22. When the outside air temperature changes during movement, for example, it may be the average value of the outside air temperature for each fixed time period during movement.

[0041] The one or more factor values ​​are not limited to such information, and may include other information such as the acceleration of the moving object 22, the motor temperature, and the temperature of the air conditioner.

[0042] FIG. 3 is a diagram showing an example of a unit of movement for which movement data is generated.

[0043] The unit of movement for which movement data is created is, for example, a continuous section in which the speed of the moving object 22 is not 0. The unit of movement for which movement data is created may also be, for example, a section that combines multiple sections excluding stop sections when the moving object 22 waits for a traffic light or stops for a rest, etc., during its movement, as shown in Fig. 3. In this way, when the unit of movement is a section that combines multiple sections excluding stop sections, the movement distance, the accumulated uphill gradient value, the accumulated downhill gradient value, and the movement time are expressed as follows:

[0044] For example, if a set of distances of multiple sections not including traffic light waits and rest stops is d1, d2, . . . , dM, the travel distance (x_dis) for one unit of travel is expressed by the following equation. x_dis=d1+d2+…+dM

[0045] For example, if the set of cumulative values ​​of uphill gradients for multiple sections not including traffic light waits and rest stops is u1, u2, . . . uM, the cumulative uphill gradient value (x_up) for one unit of movement is expressed by the following equation. x_up=u1+u2+…+uM

[0046] For example, if the set of cumulative values ​​of downward gradients of multiple sections that do not include traffic light waits or rest stops is f1, f2, . . . fM, the cumulative downward gradient value (x_fall) for one unit of movement is expressed as follows: x_fall=f1+f2+…+fM

[0047] For example, if a set of travel times for multiple sections not including traffic light waits and rest breaks is denoted as t1, t2, . . . tM, the travel time (x_time) for one unit of travel is expressed as follows: x_time=t1+T2+…+tM

[0048] 4 is a diagram showing an example of a prediction model. In this embodiment, the model storage unit 40 stores a first prediction model and a second prediction model. For example, the model storage unit 40 stores the first prediction model and the second prediction model in association with model identification information (m1, m2).

[0049] In this embodiment, the first prediction model is expressed as follows. m1:y_i=f(x_dis,x_up,x_fall,…)

[0050] f() is a function that represents a prediction model. f() may be any model as long as it is in the form of a general model, such as a multiple regression model, a neural network, deep learning, or a physical model.

[0051] In this embodiment, the second prediction model is expressed as follows. m2:y_i=f(x_dis,x_up,x_fall,…)+e

[0052] e represents an error. For the error, a value stored in an error table stored in the error storage unit 44 is used. That is, the second prediction model is expressed in a format in which an error is added to a value predicted by the first prediction model.

[0053] As shown in Fig. 4, the first prediction model and the second prediction model may be multiple regression models. In this case, f() is expressed as {β_dis, β_up, ..., β_intercept}. The regression coefficients {β_dis, β_up, ..., β_intercept} are estimated using a group of factors that affect the amount of power consumption and the actual measured value of the amount of power consumption, based on a plurality of result data stored in an error table stored in the error storage unit 44. The parameter tuning method for estimating the regression coefficients may be a general method such as the least squares method.

[0054] 5 is a diagram showing an example of an error table. The error table stores, for each of one or more past movements, actual measured values, predicted values, and errors of power consumption in addition to items similar to those in the movement table, in association with movement identification information.

[0055] 6 is a flowchart showing the flow of movement data creation processing by the prediction device 20. When the moving object 22 moves, the prediction device 20 executes movement data creation processing according to the flow shown in FIG.

[0056] First, in S101, the collection unit 32 collects one or more factor values ​​representing one or more factors that affect the amount of power consumption for the movement of the moving object 22. For example, the collection unit 32 acquires a start position, an end position, an end time, a movement time, a speed, a weight, and an inside temperature as part of the one or more factor values ​​from the moving object 22. Note that the collection unit 32 may acquire scheduled or predicted values ​​for information that is uncertain at the start of the movement, such as the end time, the movement time, the speed, and the weight.

[0057] Furthermore, the collecting unit 32 acquires the travel distance, the cumulative value of the upward gradient, and the cumulative value of the downward gradient from, for example, an external system such as a map server via a network based on the start position, the end position, and the route. Furthermore, the collecting unit 32 acquires the outside temperature from, for example, an external system such as a weather server based on the start position, the end position, and the route. For example, the collecting unit 32 acquires the outside temperature at the predicted passing position from the weather server at regular intervals from the start time to the end time. Also, for example, instead of acquiring the outside temperature from the weather server, the collecting unit 32 may acquire the current outside temperature from a thermometer provided on the moving object 22. Also, when the collecting unit 32 acquires the outside temperature from the weather server, the collecting unit 32 may correct the inside temperature using the acquired outside temperature.

[0058] Next, in S102, the movement data creation unit 34 generates movement identification information. The movement data creation unit 34 may generate the movement identification information, for example, by incrementing the movement identification information for the immediately previous movement. Next, in S103, the movement data creation unit 34 creates movement data including the generated movement identification information and the collected one or more factor values.

[0059] Next, in S104, when the movement data creation unit 34 calculates a predicted value for the current movement of the moving object 22 as the movement to be predicted, the movement data creation unit 34 sends the generated movement data to the predicted value calculation unit 48. In this case, the predicted value calculation unit 48 calculates a predicted value of the amount of power consumption for the current movement of the moving object based on the movement data acquired from the movement data creation unit 34.

[0060] Next, in S105, the movement data creation unit 34 determines whether or not the movement of the moving object 22 has ended. If the movement has not ended (No in S105), the movement data creation unit 34 waits for the process to end in S105. If the movement has ended (Yes in S105), the movement data creation unit 34 advances the process to S106.

[0061] In S106, the movement data creation unit 34 updates the factor values ​​of one or more factor values ​​included in the movement data for factors that can be replaced with actually measured values. The movement data creation unit 34 may update the movement time to the actual movement time excluding the time spent waiting for traffic lights and stopping for rest. The movement data creation unit 34 may also update the end time to the time when the movement actually ends. For example, the movement data creation unit 34 may update the speed, weight, indoor temperature, and outdoor temperature to actually measured values. In this case, when the speed, weight, indoor temperature, and outdoor temperature are sampled at regular intervals during the period when the moving object 22 is moving and measured values ​​are obtained, the movement data creation unit 34 may use the average value of the measured values. Note that the movement data creation unit 34 does not need to update any of the one or more factor values ​​included in the movement data to actually measured values.

[0062] Then, the movement data creation unit 34 stores the movement data in a movement table stored in the movement history storage unit 36. When the prediction device 20 finishes the process of S106, it ends this flow. By executing the above process, the prediction device 20 can generate movement data every time the moving object 22 moves one unit and store it in the movement table.

[0063] Fig. 7 is a flowchart showing the flow of the evaluation data creation process by the prediction device 20. Fig. 8 is a diagram showing an example of the evaluation data created by the process shown in Fig. 7. After the moving object 22 finishes moving and stores the movement data in the movement table, the prediction device 20 executes the evaluation data creation process according to the flow shown in Fig. 7.

[0064] First, in S111, the evaluation data creation unit 38 acquires from the moving object 22 an actual measurement value of the amount of power actually consumed for the movement corresponding to the movement identification information included in the movement data created by the movement data creation unit 34.

[0065] Next, in S112, the evaluation data creation unit 38 acquires the movement data stored in the movement table stored in the movement history memory unit 36 ​​for the movement corresponding to the movement identification information included in the movement data created by the movement data creation unit 34.

[0066] Next, in S113, the evaluation data creation unit 38 creates evaluation data including the movement identification information, the acquired movement data, and the acquired actual measurement value of the amount of power.

[0067] Next, in S114, the evaluation data creation unit 38 sends the created evaluation data to the error calculation unit 42. For example, the evaluation data creation unit 38 creates evaluation data as shown in FIG.

[0068] Fig. 9 is a flowchart showing the flow of a process for creating result data by the prediction device 20. Fig. 10 is a diagram showing an example of result data created by the process shown in Fig. 9. After the evaluation data is created, the prediction device 20 executes the process for creating result data according to the flow shown in Fig. 9.

[0069] First, in S121, when the moving object 22 finishes one unit of movement, the error calculation unit acquires, from the evaluation data creation unit , evaluation data for the corresponding one unit of movement.

[0070] Next, in S122, the error calculation unit 42 acquires the first prediction model from the model storage unit 40. That is, the error calculation unit 42 acquires a prediction model that predicts the power consumption amount by using one or more factor values, without using the error between the predicted value and the actual measured value.

[0071] Next, in S123, the error calculation unit 42 extracts one or more factor values ​​included in the movement data portion of the acquired evaluation data, and calculates a predicted value for past movement using the one or more factor values ​​extracted from the evaluation data and the first prediction model.

[0072] Next, in S124, the error calculation unit 42 calculates the error between the calculated predicted value and the actual measurement value included in the acquired evaluation data.

[0073] For example, the error calculation unit 42 calculates the error by the following calculation: In the following formula, e_i represents the error, y^_i represents the predicted value, and y_i represents the actual measured value. Difference: e_i=y^_i-y_i Absolute error: e_i=|y^_i-y_i| Squared error: e_i=(y^_i-y_i)^2 Relative error: e_i=(y^_i-y_i) / y_i Absolute error rate: / y_i|y^_i-y_i| / y_i Squared error rate: e_i=(y^_i-y_i)^2 / y_i

[0074] The error calculation section 42 may calculate the error using another general error calculation formula, not limited to the above formula.

[0075] Next, in S125, the error calculation section creates one piece of result data including the calculated error and the evaluation data acquired from the evaluation data creation section .

[0076] Next, in S126, the error calculation unit 42 stores the created result data in an error table stored in the error storage unit 44. For example, the error calculation unit 42 creates result data as shown in FIG. 10 and stores it in the error table. When the prediction device 20 ends the process of S126, it ends this flow. By executing the above process, the prediction device 20 can generate result data and store it in the error table every time the moving object 22 moves one unit.

[0077] 11 is a flowchart showing the flow of a process for calculating a predicted value by the prediction device 20. When predicting the amount of power consumption for a prediction target movement of the moving object 22, the prediction device 20 executes the process for calculating a predicted value according to the flow shown in FIG.

[0078] First, in S131, the prediction value calculation unit 48 acquires one or more factor values ​​representing one or more factors that affect the power consumption amount for the movement of the prediction target. For example, when the movement data creation unit 34 creates movement data prior to the movement of the moving body 22, the prediction value calculation unit 48 acquires one or more factor values ​​included in the movement data. Alternatively, when the acquisition unit 46 receives a prediction instruction input by a pilot or the like regardless of the movement of the moving body 22, the prediction value calculation unit 48 acquires one or more factor values ​​from the acquisition unit 46.

[0079] Next, in S132, the predicted value calculation unit 48 judges whether or not an error for past movement exists in the error table stored in the error storage unit 44. That is, the predicted value calculation unit 48 judges whether or not at least one result data is included in the error table. If there is no error for past movement in the error table, the predicted value calculation unit 48 advances the process to S133. If there is at least one error for past movement in the error table, the predicted value calculation unit 48 advances the process to S134.

[0080] In S133, the predicted value calculation unit 48 acquires the first prediction model from the model storage unit 40. That is, the predicted value calculation unit 48 acquires a prediction model that predicts the power consumption amount by using one or more factor values, without using an error between a predicted value and an actual measurement value. After completing the process of S133, the predicted value calculation unit 48 advances the process to S136.

[0081] In S134, the predicted value calculation unit 48 acquires a second prediction model from the model storage unit 40. That is, the predicted value calculation unit 48 acquires a prediction model that predicts the power consumption amount by using one or more factor values, using an error between a predicted value and an actual measurement value. After completing the process of S134, the predicted value calculation unit 48 advances the process to S135.

[0082] In S135, the prediction value calculation unit 48 acquires an error for a past movement from the error table stored in the error storage unit 44. In this embodiment, the prediction value calculation unit 48 acquires an error for a movement immediately before the movement to be predicted, among a plurality of past movements of the moving object 22. After completing the process of S135, the prediction value calculation unit 48 advances the process to S136.

[0083] In S136, the prediction value calculation unit 48 calculates a prediction value of the power consumption amount for the prediction target movement of the mobile object 22 based on the acquired one or more factor values ​​and the acquired prediction model. More specifically, if there is no error for the past movement, the prediction value calculation unit 48 calculates the prediction value using the acquired one or more factor values ​​and the first prediction model. If there is an error for the past movement, the prediction value calculation unit 48 calculates the prediction value using the acquired error, the acquired one or more factor values, and the second prediction model.

[0084] Next, in S137, the output unit 50 outputs the predicted value calculated by the predicted value calculation unit 48. For example, the output unit 50 transmits the predicted value to the moving object 22. In this case, the output unit 50 may transmit the movement identification information included in the corresponding movement data together with the predicted value. The output unit 50 may output the predicted value to another device. For example, the output unit 50 may output the calculated predicted value to a charging system that charges the moving object 22 with power.

[0085] As described above, the prediction device 20 according to the first embodiment calculates a predicted value of power consumption for the movement of the moving object 22 to be predicted, using the error between the predicted value and the actual measured value of the power consumption due to past movements and the prediction model. Such a prediction device 20 calculates a predicted value reflecting the error, even in cases where the power consumption varies greatly due to latent factors that are difficult to reflect in the prediction model as input variables, such as the pilot's proficiency, which is difficult to quantify, and values ​​that are difficult to measure. This allows the prediction device 20 to accurately calculate a predicted value of the power consumption consumed by the moving object 22.

[0086] Second embodiment Next, a prediction device 20 according to a second embodiment will be described. The prediction device 20 according to the second embodiment has substantially the same functions and configuration as the first embodiment, so in the second embodiment, elements having substantially the same functions and configuration as the first embodiment are given the same reference numerals, and detailed description will be omitted except for the differences. The same applies to each of the third and subsequent embodiments.

[0087] 12 is a diagram showing the configuration of a prediction device 20 according to the second embodiment, together with a moving object 22. The prediction device 20 according to the second embodiment further includes a representative value calculation unit 62 and a representative value storage unit 64.

[0088] When one or more errors for one or more past movements of the mobile object 22 are stored in the error table stored in the error storage unit 44, the representative value calculation unit 62 calculates a statistical representative value of the one or more errors. For example, the representative value calculation unit 62 calculates the average value, maximum value, minimum value, median, mode, first quartile, third quartile, variance, standard deviation, or standard error of the one or more errors as the representative value. The representative value calculation unit 62 may calculate the mean square error, mean absolute error, mean square error, mean square error rate, or mean absolute error rate as the representative value. For example, the representative value calculation unit 62 calculates the representative value every time the error calculation unit 42 creates result data and stores it in the error table.

[0089] The representative value storage unit 64 stores the representative value calculated by the representative value calculation unit 62. The representative value stored in the representative value storage unit 64 is updated to a new value every time a representative value is calculated by the representative value calculation unit 62.

[0090] When a representative value is stored in the representative value storage unit 64, that is, when an error exists for past movements, the predicted value calculation unit 48 according to the second embodiment acquires the representative value stored in the representative value storage unit 64 as an error. Then, the predicted value calculation unit 48 calculates a predicted value for the movement of the prediction target using the representative value acquired as the error and a prediction model. For example, the predicted value calculation unit 48 calculates a predicted value for the movement of the prediction target using one or more factor values, the representative value, and a second prediction model.

[0091] On the other hand, when a representative value is not stored in the representative value storage unit 64, that is, when there is no error for past movements, the predicted value calculation unit 48 according to the second embodiment calculates a predicted value for the movement of the prediction target using a prediction model without using the representative value. For example, the predicted value calculation unit 48 calculates a predicted value for the movement of the prediction target using one or more factor values ​​and the first prediction model.

[0092] 13 is a flowchart showing a process flow of the prediction device 20 of the second embodiment. In the second embodiment, when the moving object 22 finishes one unit of movement and the movement data is stored in the movement table, the prediction device 20 executes the process shown in FIG.

[0093] First, in S201, the evaluation data creation unit 38 acquires an actual measurement value of the amount of power actually consumed by the moving body 22 for the corresponding unit of movement and movement data for the corresponding unit of movement, and creates evaluation data including the acquired actual measurement value and movement data.

[0094] Next, in S202, the error calculation unit 42 calculates a predicted value of the power consumption using one or more factor values ​​extracted from the created evaluation data and the first prediction model. Furthermore, the error calculation unit 42 calculates the error between the calculated predicted value of the power consumption and the actual measured value. Then, the error calculation unit 42 creates one piece of result data including the calculated error and the evaluation data. The error calculation unit 42 stores the created result data in an error table stored in the error storage unit 44.

[0095] Next, in S203, when the error calculation section 42 creates result data and stores it in the error table, the representative value calculation section 62 calculates a representative value of one or more errors stored in the error table. Then, the representative value calculation section 62 stores the calculated representative value in the representative value storage section 64.

[0096] As described above, the prediction device 20 according to the second embodiment uses errors for multiple past movements and a prediction model to calculate a predicted value of the amount of power consumption for a prediction target movement of the moving body 22. Such a prediction device 20 calculates a predicted value reflecting errors for multiple past movements, and therefore can accurately calculate a predicted value of the amount of power consumption consumed by the moving body 22.

[0097] Third embodiment Next, a prediction device 20 according to a third embodiment will be described.

[0098] 14 is a diagram showing the configuration of a prediction device 20 according to the third embodiment, together with a moving object 22. The prediction device 20 according to the third embodiment further includes a driver table storage unit 66 and a driver-specific registration unit 68.

[0099] In the third embodiment, the collection unit 32 and the acquisition unit 46 acquire pilot identification information when acquiring one or more factor values ​​including a start position, a start time, and the like. The pilot identification information is information that identifies the pilot of the moving body 22. For example, the collection unit 32 and the acquisition unit 46 acquire, from the moving body 22, the pilot identification information that the pilot input to the moving body 22. Also, in the third embodiment, the movement data creation unit 34 creates movement data that further includes the pilot identification information.

[0100] The driver table storage unit 66 stores a driver table. The driver table stores errors for each driver. The driver table stores, for example, errors for the most recent movement of the moving body 22 by the driver indicated in the driver identification information. Alternatively, the driver table may store, as errors, one or more statistically representative values ​​of errors for one or more movements in which the driver indicated in the driver identification information has operated the moving body 22 in the past.

[0101] When the error calculation unit 42 creates new result data and stores it in the error table, the driver-specific registration unit 68 acquires the new result data from the error table. The driver-specific registration unit 68 acquires the driver identification information and the error contained in the acquired new result data. The driver-specific registration unit 68 then stores the acquired error in the driver table stored in the driver table storage unit 66 in association with the acquired driver identification information.

[0102] Alternatively, the driver-specific registration unit 68 may calculate a statistical representative value based on the driver identification information and one or more errors included in the acquired new result data. In this case, the driver-specific registration unit 68 identifies one or more result data whose driver identification information matches the new result data stored in the error table, and acquires one or more errors included in the identified one or more result data. The driver-specific registration unit 68 calculates the average value, maximum value, minimum value, median, mode, first quartile, third quartile, variance, standard deviation, standard error, or the like of the acquired one or more errors as a representative value. Then, the driver-specific registration unit 68 stores the calculated representative value as an error in the driver table stored in the driver table storage unit 66 in association with the acquired driver identification information.

[0103] When calculating a predicted value, the predicted value calculation unit 48 according to the third embodiment acquires driver identification information for the movement of the prediction target. When an error corresponding to the acquired driver identification information exists in the driver table, that is, when an error exists for a past movement by the same driver, the predicted value calculation unit 48 according to the third embodiment acquires an error for the past movement by the same driver from the driver table. Then, the predicted value calculation unit 48 calculates a predicted value for the movement of the prediction target using the acquired error and a prediction model. For example, the predicted value calculation unit 48 calculates a predicted value for the movement of the prediction target using one or more factor values, the acquired error, and a second prediction model.

[0104] On the other hand, when the error corresponding to the driver identification information does not exist in the driver table, that is, when there is no error for past movements of the same driver, the predicted value calculation unit 48 according to the third embodiment calculates a predicted value for the movement of the prediction target using a prediction model without using the error. For example, the predicted value calculation unit 48 calculates a predicted value for the movement of the prediction target using one or more factor values ​​and the first prediction model.

[0105] Fig. 15 is a diagram showing an example of a movement table according to the third embodiment. As shown in Fig. 15, the movement table according to the third embodiment further includes an item for storing operator identification information, compared to the movement table according to the first embodiment. Items other than the operator identification information in the movement table according to the third embodiment are similar to those in the movement table according to the first embodiment shown in Fig. 2.

[0106] Fig. 16 is a flowchart showing a process flow for updating errors in a driver table by the prediction device 20 according to the third embodiment. When the movement of the moving body 22 ends and the result data is stored in the error table, the prediction device 20 executes an error update process according to the flow shown in Fig. 16.

[0107] First, in S301, when the error calculation unit 42 creates new result data and stores it in the error table, the driver-specific registration unit 68 obtains the new result data from the error table.

[0108] Next, in S302, the driver-specific registration unit 68 acquires the driver identification information and error included in the acquired new result data. When calculating a statistical representative value, the driver-specific registration unit 68 identifies one or more result data stored in the error table whose driver identification information matches the new result data, and acquires one or more errors included in the identified one or more result data. Then, the driver-specific registration unit 68 calculates the average value, maximum value, minimum value, median, mode, first quartile, third quartile, variance, standard deviation, standard error, or the like of the acquired one or more errors as a representative value.

[0109] Next, in S303, the driver-specific registration unit 68 stores the acquired error or the calculated representative value in the driver table stored in the driver table storage unit 66 in association with the acquired driver identification information.

[0110] Fig. 17 is a diagram showing an example of the driver table. The driver table storage unit 66 stores the error for each driver identification information, for example, as shown in Fig. 17. This allows the predicted value calculation unit 48 to obtain the error corresponding to the driver who drives the moving object 22 from the driver table.

[0111] As described above, the prediction device 20 according to the third embodiment calculates a predicted value of the power consumption amount by using the error for past movements by the same operator and the prediction model. Since the prediction device 20 calculates a predicted value reflecting differences between operators, it can calculate a predicted value of the power consumption amount with higher accuracy.

[0112] (Fourth embodiment) Next, a prediction device 20 according to a fourth embodiment will be described.

[0113] 18 is a diagram showing a configuration of a prediction device 20 according to the fourth embodiment, together with a moving object 22. The prediction device 20 according to the fourth embodiment further includes a moving object table storage unit 70 and a moving object specific registration unit 72.

[0114] In the fourth embodiment, the collection unit 32 and the acquisition unit 46 acquire moving object identification information when acquiring one or more factor values ​​including a start position, a start time, and the like. The moving object identification information is information that identifies the moving object 22. For example, the collection unit 32 and the acquisition unit 46 acquire, from the moving object 22, moving object identification information that has been registered in advance for the moving object 22.

[0115] In the fourth embodiment, the movement data creation unit 34 creates movement data that further includes moving body identification information. Compared to the movement table according to the first embodiment, the movement table according to the fourth embodiment further includes an item for storing moving body identification information. Items in the movement table according to the fourth embodiment other than the moving body identification information are similar to those in the movement table according to the first embodiment.

[0116] The moving body table storage unit 70 stores a moving body table. The moving body table stores an error for each moving body 22. The moving body table stores, for example, an error for the most recent movement of the moving body 22 indicated in the moving body identification information. Alternatively, the moving body table may store, as an error, one or more statistically representative values ​​of errors for one or more past movements by the moving body 22 indicated in the moving body identification information.

[0117] When the error calculation unit 42 creates new result data and stores it in the error table, the moving object registration unit 72 acquires the new result data from the error table. The moving object registration unit 72 acquires the moving object identification information and the error included in the acquired new result data. The moving object registration unit 72 then stores the acquired error in the moving object table stored in the moving object table storage unit 70 in association with the acquired moving object identification information.

[0118] Alternatively, the mobile object registration unit 72 may calculate a statistical representative value based on the mobile object identification information and error included in the acquired new result data. In this case, the mobile object registration unit 72 identifies one or more result data whose mobile object identification information matches the new result data stored in the error table, and acquires one or more errors included in the identified one or more result data. The mobile object registration unit 72 calculates a statistical representative value of the acquired one or more errors. Then, the mobile object registration unit 72 stores the calculated representative value as an error in the mobile object table stored in the mobile object table storage unit 70 in association with the acquired mobile object identification information.

[0119] When calculating a predicted value, the predicted value calculation unit 48 according to the fourth embodiment acquires moving body identification information for the movement of the prediction target from the moving body 22. When an error corresponding to the acquired moving body identification information exists in the moving body table, that is, when an error exists for a past movement of the same moving body 22, the predicted value calculation unit 48 according to the fourth embodiment acquires an error for the past movement of the same moving body 22 from the moving body table. Then, the predicted value calculation unit 48 calculates a predicted value for the movement of the prediction target using the acquired error and a prediction model. For example, the predicted value calculation unit 48 calculates a predicted value for the movement of the prediction target using one or more factor values, the acquired error, and a second prediction model.

[0120] On the other hand, when there is no error corresponding to the acquired moving object identification information in the moving object table, that is, when there is no error for past movements of the same moving object 22, the predicted value calculation unit 48 according to the fourth embodiment calculates a predicted value for the movement of the prediction object using a prediction model without using the error. For example, the predicted value calculation unit 48 calculates a predicted value for the movement of the prediction object using one or more factor values ​​and a first prediction model.

[0121] As described above, the prediction device 20 according to the fourth embodiment calculates a predicted value of the power consumption amount by using a prediction model and errors for past movements of the same moving body 22. Such a prediction device 20 calculates a predicted value by reflecting differences between moving bodies 22, and therefore can calculate a predicted value of the power consumption amount with higher accuracy.

[0122] Fifth embodiment Next, a prediction device 20 according to a fifth embodiment will be described.

[0123] 19 is a diagram showing a configuration of a prediction device 20 according to the fifth embodiment, together with a moving object 22. The prediction device 20 according to the fifth embodiment further includes a route table storage unit 74 and a route-specific registration unit 76.

[0124] In the fifth embodiment, the collection unit 32 and the acquisition unit 46 acquire route identification information when acquiring one or more factor values ​​including a start position, a start time, and the like. The route identification information is information that identifies a route along which the mobile body 22 travels. For example, if the mobile body 22 is a route bus, the mobile body 22 may travel along a plurality of different routes. The route identification information identifies which of the multiple routes of such route buses, etc., the mobile body 22 travels along. For example, the collection unit 32 and the acquisition unit 46 acquire, from the mobile body 22, the route identification information that the operator inputted to the mobile body 22.

[0125] In the fifth embodiment, the movement data creation unit 34 creates movement data that further includes route identification information. Compared with the movement table according to the first embodiment, the movement table according to the fifth embodiment further includes an item for storing route identification information. Items other than the route identification information in the movement table according to the fifth embodiment are similar to those in the movement table according to the first embodiment.

[0126] The route table storage unit 74 stores a route table. The route table stores an error for each route. For example, the route table stores an error for the most recent movement of the route indicated in the route identification information. Alternatively, the route table may store, as an error, one or more statistical representative values ​​of errors for one or more past movements of the route indicated in the route identification information.

[0127] When the error calculation unit 42 creates new result data and stores it in the error table, the route-specific registration unit 76 acquires the new result data from the error table. The route-specific registration unit 76 acquires the route identification information and the error included in the acquired new result data. Then, the route-specific registration unit 76 stores the acquired error in the route table stored in the route table storage unit 74 in association with the acquired route identification information.

[0128] Alternatively, the route-specific registration unit 76 may calculate a statistical representative value based on the route identification information and error included in the acquired new result data. In this case, the route-specific registration unit 76 identifies one or more result data whose route identification information matches the new result data stored in the error table, and acquires one or more errors included in the identified one or more result data. The route-specific registration unit 76 calculates a statistical representative value of the acquired one or more errors. Then, the route-specific registration unit 76 stores the calculated representative value as an error in the route table stored in the route table storage unit 74 in association with the acquired route identification information.

[0129] When calculating a predicted value, the predicted value calculation unit 48 according to the fifth embodiment acquires route identification information for the movement of the prediction target from the moving object 22. When an error corresponding to the acquired route identification information exists in the route table, that is, when an error exists for a past movement of the same route, the predicted value calculation unit 48 according to the fifth embodiment acquires the error for the past movement of the same route from the route table. Then, the predicted value calculation unit 48 calculates a predicted value for the movement of the prediction target using the acquired error and a prediction model. For example, the predicted value calculation unit 48 calculates a predicted value for the movement of the prediction target using one or more factor values, the acquired error, and a second prediction model.

[0130] On the other hand, when there is no error corresponding to the acquired route identification information in the route table, that is, when there is no error for a past movement of the same route, the predicted value calculation unit 48 according to the fifth embodiment calculates a predicted value for the movement of the prediction target using a prediction model without using the error. For example, the predicted value calculation unit 48 calculates a predicted value for the movement of the prediction target using one or more factor values ​​and a first prediction model.

[0131] As described above, the prediction device 20 according to the fifth embodiment calculates a predicted value of the power consumption by using an error for past movements along the same route and a prediction model. Since the prediction device 20 calculates a predicted value by reflecting differences in the routes, it can calculate a predicted value of the power consumption with higher accuracy.

[0132] Sixth embodiment Next, a prediction device 20 according to a sixth embodiment will be described.

[0133] 20 is a diagram showing a configuration of a prediction device 20 according to a sixth embodiment, together with a moving object 22. The prediction device 20 according to the sixth embodiment further includes a set table storage unit 78 and a set-by-set registration unit 80.

[0134] In the sixth embodiment, when acquiring one or more factor values ​​including a start position, a start time, etc., the collection unit 32 and the acquisition unit 46 acquire a set including two or more predetermined pieces of information among the pilot identification information, the moving object identification information, and the route identification information. The pilot identification information is the same as in the third embodiment. The moving object identification information is the same as in the fourth embodiment. The route identification information is the same as in the fifth embodiment.

[0135] In the sixth embodiment, the travel data creation unit 34 creates travel data that further includes a set including two or more predetermined pieces of information from among the operator identification information, the moving body identification information, and the route identification information. Compared to the travel table according to the first embodiment, the travel table according to the sixth embodiment further includes an item for storing two or more predetermined pieces of the operator identification information, the moving body identification information, and the route identification information. Items in the travel table according to the sixth embodiment other than the operator identification information, the moving body identification information, and the route identification information are similar to those in the travel table according to the first embodiment.

[0136] When the mobile object 22 has traveled one or more times in the past along the route indicated in the route identification information, one or more statistically representative values ​​of the errors for this one or more travels may be stored as the error.

[0137] The set table storage unit 78 stores a set table. The set table stores an error for each set. For example, the set table stores an error for the most recent movement of a combination of the pilot, the moving body 22, and the route indicated in the set identification information. Alternatively, the set table may store, as an error, one or more statistical representative values ​​of errors for one or more movements in the combination of the pilot, the moving body 22, and the route indicated in the set identification information.

[0138] When the error calculation unit 42 creates new result data and stores it in the error table, the set-specific registration unit 80 acquires the new result data from the error table. The set-specific registration unit 80 acquires a set and an error, which includes two or more pieces of predetermined information among the pilot identification information, the moving object identification information, and the route identification information, included in the acquired new result data. Then, the set-specific registration unit 80 stores the acquired error in the set table stored in the set table storage unit 78 in association with the acquired set.

[0139] Alternatively, the set-specific registration unit 80 may calculate a statistical representative value based on a set including two or more pieces of predetermined information among the driver identification information, the moving object identification information, and the route identification information included in the acquired new result data, and the error. In this case, the set-specific registration unit 80 identifies one or more pieces of result data that match the new result data stored in the error table and a set including two or more pieces of predetermined information among the driver identification information, the moving object identification information, and the route identification information, and acquires one or more errors included in the identified one or more pieces of result data. Then, the set-specific registration unit 80 calculates a statistical representative value of the acquired one or more errors. Then, the set-specific registration unit 80 stores the calculated representative value in the set table stored in the set table storage unit 78 as an error, in association with the acquired set.

[0140] When calculating a predicted value, the predicted value calculation unit 48 according to the sixth embodiment acquires from the moving body 22 a set including two or more pieces of information, which are predetermined among the operator identification information, the moving body identification information, and the route identification information, for the movement of the prediction target. When an error corresponding to the acquired set exists in the set table, that is, when an error exists for a past movement in which a set including two or more pieces of information, which are predetermined among the operator, the moving body 22, and the route, exists, the predicted value calculation unit 48 according to the sixth embodiment acquires an error for the past movement in which the set is the same from the set table. Then, the predicted value calculation unit 48 calculates a predicted value for the movement of the prediction target using the acquired error and the prediction model. For example, the predicted value calculation unit 48 calculates a predicted value for the movement of the prediction target using one or more factor values, the acquired error, and the second prediction model.

[0141] On the other hand, when the error corresponding to the acquired set does not exist in the set table, that is, when there is no error for a past movement in which a set including two or more predetermined pieces of information of the pilot, the moving body 22, and the route is the same, the predicted value calculation unit 48 according to the sixth embodiment calculates a predicted value for the movement of the prediction target using a prediction model without using the error. For example, the predicted value calculation unit 48 calculates a predicted value for the movement of the prediction target using one or more factor values ​​and a first prediction model.

[0142] As described above, the prediction device 20 according to the fifth embodiment calculates a predicted value of the power consumption amount by using a prediction model and an error for a movement for a set including two or more pieces of information of the same past operator, moving body 22, and route. Since the prediction device 20 calculates a predicted value by reflecting the difference between such sets, it can calculate a predicted value of the power consumption amount with higher accuracy.

[0143] In the third to sixth embodiments, examples have been described in which the power consumption is predicted using an error or a representative value for past travels in which the operator, the moving body 22, the route, or the set is the same when the operator, the moving body 22, or the route is the same, or when a set including any two or more pieces of information of the operator, the moving body 22, and the route is the same. The prediction device 20 is not limited to these, and may perform similar processing using an error or a representative value for past travels in which other information is the same. For example, the prediction device 20 may perform similar processing using an error or a representative value for past travels in which the start time, the end time, or the time period during which the travel is the same.

[0144] Seventh embodiment Next, a prediction device 20 according to a seventh embodiment will be described.

[0145] 21 is a diagram showing a configuration of a prediction device 20 according to the seventh embodiment, together with a moving object 22. The prediction device 20 according to the seventh embodiment further includes a category table storage unit 82 and a category calculation unit 84.

[0146] The category table storage unit 82 stores a category table. The category table stores category values ​​and model input values ​​corresponding to the category values. The category values ​​are values ​​that indicate to which category an error belongs. The model input values ​​corresponding to the category values ​​are values ​​that are input to a prediction model as errors. The category values ​​and the model input values ​​may be the same value.

[0147] When the error calculation unit 42 creates new result data and stores it in the error table, the category calculation unit 84 acquires the error table from the error storage unit 44. The category calculation unit 84 calculates the travel efficiency, which indicates the travel distance relative to the actual measurement value, for each of the multiple result data included in the acquired error table. The travel efficiency is called, for example, electricity efficiency and is expressed in units such as km / kWh.

[0148] The category calculation unit 84 generates an approximation curve representing the relationship between the movement efficiency and the error of the multiple result data based on the relationship between the movement efficiency and the error for each of the multiple result data. The approximation curve may be a straight line. For example, the category calculation unit 84 generates corresponding points for each of the multiple result data with the movement efficiency as x and the error as y. Then, the category calculation unit 84 generates an approximation curve that is most approximate to the generated corresponding points. The category calculation unit 84 generates an approximation curve represented by, for example, a function y=h(x).

[0149] Furthermore, the category calculation unit 84 reads out an efficiency setting value for classifying the movement efficiency into categories from the setting file. The category calculation unit 84 inputs the read out efficiency setting value into an approximation curve and calculates a threshold value θ. For example, the category calculation unit 84 inputs the efficiency setting value into x of a function y=h(x) representing the approximation curve and calculates a threshold value θ.

[0150] Furthermore, the category calculation unit 84 obtains an error for the most recent movement among the multiple past movements included in the error table, and compares the obtained error with the calculated threshold value θ to determine which category the error for the most recent movement belongs to. Then, the category calculation unit 84 calculates a category value representing the determined category. Alternatively, the category calculation unit 84 may obtain multiple errors for the multiple past movements included in the error table, and calculate a category value representing a category that includes the most multiple errors, or a category value representing a category that includes a statistically representative value of the multiple errors.

[0151] The category calculation unit 84 stores the calculated category value in the category table.

[0152] When calculating a predicted value, the predicted value calculation unit 48 according to the seventh embodiment acquires a category table from the category table storage unit 82. When a category value exists in the category table, that is, when an error exists for a past movement, the predicted value calculation unit 48 according to the seventh embodiment acquires a model input value corresponding to the category value from the category table as an error. Then, the predicted value calculation unit 48 calculates a predicted value for the movement to be predicted using the acquired model input value and the prediction model.

[0153] On the other hand, when there is no category value in the category table, i.e., when there is no error for past movements, the predicted value calculation unit 48 in the seventh embodiment calculates a predicted value for the movement to be predicted using a prediction model without using model input values.

[0154] The category calculation unit 84 may calculate a category value for each driver, moving body 22, or route, or for each set including any two or more pieces of information of the driver, moving body 22, and route, as in the third to sixth embodiments. In this case, the prediction value calculation unit 48 calculates a prediction value using a model input value corresponding to a category value for a past movement in which the driver, moving body 22, route, or set is the same, as in the third to sixth embodiments, when the driver, moving body 22, or route is the same, or when a set including any two or more pieces of information of the driver, moving body 22, and route is the same. Furthermore, the prediction device 20 may perform a similar process using a category value for a past movement in which other information is the same, without being limited to these. For example, the prediction device 20 may perform a similar process using a category value for a past movement in which the start time, end time, or time period during which the movement is the same.

[0155] FIG. 22 is a diagram showing an example of an approximation curve. The error may be closely related to the movement efficiency. For example, as shown in FIG. 22, the movement efficiency may be approximated by a function y=h(x) where x is the movement efficiency and y is the error. The function y=h(x) may be a general function such as a linear function, a quadratic function, an exponential function, a logarithmic function, and a trigonometric function. The prediction device 20 can uniquely determine the range of the threshold for each category range of the movement efficiency by using such a function y=h(x).

[0156] Fig. 23 is a flowchart showing a process flow for updating a category value by the prediction device 20 according to the seventh embodiment. Fig. 24 is a diagram showing an example of a setting file read in S704 in Fig. 23. Fig. 25 is a diagram showing an example of a threshold value θ determined in S705 in Fig. 23.

[0157] When the movement of the moving object 22 ends and the result data is stored in the error table, the prediction device 20 executes an error update process according to the flow shown in FIG.

[0158] First, in S701, when the error calculation section 42 creates new result data and stores it in the error table, the category calculation section 84 acquires the error table from the error storage section 44.

[0159] Next, in S702, the category calculation unit 84 calculates the movement efficiency representing the movement distance relative to the actual measurement value for each of the multiple result data included in the acquired error table. For example, when the actual measurement value included in the i-th (i is an integer equal to or greater than 1) result data is y_i and the movement distance is x_dis^(i), the category calculation unit 84 calculates the movement efficiency γ_i by the following formula. γ_i=y_i / x_dis^(i)

[0160] Next, in S703, the category calculation unit 84 generates an approximation curve representing the relationship between the movement efficiency and the error for each of the multiple result data, based on the relationship between the movement efficiency and the error for each of the multiple result data. For example, the category calculation unit 84 generates corresponding points for each of the multiple result data, with the movement efficiency being x and the error being y. Then, the category calculation unit 84 fits the coefficients of a predetermined function, etc., so as to be most approximate to the multiple corresponding points generated, to generate a function y=h(x) representing the approximation curve.

[0161] Next, in S704, the category calculation unit 84 reads out an efficiency setting value for classifying the movement efficiency into categories from a setting file created in advance. The setting file is created in advance by, for example, a designer or the like. For example, the setting file may be table data in which efficiency setting values ​​and category values ​​are stored in columns as shown in FIG. 24. For example, when the movement efficiency is classified into a first category with low efficiency and a second category with high efficiency, the setting file stores, for example, the maximum movement efficiency of the first category as the efficiency setting value. Note that, when further classification into more categories is performed, the setting file may store two or more efficiency setting values ​​representing the maximum movement efficiency of each category.

[0162] Next, in S705, the category calculation unit 84 inputs the read efficiency setting value into the approximation curve to calculate the threshold value θ. For example, the category calculation unit 84 inputs the efficiency setting value into x of a function y=h(x) representing the approximation curve to calculate the threshold value θ. For example, the category calculation unit 84 may manage the threshold value θ using table data in which the threshold value θ and the category value are stored in columns, as shown in FIG. 25. Note that when multiple efficiency setting values ​​are read from the setting file, the category calculation unit 84 calculates the threshold value θ corresponding to each of the multiple efficiency setting values.

[0163] Next, in S706, the category calculation unit 84 calculates a category value based on the calculated threshold value θ and one or more errors stored in the error table. The method of calculating the category value will be described later with reference to Figs. 26 and 27.

[0164] Then, in S707, the category calculation section 84 stores the calculated category value in the category table.

[0165] FIG. 26 is a diagram showing a first example of the calculation process of the category value in S706.

[0166] When multiple pieces of result data for multiple past movements are stored in the error table, the category calculation unit 84, for example, acquires an error included in the result data for the most recent movement among the multiple past movements. The category calculation unit 84 may then compare the acquired error with one or more calculated thresholds to determine which category the error for the most recent movement belongs to. For example, as shown in FIG. 26, in the case of classifying into one of the first category, the second category, and the third category, the threshold θ representing the boundary between the first category and the second category is θ1, and the threshold θ representing the boundary between the second category and the third category is θ2. In this case, the category calculation unit 84 determines that the error belongs to the second category if it is between θ1 and θ2.

[0167] FIG. 27 is a diagram showing a second example of the category value calculation process in S706.

[0168] When multiple result data for multiple past movements are stored in the error table, the category calculation unit 84, for example, acquires multiple errors included in the multiple result data for multiple past movements. The category calculation unit 84 may then calculate a category value representing the category that includes the largest number of errors among the acquired multiple errors. For example, as shown in FIG. 27, in the case of classifying errors into any of the first category, the second category, and the third category, assume that the first category includes three errors, the second category includes two errors, and the first category includes one error. In such a case, the category calculation unit 84 takes a majority vote of the number of errors and determines that the error is included in the first category.

[0169] Furthermore, the category calculation unit 84 may calculate a statistical representative value of the acquired errors, and calculate a category value that indicates the category that the calculated representative value is included in. For example, the category calculation unit 84 may calculate the average value, median value, maximum value, minimum value, first quartile, third quartile, or mode of the errors as the representative value.

[0170] FIG. 28 is a diagram showing the correspondence between category values ​​and model input values.

[0171] The category table storage unit 82 stores a category table that stores category values ​​and model input values ​​corresponding to the category values. The model input values ​​are values ​​obtained by converting the category values ​​into numerical values ​​that can be input to a prediction model. The model input values ​​may have, for example, dummy variables in a number corresponding to a number of categories. In this case, each of the multiple dummy variables indicates whether or not it belongs to the corresponding category. For example, among the multiple dummy variables, the dummy variable corresponding to the category to which the error belongs is set to 1, and the other dummy variables are set to 0.

[0172] As described above, the prediction device 20 according to the seventh embodiment calculates a category value indicating which of a plurality of categories classified by one or a plurality of threshold values ​​the error for the past movement belongs to. Then, the prediction device 20 according to the seventh embodiment calculates a predicted value of the amount of power consumption for the movement of the moving object 22 to be predicted, using the category value and the prediction model. When the error is closely related to the movement efficiency, such a prediction device 20 can easily improve the accuracy of the predicted value reflecting the error.

[0173] Eighth embodiment Next, a prediction device 20 according to an eighth embodiment will be described.

[0174] The prediction device 20 according to the eighth embodiment has the same configuration as that of the first embodiment. However, the prediction device 20 according to the eighth embodiment stores, as a second prediction model, a model in which an error and one or more factor values ​​are input variables and the power consumption is an objective variable, as shown in the following formula. m2:y_i=f(e,x_dis,x_up,x_fall,…)

[0175] f() is a function representing a prediction model. f() may be any model as long as it is in the form of a general model such as a multiple regression model, a neural network, deep learning, or a physical model. e represents an error. Parameters such as regression coefficients in such a second prediction model are calculated by a general learning method such as the least squares method using, for example, the error, multiple factor values, and actual measured values ​​contained in the result data for multiple past movements stored in the error table as learning data.

[0176] The prediction device 20 of the eighth embodiment can improve the accuracy of the predicted value that reflects the error by using a prediction model in which the error and one or more factor values ​​are a group of input variables and the power consumption is the objective variable.

[0177] Ninth embodiment Next, a prediction device 20 according to a ninth embodiment will be described.

[0178] FIG. 29 is a diagram showing an example of an output image 90 that the prediction device 20 causes to be displayed on a display device.

[0179] The output unit 50 of the prediction device 20 according to the ninth embodiment causes a display device to display an output image 90 as shown in Fig. 29. The output image 90 includes an error log image 92, a movement data log image 94, an error image 96, and a comparison image 98.

[0180] The error log image 92 is a graph plotting errors included in multiple result data stored in an error table. For example, the error log image 92 is a graph showing date and time on the horizontal axis and errors on the vertical axis. Note that the error log image 92 is not limited to such a graph, and may be a graph showing other factors such as travel time or travel distance on the horizontal axis. By displaying such an error log image 92, the prediction device 20 can show the administrator, operator, etc. the trend of errors for each factor.

[0181] The movement data log image 94 shows movement data for, for example, a specified movement stored in the movement table. For example, the movement data log image 94 shows movement data at a specified date and time.

[0182] The error image 96 is an image representing an error to be input to the prediction model. When a representative value of a plurality of errors is input to the prediction model, the error image 96 may be an image representing the representative value.

[0183] The comparison image 98 is an image showing the error between the predicted value and the actual measurement value when calculated using the error and the prediction model, and the error between the predicted value and the actual measurement value when calculated without using the error. By displaying such a comparison image 98, the prediction device 20 can show the manager, pilot, etc., the degree to which the predicted value has been improved by inputting the error into the prediction model.

[0184] The output unit 50 included in the prediction device 20 according to the second to eighth embodiments may also display a similar output image 90 on a display device. The output unit 50 according to the third to sixth embodiments may display an output image 90 for an error for each set including a pilot, a moving object 22, or a route, or two or more of these. The output unit 50 according to the sixth embodiment may display an output image 90 in which the error includes a category value.

[0185] The prediction device 20 according to the ninth embodiment as described above can show the degree or tendency of improvement of the predicted value to a manager, operator, or the like.

[0186] (Hardware configuration) Fig. 30 is a diagram showing an example of a hardware configuration of the prediction device 20 according to each embodiment. The prediction device 20 is realized by a computer having a hardware configuration as shown in Fig. 30, for example. The prediction device 20 includes a CPU (Central Processing Unit) 301, a RAM (Random Access Memory) 302, a ROM (Read Only Memory) 303, an operation input device 304, a display device 305, a storage device 306, and a communication device 307. These units are connected by a bus.

[0187] The CPU 301 is a processor that executes arithmetic processing, control processing, etc. according to a program. The CPU 301 uses a predetermined area of ​​the RAM 302 as a working area and executes various processes in cooperation with programs stored in the ROM 303, the storage device 306, etc.

[0188] The RAM 302 is a memory such as a Synchronous Dynamic Random Access Memory (SDRAM), and functions as a work area for the CPU 301. The ROM 303 is a memory that stores programs and various information in a non-rewritable manner.

[0189] The operation input device 304 is an input device such as a mouse, a keyboard, etc. The operation input device 304 receives information input by a user as an instruction signal, and outputs the instruction signal to the CPU 301.

[0190] The display device 305 is a display device such as an LCD (Liquid Crystal Display), etc. The display device 305 displays various information based on a display signal from the CPU 301.

[0191] The storage device 306 is a device that writes and reads data to a semiconductor storage medium such as a flash memory, or a magnetically or optically recordable storage medium, etc. The storage device 306 writes and reads data to the storage medium in response to control from the CPU 301. The communication device 307 communicates with external devices via a network in response to control from the CPU 301.

[0192] The program executed by the computer has a modular configuration including a collection module, a movement data creation module, an evaluation data creation module, an error calculation module, an acquisition module, a predicted value calculation module, and an output module. The program may further include at least one of a representative value calculation module, a pilot-specific registration module, a moving object-specific registration module, a route-specific registration module, and a category calculation module.

[0193] This program is deployed on the RAM 302 by the CPU 301 (processor) and executed, thereby causing the computer to function as the collection unit 32, the movement data creation unit 34, the evaluation data creation unit 38, the error calculation unit 42, the acquisition unit 46, the prediction value calculation unit 48, and the output unit 50. This program may further cause the computer to function as at least one of the representative value calculation unit 62, the pilot-specific registration unit 68, the moving object registration unit 72, the route-specific registration unit 76, the set-specific registration unit 80, and the category calculation unit 84. Note that some or all of the collection unit 32, the movement data creation unit 34, the evaluation data creation unit 38, the error calculation unit 42, the acquisition unit 46, the prediction value calculation unit 48, the output unit 50, the representative value calculation unit 62, the pilot-specific registration unit 68, the moving object registration unit 72, the route-specific registration unit 76, the set-specific registration unit 80, and the category calculation unit 84 may be realized by a hardware circuit. Additionally, the RAM 302 and the storage device 306 function as the movement history storage unit 36, the model storage unit 40, and the error storage unit 44. Additionally, the RAM 302 and the storage device 306 may further function as at least one of the representative value storage unit 64, the operator table storage unit 66, the moving object table storage unit 70, the route table storage unit 74, the set table storage unit 78, and the category table storage unit 82.

[0194] In addition, the program executed by the computer is provided as a file in a format that can be installed on the computer or in a format that can be executed by the computer, and is recorded on a computer-readable recording medium such as a CD-ROM, a flexible disk, a CD-R, or a DVD (Digital Versatile Disk).

[0195] The program may be stored on a computer connected to a network such as the Internet and provided by being downloaded via the network. The program may be provided or distributed via a network such as the Internet. The program executed by the prediction device 20 may be provided by being pre-installed in the ROM 303 or the like.

[0196] Although some embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their modifications are included in the scope and spirit of the invention, and are included in the scope of the invention and its equivalents described in the claims. [Explanation of symbols]

[0197] 20 Prediction Device 22 Mobile 32 Collection Department 34. Mobile Data Creation Department 36 Movement history memory unit 38 Evaluation Data Creation Department 40 Model memory section 42 Error calculation section 44 Error storage section 46 Acquisition Department 48 Prediction value calculation section 50 Output section 62 Representative value calculation section 64 Representative value storage section 66 Pilot table memory section 68 Pilot Registration Section 70 Mobile table storage unit 72 Mobile Registration Section 74 Route table storage unit 76 Route Registration Section 78 Set table storage unit 80 Set Registration Section 82 Category table storage section 84 Category Calculation Unit

Claims

1. a prediction value calculation unit that calculates a prediction value of the power consumption in the movement of a prediction target based on a prediction model using a power consumption amount of a moving body as a response variable, one or more factor values ​​that affect the power consumption amount for the movement of the prediction target of the moving body, and an error between a prediction value based on the prediction model and an actual measurement value; The prediction value calculation unit calculates the prediction value for the movement of the prediction target by using the error for a movement immediately preceding the movement of the prediction target among a plurality of past movements of the moving object and the prediction model. Information processing device.

2. an acquisition unit that acquires the one or more factor values ​​for a movement of the moving body that is to be predicted; an output unit that outputs the predicted value; The information processing device according to claim 1 , further comprising:

3. The predicted value calculation unit calculates the predicted value for the movement to be predicted by using the prediction model, an error between a past predicted value of the amount of power consumption for the past movement of the moving object and an actual measured value of the amount of power consumption consumed in the past movement, the one or more acquired factor values, and the prediction model. The information processing device according to claim 2 .

4. The predicted value calculation unit when there is no error for the past movement, calculating the predicted value for the movement to be predicted using a first prediction model having the one or more factor values ​​as a group of input variables and the power consumption as a response variable; When the error exists for the past movement, the predicted value is calculated by adding the error to a value predicted using the first prediction model. The predicted value for the movement of the prediction target is calculated using a second prediction model. The information processing device according to claim 3 .

5. The predicted value calculation unit when there is no error for the past movement, calculating the predicted value for the movement to be predicted using a first prediction model having the one or more factor values ​​as a group of input variables and the power consumption as a response variable; When the error exists for the past movement, the error and the one or more factor values ​​are set as a group of input variables, and the power consumption amount is set as a response variable to calculate the predicted value for the movement to be predicted using a second prediction model. The information processing device according to claim 3 .

6. The predicted value calculation unit calculates the predicted value for the movement of the prediction target by using a statistical representative value of one or more of the errors for one or more past movements of the moving object and the prediction model. The information processing device according to claim 3 .

7. A prediction model having a power consumption of a moving body as a target variable, a prediction value calculation unit that calculates a prediction value of the power consumption of the moving body in the movement of a prediction target based on one or more factor values ​​that affect the power consumption of the moving body in the movement of the prediction target, and an error between a prediction value by the prediction model and an actual measurement value, When the error exists for a past movement by the same operator, the predicted value calculation unit calculates the predicted value for the movement of the prediction target by using the error for the past movement by the same operator and the prediction model. Information processing device.

8. A prediction model having a power consumption amount of a moving body as a target variable, a prediction value calculation unit that calculates a prediction value of the power consumption amount in the movement of a prediction target based on one or more factor values ​​that affect the power consumption amount of the moving body for the prediction target, and an error between a prediction value by the prediction model and an actual measurement value, The predicted value calculation unit, when there is an error for a past movement of the same moving body, calculates the predicted value for the movement of the prediction target by using the error for the past movement of the same moving body and the prediction model. Information processing device.

9. A prediction model having a power consumption amount of a moving body as a target variable, a prediction value calculation unit that calculates a prediction value of the power consumption amount in the movement of a prediction target based on one or more factor values ​​that affect the power consumption amount of the moving body for the prediction target, and an error between a prediction value by the prediction model and an actual measurement value, The predicted value calculation unit, when there is an error for a past movement in which the moving body moves along the same route, calculates the predicted value for the movement of the prediction target by using the error for the past movement in which the route is the same and the prediction model. Information processing device.

10. A prediction model having a power consumption of a moving body as a target variable, and a prediction value calculation unit which calculates a prediction value of the power consumption of the moving body in the movement of a prediction target based on one or more factor values ​​which affect the power consumption of the moving body in the movement of the prediction target, and an error between a prediction value by the prediction model and an actual measurement value, When there is an error for a past movement for which a set including any two or more predetermined pieces of information of a pilot, the moving body, and a route along which the moving body moves is the same, the predicted value calculation unit calculates the predicted value for the movement of the prediction target using the error for the past movement for which the set is the same and the prediction model. Information processing device.

11. A prediction value calculation unit that calculates a prediction value of the power consumption of a moving object based on a prediction model having the power consumption of a moving object as a target variable, one or more factor values ​​that affect the power consumption of the moving object relative to the movement of the moving object, and an error between a prediction value by the prediction model and an actual measurement value; a category calculation unit that calculates one or more thresholds for categorizing the error based on an approximation curve that indicates a relationship between a movement efficiency that indicates a ratio of the actual measurement value to a movement distance of the moving body and the error, and calculates a category value that indicates which of a plurality of categories classified by the one or more thresholds the error for a past movement belongs to; Equipped with The predicted value calculation unit calculates the predicted value for the movement of the prediction target by using a model input value corresponding to the category value as the error when the error exists for the past movement. Information processing device.

12. The category calculation unit calculates the category value representing a category including the error for a most recent movement of the prediction target among a plurality of past movements of the moving object. The information processing device according to claim 11.

13. The category calculation unit calculates the category value representing a category that most includes the plurality of errors for the plurality of past movements of the moving object, or the category value representing a category that includes a statistically representative value of the plurality of errors for the plurality of past movements of the moving object. The information processing device according to claim 11.

14. A prediction value calculation unit that calculates a prediction value of the power consumption of a moving object based on a prediction model having a power consumption amount of a moving object as a target variable, one or more factor values ​​that affect the power consumption amount of the moving object relative to the movement of the moving object, and an error between a prediction value by the prediction model and an actual measurement value; an output unit that outputs the predicted value; Equipped with The output unit displays the error when calculated using the error and the prediction model, and the error between the predicted value and the actual measurement value when calculated without using the error. Information processing device.

15. An information processing method for predicting power consumption of a moving object by an information processing device, comprising: calculating a predicted value of the power consumption in the movement of the prediction target based on a prediction model having a power consumption amount of the moving body as a response variable, one or more factor values ​​that affect the power consumption amount for the movement of the prediction target of the moving body, and an error between a predicted value by the prediction model and an actual measured value, In calculating the predicted value of the power consumption amount, the information processing device calculates the predicted value for the movement of the prediction target by using the error for a movement immediately before the movement of the prediction target among a plurality of past movements of the moving object and the prediction model. Information processing methods.

16. An information processing method for predicting power consumption of a moving object by an information processing device, comprising: calculating a predicted value of the power consumption in the movement of the prediction target based on a prediction model having a power consumption amount of the moving body as a response variable, one or more factor values ​​that affect the power consumption amount for the movement of the prediction target of the moving body, and an error between a predicted value by the prediction model and an actual measured value, In calculating the predicted value of the power consumption, if there is an error for a past movement in which the operator is the same, the information processing device calculates the predicted value for the movement to be predicted using the error for the past movement in which the operator is the same and the prediction model. Information processing methods.

17. An information processing method for predicting power consumption of a moving object by an information processing device, comprising: calculating a predicted value of the power consumption in the movement of the prediction target based on a prediction model having a power consumption amount of the moving body as a response variable, one or more factor values ​​that affect the power consumption amount for the movement of the prediction target of the moving body, and an error between a predicted value by the prediction model and an actual measured value, In calculating the predicted value of the power consumption, if there is an error for a past movement of the same moving body, the information processing device calculates the predicted value for the movement of the prediction target using the error for the past movement of the same moving body and the prediction model. Information processing methods.

18. An information processing method for predicting power consumption of a moving object by an information processing device, comprising: calculating a predicted value of the power consumption in the movement of the prediction target based on a prediction model having a power consumption amount of the moving body as a response variable, one or more factor values ​​that affect the power consumption amount for the movement of the prediction target of the moving body, and an error between a predicted value by the prediction model and an actual measured value, In calculating the predicted value of the power consumption, if there is an error for a past movement in which the moving object moves along the same route, the information processing device calculates the predicted value for the movement to be predicted using the error for the past movement in which the moving object moves along the same route and the prediction model. Information processing methods.

19. An information processing method for predicting power consumption of a moving object by an information processing device, comprising: calculating a predicted value of the power consumption in the movement of the prediction target based on a prediction model having a power consumption amount of the moving body as a response variable, one or more factor values ​​that affect the power consumption amount for the movement of the prediction target of the moving body, and an error between a predicted value by the prediction model and an actual measured value, In calculating the predicted value of the power consumption amount, if there is an error for a past movement for which a set including any two or more predetermined pieces of information of a pilot, the moving body, and a route along which the moving body moves is the same, the information processing device calculates the predicted value for the movement of the prediction target using the error for the past movement for which the set is the same and the prediction model. Information processing methods.

20. An information processing method for predicting power consumption of a moving object by an information processing device, comprising: calculating a predicted value of the power consumption in the movement of the prediction target based on a prediction model having a power consumption amount of the moving body as a response variable, one or more factor values ​​that affect the power consumption amount for the movement of the prediction target of the moving body, and an error between a predicted value by the prediction model and an actual measured value, the information processing device calculates one or more thresholds for categorizing the error based on an approximation curve that represents a relationship between a movement efficiency that represents a ratio of the actual measurement value to a movement distance of the moving body and the error, and calculates a category value that represents in which of a plurality of categories classified by the one or more thresholds the error for the past movement is included; In the calculation of the predicted value of the power consumption, if there is an error for the past movement, the information processing device calculates the predicted value for the movement to be predicted using a model input value corresponding to the category value as the error. Information processing methods.

21. An information processing method for predicting power consumption of a moving object by an information processing device, comprising: calculating a predicted value of the power consumption in the movement of the prediction target based on a prediction model having a power consumption amount of the moving body as a response variable, one or more factor values ​​that affect the power consumption amount for the movement of the prediction target of the moving body, and an error between a predicted value by the prediction model and an actual measured value, The information processing device displays the error when calculated using the error and the prediction model, and the error between the predicted value and the actual measurement value when calculated without using the error. Information processing methods.

22. A program for causing a computer to function as an information processing device described in any one of claims 1 to 14.

23. A mobile object comprising the information processing device according to any one of claims 1 to 14.

Images