JP2021033583A - Control apparatus, control system, and control method - Google Patents

Abstract

To download an appropriate model.SOLUTION: A control apparatus includes a calculation unit and a download unit. The calculation unit calculates an evaluation value of a mode, on the basis of output data to be obtained by inputting evaluation data, which is input data for evaluation, to the model generated by machine learning. When the evaluation value calculated by the calculation unit is equal to or smaller than a threshold, the download unit downloads an update model different from the above model from a server device. The download unit transmits evaluation data and a teacher label for the evaluation data to the server device, and downloads an update model selected based on the evaluation data and the teacher data from the server device.SELECTED DRAWING: Figure 2

Description

The present invention relates to control devices, control systems and control methods.

Conventionally, there is a control device that downloads a new model from a server device when the reliability of the model generated by machine learning decreases (see, for example, Patent Document 1). Such a control device determines whether or not the model needs to be downloaded, and then downloads a new model.

JP-A-2018-173860

However, with the prior art, there was room for improvement in downloading the optimal model. Specifically, in the prior art, it is not known to which model the server device should be updated, and there is a risk that the reliability of the downloaded model will be lower than that of the model before downloading.

The present invention has been made in view of the above, and an object of the present invention is to provide a control device, a control system, and a control method capable of downloading an appropriate model.

In order to solve the above-mentioned problems and achieve the object, the control device according to the embodiment includes a calculation unit and a download unit. The calculation unit calculates the evaluation value of the model based on the output data obtained by inputting the evaluation data, which is the input data for evaluation, into the model generated by machine learning. When the evaluation value calculated by the calculation unit is equal to or less than the threshold value, the download unit downloads an update model different from the model from the server device. Further, the download unit transmits the evaluation data and the teacher label of the evaluation data to the server device, and the update model selected based on the evaluation data and the teacher label is transmitted to the server device. Download from.

According to the present invention, an appropriate model can be downloaded.

FIG. 1A is a diagram showing a mounting example of a control device. FIG. 1B is a diagram showing an outline of a control system. FIG. 2 is a block diagram of the control system. FIG. 3 is a diagram showing a specific example of the model DB. FIG. 4 is a diagram showing a specific example of the applicable range. FIG. 5 is a diagram showing a specific example of the evaluation DB. FIG. 6 is a diagram showing an example of the conditions of the evaluation data. FIG. 7 is a schematic diagram showing an example of the transition of the evaluation value. FIG. 8 is a flowchart showing a processing procedure executed by the control device. FIG. 9 is a flowchart showing a processing procedure executed by the server device.

Hereinafter, the control device, the control system, and the control method according to the embodiment will be described with reference to the accompanying drawings. The present invention is not limited to the embodiments shown below.

First, the outline of the control system and the control method according to the embodiment will be described with reference to FIGS. 1A and 1B. FIG. 1A is a diagram showing a mounting example of a control device. FIG. 1B is a diagram showing an outline of a control system.

As shown in FIG. 1A, the control device 1 according to the embodiment is mounted on the vehicle 100. The control device 1 generates output data Do by inputting input data Di based on sensor information input from the vehicle-mounted sensor of the vehicle 100 into the model M. Here, the model M has layers called an input layer, an intermediate layer (hidden layer), and an output layer, and transmits the input data Di input from the input layer to the output layer while propagating each connection path. ..

At this time, the output data Do corresponding to the input data Di is generated by executing the arithmetic processing based on the parameters set in each connection path. The parameters include weight parameters, bias parameters, and the like.

By the way, as described above, the control device 1 performs arithmetic processing using the model M based on the sensor information input from the in-vehicle sensor. Here, the in-vehicle sensor may have an installation error for each vehicle 100, or the parts of the vehicle 100 may deteriorate.

Therefore, it is necessary to optimize the model M according to the current state of the vehicle 100. Here, for example, it is conceivable to optimize the model by downloading a new model from the server.

In this case, if the model to be downloaded is selected incorrectly, the reliability of the model after download may be lower than the reliability of the model before download. Therefore, when downloading a model, it is necessary to select a model that suits the current situation.

Therefore, in the control method according to the embodiment, it is decided to provide the data for selecting the update model to the server device that provides the model to the control device 1.

Specifically, as shown in FIG. 1B, in the control system S according to the embodiment, the control device 1 evaluates the model M. Here, the control device 1 can evaluate the model M by generating a predetermined state (for example, knocking of the engine) and inputting the sensor information at that time into the model M as evaluation data.

That is, if it is shown that the output data Do obtained by inputting the evaluation data into the model M is knocking, it should be shown that the model M generates the correct output data Do and the output data Do is not knocking. For example, it means that the model M generated erroneous output data Do.

The control device 1 inputs evaluation data of a plurality of conditions to the model M, and calculates an evaluation value of the model M based on the output data Do (step S1). Then, when the evaluation value is equal to or less than the threshold value, that is, when it is determined that the model M does not correspond to the current state, the control device 1 transmits the selection data to the server device 50 (step S2).

Here, the selection data is data for selecting an update model newly provided by the server device 50 to the control device 1, and is a data set of the above evaluation data and a teacher label of the evaluation data. Consists of.

Subsequently, the server device 50 selects the update model based on the acquired selection data (step S3). For example, the server device 50 can select the update model by inputting the evaluation data into each update model and comparing the obtained evaluation output data with the teacher label of the evaluation data.

In other words, the more the evaluation output data and the teacher label match, the more the updated model is the current model. The server device 50 selects the update model having the highest matching rate between the evaluation output data and the teacher label. Then, the control device 1 downloads the update model selected by the server device 50 (step S4).

As a result, the control device 1 can update the model M to the current update model of the vehicle 100. That is, it is possible to switch the model M to an update model in which a certain degree of accuracy is guaranteed.

As described above, in the control method according to the embodiment, the evaluation data and the teacher label are transmitted to the server device 50. As a result, the server device 50 can select an updated model suitable for the current state of the vehicle 100.

Therefore, according to the control method according to the embodiment, it is possible to download an appropriate model.

Next, a configuration example of the control system S according to the embodiment will be described with reference to FIG. FIG. 2 is a block diagram of the control system S. Note that FIG. 2 also shows the in-vehicle sensors 101.

The in-vehicle sensors 101 are, for example, sensors that detect the state of the internal combustion engine (engine) of the vehicle 100. The control device 1 can determine a state such as an engine misfire by inputting the sensor information input from the in-vehicle sensors 101 into the model M.

As shown in FIG. 2, the control device 1 includes a storage unit 2 and a control unit 3. The storage unit 2 is realized by, for example, a semiconductor memory element such as a RAM or a flash memory, or a storage device such as a hard disk or an optical disk.

Further, as shown in FIG. 2, the storage unit 2 includes a model DB 21 and an evaluation data DB 22. The model DB 21 is a database that stores the model M. In this embodiment, it is also possible to store a plurality of models M in the model DB 21.

FIG. 3 is a diagram showing a specific example of the model DB 21. As shown in FIG. 3, in the model DB 21, "model ID", "applicable range", "model" and the like are stored in association with each other.

The "model ID" indicates an identifier that identifies the model M. “Applicable range” indicates the applicable range of the corresponding model M. A specific example of the applicable range will be described later with reference to FIG. “Model” indicates a model M main body.

Next, a specific example of the applicable range will be described with reference to FIG. FIG. 4 is a diagram showing a specific example of the applicable range. As shown in FIG. 4, the applicable range is composed of a first condition (engine load factor) and a second condition (engine speed).

In the example of FIG. 4, it is shown that the first conditions of the applicable ranges R1 to R3 are different ranges. That is, in the present embodiment, the accuracy of the output data Do can be ensured by appropriately using the plurality of models M according to the applicable range. Although FIG. 4 shows a case where the total number of applicable ranges is three, the number of applicable ranges is changed according to the applicable range of each model M.

Returning to the description of FIG. 2, the evaluation data DB 22 will be described. The evaluation data DB 22 is a database that stores the above-mentioned evaluation data, teacher labels, and the like. FIG. 5 is a diagram showing a specific example of the evaluation data DB 22.

As shown in FIG. 5, in the evaluation data DB 22, "model ID", "data body", "evaluation value" and the like are stored in association with each other. As described above, the model ID is an identifier that identifies the model M.

The “data body” indicates the data body of the evaluation data, the output data Do for the evaluation data, and the teacher data. Further, in the example shown in FIG. 5, the case where the data main body is classified by "condition 1", "condition 2" ... "condition n" is shown.

The "evaluation value" is an evaluation value of the corresponding model M, and is calculated by the calculation unit 33 described later. Subsequently, a specific example of the conditions shown in FIG. 6 will be described. FIG. 6 is a diagram showing an example of the conditions of the evaluation data.

In the example shown in FIG. 6, the condition C indicates the range of the first condition and the second condition, and one square shown in FIG. 6 corresponds to one condition C. The range of each condition C can be arbitrarily changed.

Returning to the description of FIG. 2, the control unit 3 will be described. The control unit 3 is a controller, and is realized by, for example, a CPU, an MPU, or the like executing various programs stored in the storage unit 2 with the RAM as a work area. Further, the control unit 3 can be realized by, for example, an integrated circuit such as an ASIC or FPGA.

In the example shown in FIG. 2, the control unit 3 includes an acquisition unit 31, a calculation unit 32, a calculation unit 33, and a download unit 34. The acquisition unit 31 acquires sensor information from the in-vehicle sensors 101. The sensor information acquired by the acquisition unit 31 is notified to the calculation unit 32.

The calculation unit 32 generates output data Do by inputting the sensor information acquired by the acquisition unit 31 into the model M as input data Di. As described above, a plurality of models M may be stored in the model DB 21 depending on the applicable range of the model M.

In this case, the calculation unit 32 selects the model M according to the applicable range of the model M, and then inputs the input data Di. This makes it possible to appropriately obtain the output data Do according to the applicable range.

The calculation unit 33 calculates the evaluation value of the model M based on the output data Do obtained by inputting the evaluation data which is the input data for the evaluation into the model M. First, the calculation unit 33 instructs an engine control device (not shown) to generate a predetermined state (for example, knocking of the engine), so that the evaluation data is evaluated from the in-vehicle sensors 101 via the acquisition unit 31. To get.

Then, the calculation unit 33 inputs the evaluation data into the model M, and calculates the evaluation value of the model M based on the obtained output data Do and the teacher label of the evaluation data. The calculation unit 33 can sequentially calculate the evaluation value by repeating the above process instead of the condition C.

Here, the evaluation value is a match rate between the output data Do and the teacher label. That is, it can be expressed as evaluation value = number of correct answers ÷ number of tests. By setting the evaluation value as the matching rate in this way, the model M can be easily evaluated.

For example, the calculation unit 33 calculates the evaluation value when the number of tests reaches a predetermined number of times (for example, 20 times) as the current evaluation value. If the number of tests is 20 and the threshold value for the evaluation value is 90%, it is confirmed that the evaluation value does not exceed 90% when the answer is incorrect 3 times. Therefore, the calculation of the evaluation value may be stopped when it is determined that the evaluation value is equal to or less than the threshold value.

Further, the calculation unit 33 calculates the current evaluation value by using the evaluation data under the plurality of conditions C. In other words, the evaluation value of the model M is calculated under a plurality of conditions C. As a result, the model M can be evaluated using the evaluation data of various conditions C, so that the model M can be evaluated in a wide range. Note that, for example, only the conditions at which a plurality of conditions C are frequently generated may be used.

The calculation unit 33 resets the evaluation data and the evaluation value when the number of tests reaches a predetermined number of times (for example, 20 times) or when a predetermined period elapses. As a result, after the reset, the evaluation value can be newly calculated, so that the evaluation value suitable for the current situation can be calculated.

When the evaluation value calculated by the calculation unit 33 is equal to or less than the threshold value, the download unit 34 downloads an update model different from the model M to be evaluated from the server device 50.

First, the download unit 34 sets the evaluation data of each condition C and the teacher label of the evaluation data, and transmits the data to the server device 50. As a result, the server device 50 selects the update model based on the evaluation data and the teacher label.

Then, the download unit 34 downloads the update model from the server device 50 and updates the model DB 21 using the update model. Note that downloading the updated model includes both downloading all the data of the updated model and downloading the parameters of the model. That is, if the neural network structure is the same before and after the download, only the parameters may be downloaded.

Further, as will be described later, when the server device 50 selects a plurality of update models, the download unit 34 can download a plurality of update models for one model M.

Next, the transition of the evaluation value before and after the download will be described. FIG. 7 is a schematic diagram showing an example of the transition of the evaluation value. As shown in FIG. 7, the download unit 34 downloads the updated model at the time t1 when the evaluation value of the model M gradually decreases and the evaluation value reaches the threshold value Th. This makes it possible to use an updated model that is more suitable for the current situation than the model M. Therefore, the evaluation value is recovered after the time t1.

Returning to the description of FIG. 2, the server device 50 will be described. The server device 50 includes a storage unit 51 and a control unit 52. The storage unit 51 is realized by, for example, a semiconductor memory element such as a RAM or a flash memory, or a storage device such as a hard disk or an optical disk.

As shown in FIG. 2, the storage unit 51 stores the update model DB 51a. The update model DB 51a is a database that stores a plurality of update models. For example, the update model DB 51a stores various update models according to the state of the vehicle 100.

The control unit 52 is a controller, and is realized by, for example, the CPU, MPU, or the like executing various programs stored in the storage unit 2 with the RAM as a work area. Further, the control unit 3 can be realized by, for example, an integrated circuit such as an ASIC or FPGA.

As shown in FIG. 2, the control unit 52 includes a selection unit 52a and a learning unit 52b. The selection unit 52a selects the update model to be provided to the control device 1 from the update model DB 51a. First, when the selection unit 52a acquires the evaluation data and the teacher label from the control device 1, the selection unit 52a inputs the evaluation data to each update model of the update model DB 51a and obtains the output data Do.

Subsequently, the selection unit 52a compares the output data Do with the teacher label, and selects an update model having a high matching rate between the output data Do and the teacher label. Here, when one update model cannot cover the entire applicable range, the selection unit 52a can select a plurality of updated models according to the applicable range.

In this way, the selection unit 52a can expand the applicable range by selecting a plurality of update models as compared with the case where one update model is used. At this time, the selection unit 52a may determine the number of update models to be provided according to the memory capacity of the storage unit 2. The selection unit 52a can also provide one update model for a plurality of models.

The learning unit 52b generates an update model based on the evaluation data and the teacher label. Specifically, the learning unit 52b generates a new update model by performing machine learning using the evaluation data and the teacher label as teacher data. That is, the evaluation data and the teacher label can be used not only for selecting the update model but also for generating the update model.

Next, the processing procedure executed by the control device 1 and the server device 50 according to the embodiment will be described with reference to FIGS. 8 and 9. FIG. 8 is a flowchart showing a processing procedure executed by the control device 1.

As shown in FIG. 8, when the control device 1 first acquires the evaluation data (step S101), the control device 1 calculates the evaluation value based on the output data Do obtained by inputting the evaluation data into the model M (step S101). S102).

Subsequently, the control device 1 determines whether or not the evaluation value is larger than the threshold value Th (step S103), and if the evaluation value is larger than the threshold value Th (step S103, Yes), the process proceeds to the process of step S101. ..

Further, in the determination process of step S103, when the evaluation value is equal to or less than the threshold value Th (steps S103, No), the control device 1 transmits the evaluation data and the teacher label to the server device 50 (step S104).

After that, the control device 1 acquires the update model from the server device 50 (step S105), updates the model DB 21 based on the update model (step S106), and ends the process.

Next, the processing procedure of the server device 50 will be described with reference to FIG. FIG. 9 is a flowchart showing a processing procedure of the server device 50. As shown in FIG. 9, first, the server device 50 determines whether or not an update request has been acquired from the control device 1 (step S201).

When the server device 50 acquires the update request in the determination process of step S201 (step S201, Yes), the server device 50 selects the update model for each condition C (step S202). Subsequently, the server device 50 determines whether or not there is one selected update model (step S203), and if there is one update model (step S203, Yes), the server device 50 selects one selected update model. Provided (step S204), and the process ends.

Further, when the server device 50 has selected a plurality of update models in the determination process of step S203 (steps S203, No), the server device 50 provides a plurality of update models (step S205) and ends the process. If the update request has not been acquired in the determination process of step S201 (steps S201, No), the server device 50 ends the process.

As described above, the control device 1 according to the embodiment includes a calculation unit 33 and a download unit 34. The calculation unit 33 calculates the evaluation value of the model M based on the output data Do obtained by inputting the evaluation data, which is the input data for evaluation, into the model M generated by machine learning.

When the evaluation value calculated by the calculation unit 33 is equal to or less than the threshold value, the download unit 34 downloads an update model different from the model M from the server device 50. Further, the download unit 34 transmits the evaluation data and the teacher label of the evaluation data to the server device 50, and downloads the update model selected based on the evaluation data and the teacher label from the server device 50. Therefore, according to the control device 1 according to the embodiment, an appropriate model M can be downloaded.

By the way, in the above-described embodiment, the case where the control device 1 is mounted on the vehicle 100 has been described, but the present invention is not limited to this. That is, the present invention can be applied to various devices such as personal computers.

Further effects and variations can be easily derived by those skilled in the art. For this reason, the broader aspects of the invention are not limited to the particular details and representative embodiments expressed and described as described above. Therefore, various modifications can be made without departing from the spirit or scope of the general concept of the invention as defined by the appended claims and their equivalents.

1 Control device 31 Acquisition unit 32 Calculation unit 33 Calculation unit 34 Download unit 50 Server device M model S Control system

Claims (8)

A calculation unit that calculates the evaluation value of the model based on the output data obtained by inputting the evaluation data, which is the input data for evaluation, into the model generated by machine learning.
It is provided with a download unit that downloads an update model different from the model from the server device when the evaluation value calculated by the calculation unit is equal to or less than the threshold value.
The download section
The feature is that the evaluation data and the teacher label of the evaluation data are transmitted to the server device, and the update model selected based on the evaluation data and the teacher label is downloaded from the server device. Control device. The download section
The control device according to claim 1, wherein a plurality of the updated models are downloaded with respect to the model of 1. The calculation unit
The control device according to claim 1 or 2, wherein the matching rate between the output data and the teacher label is calculated as the evaluation value. The download section
Based on the evaluation value when the number of data for evaluation reaches a predetermined number, it is determined whether or not the model needs to be downloaded.
The calculation unit
The control device according to claim 1, 2 or 3, wherein the evaluation value and the evaluation data are reset at predetermined intervals. The calculation unit
The control device according to any one of claims 1 to 4, wherein the evaluation value is calculated using the evaluation data having different conditions. The control device according to any one of claims 1 to 5, further comprising an acquisition unit that acquires the evaluation data from an in-vehicle sensor that detects the state of the vehicle. The control device according to any one of claims 1 to 6.
A control system including a server device that provides the update model to the control device based on data transmitted from the control device. A calculation process for calculating the evaluation value of the model based on the output data obtained by inputting the evaluation data, which is the input data for evaluation, into the model generated by machine learning.
Including a download step of downloading an update model different from the model from the server device when the evaluation value calculated by the calculation step is equal to or less than the threshold value.
The download process is
The feature is that the evaluation data and the teacher label of the evaluation data are transmitted to the server device, and the update model selected based on the evaluation data and the teacher label is downloaded from the server device. Control method to do.

Images