JP7337510B2 - Output device, learning model generation method, computer program, and storage medium - Google Patents

Description

The present invention relates to an information output device, a learning model generation method, a computer program, and a storage medium for proposing an appropriate mounting state for a component of a manpowered vehicle.

There are human-powered vehicles that use human power at least partially, such as bicycles, electrically assisted bicycles, and electric bicycles called E-bikes. A human powered vehicle comprises multiple components such as a front derailleur, a rear derailleur, a seat post, or a suspension.

There is known a technique of connecting a diagnostic device to a plurality of components and diagnosing whether or not there is a failure (see Patent Documents 1 and 2). A conventional diagnostic device communicates with a component and diagnoses a failure based on whether or not the operation is completed when the component is instructed to perform a predetermined operation.

U.S. Pat. No. 7,819,032 U.S. Pat. No. 9,227,697

Even if there is no failure of individual components, the combination may not be appropriate. Depending on the relationship with other components and the personality of the rider, there are cases where a more appropriate assembly state should be proposed. Appropriate assembly states based on multiple conditions should be proposed by comprehensively judging the states of each of the plurality of components.

An object of the present invention is to provide an output device, a learning model generation method, a computer program, and a storage medium for outputting information for proposing an appropriate mounting state of parts including components of a manpowered vehicle.

(1) The output device according to the first aspect of the present invention uses a learning model learned by a learning algorithm to output output information regarding the mounting state of parts of a human-powered vehicle according to input information regarding the mounting state. have a department.
Therefore, the learning model outputs output information corresponding to various information regarding the mounting state, and it is possible to propose an appropriate mounting state of the part based on the output information.

(2) In the output device according to the second aspect of the present invention, the input information includes identification information of the component.
Therefore, output information corresponding to the combination of parts to be attached to the manpowered vehicle is output from the learning model, and it is possible to propose an appropriate attachment state of the parts based on the output information.

(3) In the output device according to the third aspect of the present invention, the input information includes information regarding the vehicle body of the human-powered vehicle.
Therefore, output information corresponding to information such as the size or type of the vehicle body of the manpowered vehicle is output from the learning model, and it is possible to propose an appropriate mounting state of parts based on the output information.

(4) In the output device according to the fourth aspect of the present invention, the input information includes physical information of the user of the manpowered vehicle.
Therefore, with respect to the parts of the manpowered vehicle, it is possible to propose an appropriate mounting state of the parts according to the weight, height, leg length, or the like of the user who rides the manpowered vehicle.

(5) In the output device according to the fifth aspect of the present invention, the input information includes at least one of still image data of the manpowered vehicle and moving image data of the manpowered vehicle.
Therefore, regarding parts of a manpowered vehicle, it is possible to propose an appropriate mounting state of parts according to the observation result of the whole manpowered vehicle by inputting still image data or moving image data of the manpowered vehicle.

(6) In the output device according to the sixth aspect of the present invention, the input information includes data regarding vibrations emitted from the manpowered vehicle.
For this reason, it is possible to propose an appropriate mounting state of the parts of the manpowered vehicle based on the vibration during driving of the manpowered vehicle.

(7) In the output device according to the seventh aspect of the present invention, the vibration-related data includes audio data.
For this reason, regarding the parts of a human-powered vehicle, it is possible to propose the appropriate mounting state of the parts according to various generated sounds including abnormal sounds by inputting sound data collected while the human-powered vehicle is driving. .

(8) In the output device according to the eighth aspect of the present invention, the output information includes suitability information regarding suitability of the mounting state.
Therefore, with respect to the parts of the manpowered vehicle, the appropriateness of the mounting state of the parts is output from the learning model according to various input information, and it is possible to present the appropriateness of the mounting state of the parts.

(9) In the output device according to the ninth aspect of the present invention, the output information includes proposal information regarding proposals for the mounting state.
For this reason, with respect to the parts of the human-powered vehicle, since the learning model outputs proposals for other appropriate mounting states for the mounting states of the parts corresponding to various input information, the appropriate mounting states of the parts can be determined from the output information. can be proposed.

(10) In the output device according to the tenth aspect of the present invention, the proposal information includes at least one of information on how to install the part, information on adjustment of the part, and information on replacement of the part. .
Therefore, with respect to the parts of the manpowered vehicle, it is possible to propose an appropriate mounting method, adjustment amount, adjustment method, and replacement of the parts according to the mounting state of the parts corresponding to various input information.

(11) A method of generating a learning model according to the eleventh aspect of the present invention uses a neural network configured to output output information regarding the mounting state of parts of a manpowered vehicle in response to input information regarding the mounting state of the parts. In the learning model generation method, a computer adjusts parameters in the intermediate layer of the neural network in accordance with teacher data labeled with evaluations of the input information.
For this reason, a learning model is generated based on various input information regarding the mounting state of the parts of the manpowered vehicle and known teacher data corresponding to appropriate mounting states corresponding to the input information. By using the output information output from the learning model, it is possible to propose an appropriate mounting state of the part.

(12) The method of generating a learning model according to the twelfth aspect of the present invention generates the learning model based on additional teacher data labeled with evaluations of output information outputted by giving the input information to the neural network. Update by computer.
Therefore, the neural network is sequentially updated based on the evaluation of the output information output from the neural network that is learning about the parts of the manpowered vehicle, and the proposal of the mounting state of the parts becomes more appropriate.

(13) A method of generating a learning model according to the thirteenth aspect of the present invention uses a neural network configured to output output information regarding the mounting state of parts of a manpowered vehicle in response to input information regarding the mounting state of the parts. A method of generating a learning model, wherein a learning model is generated by applying the input information to the neural network in an intermediate layer of the neural network according to teacher data labeled with an evaluation of the output information output Adjust parameters by computer.
Therefore, the neural network is sequentially updated based on the evaluation of the output information output from the neural network during learning without using teaching data in advance, and the proposal of the mounting state of the part becomes more appropriate.

(14) The computer program according to the fourteenth aspect of the present invention provides a learning model trained by a learning algorithm to output output information regarding the mounting state in response to input information regarding the mounting state of parts of a manpowered vehicle, Given the acquired input information, based on the output information output from the learning model, information on the mounting state of parts in the manpowered vehicle is superimposed and arranged on the image on a display unit that displays the image of the manpowered vehicle. Make the computer execute the processing to be displayed.
For this reason, output information for proposing an appropriate mounting state based on various input information regarding the mounting state of the parts of the human-powered vehicle is output from the learning model according to various conditions such as vehicle body information and rider information. be done. Based on the information output from the learning model, an image showing the appropriate mounting state of parts corresponding to the image of the human-powered vehicle is displayed, making it possible to make visually easy-to-understand proposals.

(15) A storage medium according to the fifteenth aspect of the present invention stores the computer program of (14).
For this reason, based on the computer program read by the computer from the storage medium, the computer displays an image showing the appropriate mounting state of the parts corresponding to the image of the manpowered vehicle based on the information output from the learning model. , a visually easy-to-understand proposal is possible.

According to the output device for outputting information about the mounting state of parts of a manpowered vehicle according to the present invention, it is possible to output information for proposing an appropriate mounting state of parts including components of a manpowered vehicle.

It is a block diagram which shows the structure of the output device of 1st Embodiment, and a terminal device. 4 is a flow chart showing an example of a method of generating a learning model; 10 is a flow chart showing another example of a method of generating a learning model; It is a figure which shows the outline|summary of a learning model. 7 is a flow chart showing an example of a procedure for proposing an installation state; It is a figure which shows the example of a screen displayed on a terminal device. It is a figure which shows the example of a screen displayed on a terminal device. It is a block diagram which shows the structure of the output device of 2nd Embodiment, and a terminal device. 9 is a flow chart showing an example of a learning model generation method according to the second embodiment. It is a figure which shows the outline|summary of the learning model of 2nd Embodiment. 7 is a flow chart showing an example of a procedure for proposing an installation state; 7 is a flow chart showing an example of a procedure for proposing an installation state; It is a figure which shows the example of a screen displayed on the terminal device of 2nd Embodiment. It is a figure which shows the example of a screen displayed on the terminal device of 2nd Embodiment. FIG. 11 is a block diagram showing the configuration of a terminal device according to the third embodiment; FIG. It is a side view of a human-powered vehicle to which the output device of 4th Embodiment is applied. It is a block diagram which shows the structure of the output device in 4th Embodiment. FIG. 14 is a flow chart showing an example of a method of generating a learning model in the fourth embodiment; FIG. It is a figure which shows the outline|summary of the learning model in 4th Embodiment. FIG. 14 is a flow chart showing an example of a processing procedure for judging whether the mounting state is appropriate in the fourth embodiment; FIG.

The following descriptions of the respective embodiments are examples of forms that the output device of the present invention can take, and are not intended to limit the forms. An output device related to the present invention can take a form different from each embodiment, such as a modification of each embodiment, or a form in which at least two mutually consistent modifications are combined.

In the description of each embodiment below, terms such as forward, rearward, forward, rearward, left, right, sideways, up, and down refer to directions when the user is seated on the saddle of the manpowered vehicle. Used as a reference.

(First embodiment)
In the first embodiment, an output device 1 managed by a parts manufacturer of the manpowered vehicle and a terminal used for maintenance of the manpowered vehicle based on information on the mounting state of the parts of the manpowered vehicle provided from the output device 1. A maintenance method using the device 2 will be described.

FIG. 1 is a block diagram showing configurations of an output device 1 and a terminal device 2 according to the first embodiment. In the first embodiment, the output device 1 uses a server computer. The output device 1 may be a personal computer. The output device 1 includes a processing unit 10 , a storage unit 12 and a communication unit 14 . In the first embodiment, the output device 1 is described below as one server computer, but the functions or processes may be distributed among a plurality of server computers, or the output device 1 may be virtually generated in one large computer. It may be one of multiple instances.

The processing unit 10 is a processor using a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit). The processing unit 10 uses built-in memories such as ROM (Read Only Memory) and RAM (Random Access Memory) to control each component and execute processing. The processing unit 10 generates the learning model 1M based on the learning program 1P including the learning model 1M stored in the storage unit 12, and uses the generated learning model 1M to transmit information regarding the assembly of the human-powered vehicle to the terminal device. Output to 2.

The storage unit 12 uses a hard disk or an SSD (Solid State Drive) and stores programs and data referred to by the processing unit 10 including the learning program 1P. The storage unit 12 stores a learning model 1M generated based on the learning program 1P. The storage unit 12 stores data of manuals explaining how to install parts handled by parts manufacturers. The learning program 1P and the learning model 1M may be obtained by reading the learning program 30P and the learning model 30M stored in the storage medium 3 and duplicating them in the storage unit 12 .

The communication unit 14 is a communication module that realizes communication connection with the terminal device 2 via the network N including the public communication network N1. The communication unit 14 is, for example, a wireless communication device compatible with Wi-Fi. In the second example, the communication unit 14 is a carrier communication module that realizes communication connection via a carrier network N2 included in the network N. FIG. The communication unit 14 is a network card for wired connection in the third example.

The terminal device 2 is a communication terminal such as a tablet owned by a user of the manpowered vehicle performing maintenance or a dealer of the manpowered vehicle. The terminal device 2 may be a so-called smart phone. The terminal device 2 may be a laptop personal computer. The terminal device 2 includes a processing section 20 , a storage section 22 , a communication section 24 , a display section 26 and an operation section 28 .

The processing unit 20 is a processor using a CPU or GPU. The processing unit 20 uses internal memories such as ROM and RAM to control each component and execute processing. The processing unit 20 may be configured as one piece of hardware, SoC (System On a Chip), in which the processor, memory, storage unit 22 and communication unit 24 are integrated. Based on the maintenance application program 2P stored in the storage unit 22, the processing unit 20 displays information regarding the assembly of the human-powered vehicle on the display unit 26. FIG.

The storage unit 22 uses a flash memory and stores programs and data referred to by the processing unit 20 including the maintenance application program 2P. The maintenance application program 2P stored in the storage unit 22 is provided by a parts manufacturer of the human-powered vehicle, distributed from an arbitrary distribution server, and installed in the terminal device 2, which is a general-purpose computer. The maintenance application program 2P may be obtained by reading the maintenance application program 50P stored in the storage medium 5 and duplicating it in the storage unit 22 .

The communication unit 24 is a communication module that realizes communication connection to the public communication network N1. The communication unit 24 is, for example, a wireless communication device compatible with Wi-Fi. In the second example, the communication unit 24 is a carrier communication module that realizes communication via the carrier network N2. The communication unit 24 is a network card in the third example.

The display unit 26 uses a display device such as a liquid crystal panel or an organic EL display. The display unit 26 displays information about the mounting state of the parts in the manpowered vehicle together with the image of the manpowered vehicle.

The operation unit 28 is an interface that receives user operations, and uses physical buttons, a touch panel device with a built-in display, a speaker, a microphone, and the like. The operation unit 28 may accept operations on the screen displayed on the display unit 26 using physical buttons or a touch panel, or recognize the operation content from the input voice with the microphone and interact with the voice output from the speaker. You may accept the operation in the form.

The output device 1 of the first embodiment having the configuration described above uses a learning model 1M learned by a learning algorithm to be described later to determine the mounting state of the parts according to the input information regarding the mounting state of the parts of the human-powered vehicle. Print output information about . The output information is transmitted to the terminal device 2, and the terminal device 2 receives it based on the maintenance application program 2P, and displays information on the mounting state of the parts in the human-powered vehicle on the display unit 26 based on the received output information. indicate.

First, the learning model 1M generated by the output device 1 will be described. The learning model 1M is generated in advance by the output device 1 or an external device. FIG. 2 is a flow chart showing an example of a method of generating the learning model 1M. The output device 1 generates a learning model 1M that outputs output information in response to input information regarding the mounting state of parts of the human-powered vehicle by a supervised deep learning algorithm using a neural network (hereinafter referred to as NN). The learning algorithm may be an unsupervised learning algorithm or a recurrent neural network. The learning algorithm may be reinforcement learning.

The processing unit 10 collects information to be teacher data and stores it in the storage unit 12 before starting generation. The training data is a group of information in which information about the mounting state of the parts of the manpowered vehicle and judgment results as to whether the mounting state is appropriate or not are given. These groups of information are collected by manufacturers of manpowered vehicles or parts, and are information of manpowered vehicles in installed states determined to be appropriate or inappropriate. Information clusters are collected at a human powered vehicle dealership, assembly plant or proving ground. Assembling engineers, manufacturer engineers, etc. are responsible for judging suitability. The subject of determination may be the user. The input information is, for example, part identification information, such as the manufacturer name, model number, part number, and part name. The input information includes vehicle body information. The vehicle body information includes the size, type, and part number of each frame, front fork, handle, stem, and seat post. The input information includes the physical information of the user riding in the manpowered vehicle corresponding to the part identification information and the vehicle body information. The physical information includes at least one of the user's weight, height, leg length, arm length, and the like.

The processing unit 10 executes the following processes using the collected information group.
The processing unit 10 acquires identification information of parts to be attached to the manpowered vehicle, regarding the mounting state of the parts of the manpowered vehicle stored in the storage unit 12 (step S101). The processing unit 10 acquires information about the body of the human-powered vehicle to which the one or more parts acquired in step S101 are attached (step S103). The processing unit acquires the physical information of the user of the manpowered vehicle to which the one or more parts acquired in step S101 are attached (step S105).

The processing unit 10 creates teacher data by assigning a label indicating the judgment result of suitability given in advance to the input information acquired in steps S101 to S105 (step S107).

The processing unit 10 determines whether or not teacher data has been created from all information groups that satisfy a predetermined condition among the information groups stored in the storage unit 12 (step S109). The predetermined condition is, for example, the number of pieces of information used as teacher data for obtaining input information for the same mounting state. If it is determined in step S109 that the teacher data has not been created from all information groups (S109: NO), the processing unit 10 returns the process to step S101.

If it is determined in step S109 that the teacher data has been created from all information groups (S109: YES), the processor 10 provides the created teacher data to the NN (step S111). The processing unit 10 learns parameters including at least one of weights and biases in the intermediate layers of the NN from the given teacher data (step S113), and ends the process.

As a result, the learning model 1M is generated by adjusting the parameters in the intermediate layer of the NN according to the teacher data pre-labeled for the evaluation of the input information by a computer.

Generation of the learning model 1M is not limited to the labeled supervised learning method described above. FIG. 3 is a flow chart showing another example of the method of generating the learning model 1M. Among the processing procedures shown in the flowchart of FIG. 3, the procedures common to the flowchart of FIG. 2 are given the same step numbers, and detailed descriptions thereof are omitted. In the flowchart of FIG. 3, the processing unit 10 executes processing based on information regarding one mounting state or one manpowered vehicle. The processing unit 10 performs the following processes for each of the various manpowered vehicles or the various mounting conditions.

The processing unit 10 gives the input information acquired in steps S103 to S105 for one mounting state to the learning NN (step S121). The processing unit 10 specifies the information, for example, the score, regarding the suitability of the mounting state as a whole, which is output from the NN when the input information is given to the NN (step S123). The processing unit 10 compares the specified suitability information with the judgment result given in advance to derive an evaluation value (step S125). The evaluation value is, for example, the difference between the score specified in step S123 and the score corresponding to the determination result. Error parameters are, for example, angles, distances, dimensions, orientations, or weights.

The processing unit 10 gives the derived evaluation value to the intermediate layer to update the parameters in the intermediate layer (step S127), and ends the process. In step S127, the parameters are updated to minimize the derived error. In step S127, the learning model 1M is generated by adjusting the evaluation teacher data for the output information output when the input information obtained in steps S101 to S105 is input to the NN.

The processing unit 10 may generate the learning model 1M by either the processing procedure shown in the flowchart of FIG. 2 or the processing procedure shown in the flowchart of FIG. good. When executing any of the processing procedures, the processing unit 10, for example, initially generates the learning model 1M according to the procedure shown in the flowchart of FIG. It is preferable to adjust and re-update according to the processing procedure shown in the flow chart.

FIG. 4 is a diagram showing an overview of the learning model 1M. The learning model 1M includes an input layer 31 for inputting input information about parts of the manpowered vehicle, and an output layer 32 for outputting output information about the adequacy of the mounting state of the parts of the manpowered vehicle. The learning model 1M includes an intermediate layer 33 having parameters that have been learned using teacher data so as to output suitability when input information is input. When input information about a part is input to the input layer 31 , computation is performed in the intermediate layer 33 using learned parameters, and output information about the propriety of the input part is output from the output layer 32 .

In the specific example of FIG. 4, the input layer 31 is provided with identification information for each of different types of parts regarding the adequacy of the mounting state of the parts. In the example of FIG. 4, the identification information of the part is given the identification information about the electric component. A front derailleur, a rear derailleur, and a disk rotor are illustrated.

The input layer 31 is provided with vehicle body information. The vehicle body information includes the size, type, and part number of each of the frame, front fork, handle, stem, and seat post, as described above.

The input layer 31 is given physical information of the user. Physical information includes height and weight. Physical information may also be leg length, hand length, or gender.

The specific example of FIG. 4 is a configuration in which all of identification information of parts, vehicle body information, and physical information are given to the input layer 31 . However, the processing in the output device 1 is not limited to the configuration that provides everything. Part identification information, vehicle body information, and part of physical information may be used. The information to be input may be distinguished according to the object of suitability. For example, in the case of the learning model 1M that outputs information on whether or not the rear derailleur is installed properly, the type and length of the frame and the model number of the front derailleur may be input.

The learning model 1M generated in this manner is used in the following process of proposing the mounting state of parts. FIG. 5 is a flow chart showing an example of a processing procedure for proposing an installation state. Proposal of the installation state is started when the terminal device 2 used by the user executes the maintenance application program 2P and starts maintenance.

The processing unit 20 of the terminal device 2 receives an input on the user's information input screen about the manpower-driven vehicle based on the maintenance application program 2P (step S201). Based on the input to the input screen, the processing unit 20 receives the identification information of the parts of the manpowered vehicle to be maintained or assembled from the manpowered vehicle, the vehicle body information, and the physical information of the user ( step S203).

The processing unit 20 transmits a determination request including the received part identification information, vehicle body information, and user physical information from the communication unit 24 to the output device 1 (step S205).

The output device 1 receives the determination request through the communication unit 14 (step S131), and the processing unit 10 acquires the identification information of the parts attached to the human-powered vehicle from the received determination request (step S133). The processing unit 10 acquires information about the body of the human-powered vehicle from the determination request (step S135). The vehicle body information is either the size, type, or product number of the frame, front fork, handle, stem, and seat post of the human-powered vehicle. The processing unit 10 acquires the physical information of the user of the human-powered vehicle from the determination request (step S137). The physical information in step S137 includes at least one of the user's weight, height, leg length, arm length, and the like.

Only one of steps S135 and S137 may be performed. That is, the information to be input to the learning model 1M may be both the vehicle body information and the body information, or may be only one of them.

The processing unit 10 of the output device 1 gives the input information acquired in steps S133, S135, and S137 to the learning model 1M (step S139). The processing unit 10 identifies information, for example, a score, indicating whether the attachment state as a whole is output from the learning model 1M when the input information is given to the learning model 1M (step S141).

Based on the identification information of the part acquired in step S133, the processing unit 10 reads the manual data corresponding to the target part from the storage unit 12 (step S143). The manual data related to the vehicle body to be processed is read out from the storage unit 12 (step S145). The processing unit 10 may read the instruction manual data of parts recommended for each body type based on the physical information.

The processing unit 10 transmits the suitability information specified in step S141 and the instruction manual data read out in steps S143 and S145 from the communication unit 14 to the terminal device 2 that requested the determination (step S147).

The terminal device 2 receives the information on suitability and the manual data from the communication unit 24 (step S207), and the processing unit 20 displays the received information on the suitability on the display unit 26 (step S209). The processing unit 20 determines suitability based on the received information (step S211), and if it is determined to be suitable (S211: YES), the process ends.

If it is determined to be inappropriate in step S211 (S211: NO), the processing unit 20 causes the display unit 26 to display a sample of how to install the parts based on the instruction manual data received in step S207 (step S213), and performs processing. exit.

The processing unit 20 may display a sample of how to install the component on the display unit 26 regardless of the determination in step S211.

6 and 7 are diagrams showing examples of screens displayed on the terminal device 2. FIG. When the user selects and activates the maintenance application program 2P using the terminal device 2, the processing unit 20 displays an input screen 261 for accepting input information regarding the mounting state of the parts based on the maintenance application program 2P. 26. The input screen 261 includes, as shown in FIG. 6, a selection screen 263 of identification information of parts for each component, a selection screen 265 of vehicle body information, and a selection screen 267 of physical information. The input screen 261 includes a send button 269 for sending a request for judgment of suitability. When the send button 269 is selected, the processing unit 20 sends a determination request including the information received on the selection screens 263, 265, and 267 from the communication unit 24 (S205). A free input field may be used instead of the selection screen 267 .

FIG. 7 shows an example of a proposal screen 271 that displays information on suitability corresponding to a judgment request. As shown in FIG. 7, the proposal screen 271 includes a link 273 to instruction manual data on how to install together with the result of suitability. When the link 273 is selected, the instruction manual, that is, a proposal for an appropriate mounting method, is displayed on the display unit 26 . Further, as shown in FIG. 7, the proposal screen 271 shows the recommended parts together with the result of suitability, and preferably includes information on replacement of the parts.

As described above, the learned learning model 1M is assumed to be used as a program module of the maintenance application program 20P, which is part of the artificial intelligence software. The learning model 1M of the present disclosure is used in cooperative processing with the terminal device 2 in the output device 1 including the processing unit 10 and the storage unit 12, as described above.

In this way, based on the learning model 1M, various information about the mounting state, information for identifying parts, information about the type or size of the vehicle body, and output information according to the body of the user who is the rider are output. It is possible to propose an appropriate mounting state of parts based on this.

(Second embodiment)
FIG. 8 is a block diagram showing configurations of the output device 1 and the terminal device 2 of the second embodiment. The output device 1 according to the second embodiment stores the learning model 12M in the storage unit 12. FIG. The learning model 12M is learned for each component that is an object of suitability. The learning model 12M receives still image data or moving image data of an imaged human-powered vehicle as input information regarding the mounting state. The terminal device 2 includes an imaging unit 29 in addition to the processing unit 20 , the storage unit 22 , the communication unit 24 , the display unit 26 and the operation unit 28 . The imaging unit 29 is a device that includes an imaging element, captures a still image or a moving image based on an instruction from the processing unit 20, and outputs data obtained by imaging.

The configuration of the output device 1 and the terminal device 2, the method of generating the learning model 12M, the content of the learning model 12M, and the processing procedure related to the method of outputting information on the adequacy of the mounting state based on the learning model 12M are the same as those of the first embodiment. is. Therefore, among the configurations of the second embodiment, the configurations common to those of the first embodiment are denoted by the same reference numerals, and detailed description thereof will be omitted.

FIG. 9 is a flow chart showing an example of a method of generating the learning model 12M in the second embodiment. Also in the second embodiment, the output device 1 generates a learning model 12M that outputs output information about the mounting state of parts by a supervised deep learning algorithm using NN.

The processing unit 10 collects information that becomes teacher data and stores it in the storage unit 12 before starting generation. The training data is a group of information in which input information, which is still image data or moving image data obtained by picking up the mounting state of the parts of the manpowered vehicle, is added with the judgment result of the propriety of the mounting state being imaged. These groups of information are collected by manufacturers of manpowered vehicles or parts, and are information of manpowered vehicles in installed states determined to be appropriate or inappropriate. Still image data, or motion image data, is collected at a manpowered vehicle dealership, assembly plant, or proving ground. The still image data or the moving image data is classified according to the type of target parts. The input information may include, in addition to still image data or moving image data, component identification information, vehicle body information, or user physical information described in the first embodiment.

The processing unit 10 executes the following generation processing for each part. For example, the processing unit 10 performs the following processing using still image data or moving image data classified as images of the front derailleur as teacher data.

The processing unit 10 selects a target part with respect to the mounting state of the parts of the manpowered vehicle stored in the storage unit 12 (step S151). At least one of still image data and moving image data of the human-powered vehicle with the parts attached is acquired (step S153). The still image data or moving image data may be data obtained by imaging the entire human-powered vehicle, or may be data obtained by imaging an enlarged image of an object for determining whether the mounting state is appropriate.

The processing unit 10 acquires the identification information of the parts attached to the manpowered vehicle for which the still image or the moving image is captured (step S155). At step S155, the processing unit 10 acquires, for example, the product number or model name of the front derailleur. The processing unit 10 acquires vehicle body information (step S157) and acquires physical information of the user (step S159). All of the processes of steps S155, S157, and S159 may be omitted, or one or two of them may be executed.

The processing unit 10 creates teacher data by assigning a label indicating the judgment result of suitability given in advance to the input information acquired in steps S153 to S159 (step S161).

The processing unit 10 determines whether or not the teacher data has been created from all the information groups that satisfy a predetermined condition for the target part among the information groups stored in the storage unit 12 (step S163). The predetermined condition is, for example, an upper limit for the number of pieces of information to be used as teacher data regarding acquisition of input information for the same mounting state. If it is determined in step S163 that the teacher data has not been created from all the information groups (S163: NO), the processing unit 10 returns the process to step S153.

If it is determined in step S163 that the teacher data has been created from all information groups (S163: YES), the processor 10 provides the created teacher data to the NN (step S165). The processing unit 10 learns parameters including at least one of weights and biases in the intermediate layers of the NN from the given teacher data (step S167), and ends the process.

As a result, the learning model 12M is generated by adjusting the parameters in the intermediate layer of the NN according to the teacher data pre-labeled for the evaluation of the input information by a computer.

In the second embodiment, as shown in the flowchart of FIG. 3 of the first embodiment, the processing unit 10 inputs one piece of still image data or moving image data and outputs information about suitability output from the NN, A learning model 12M may be generated by error propagation by obtaining a difference from a known determination result. Flowcharts and detailed descriptions of corresponding processing procedures are omitted. The processing unit 10 may adjust the parameters of the intermediate layer by error propagation after the processing procedure shown in the flowchart of FIG. 9 .

FIG. 10 is a diagram showing an overview of the learning model 12M of the second embodiment. The learning model 12M includes an input layer 31 for inputting input information about parts of the manpowered vehicle, and an output layer 32 for outputting output information about the adequacy of the mounting state of the parts of the manpowered vehicle. The learning model 12M is provided with an intermediate layer 33 having parameters that have been learned using teacher data so as to output suitability when input information is input. When input information about a part is input to the input layer 31 , computation is performed in the intermediate layer 33 using learned parameters, and output information about the propriety of the input part is output from the output layer 32 .

In the specific example of FIG. 10, the input layer 31 is provided with still image data or moving image data obtained by picking up a human-powered vehicle regarding the propriety of the mounting state of parts. In the input layer 31, apart from still image data or moving image data, input information such as that shown in the first embodiment, such as identification information of parts, vehicle body information, and physical information of the user, or All may be given.

The specific example of FIG. 10 is a configuration in which still image data or moving image data, part identification information and vehicle body information are input. However, the processing in the output device 1 is not limited to the configuration that provides everything. Only still image data or moving image data may be used. As shown in FIG. 10, part information and vehicle body information may be input to the intermediate layer 33 after still image data or moving image data feature amounts are output, still image data, or The moving image data may be input to the input layer 31 after the feature amount is derived in advance using a convolutional neural network.

The input information of the learning model 12M may be classified according to the object of suitability. As described above, when generating the learning model 12M that outputs information on whether or not the front derailleur is installed properly, the still image data of the front derailleur and the identification information of the parts are input, and the disk rotor is used as the target. In some cases, a distinction may be made so as to input still image data of the disk rotor and vehicle body information.

In FIG. 10, the learning model 12M outputs, as output information, the probabilities of whether it is appropriate or, if inappropriate, what type of inappropriate state it is. The processing unit 10 identifies the appropriate state, the type A inappropriate state, and the type B inappropriate state based on the output probabilities of the appropriate state. The learning model 12M is not limited to this, and may output the probability of any one of a plurality of states including a suitable type G state and a suitable type E state.

The learning model 12M learned in this way outputs the probability of the adequacy of the mounting state of the parts shown in the image when new still image data or moving image data is input.

The learning model 12M is used in the process of proposing the mounting state of the following parts. 11 and 12 are flowcharts showing an example of a processing procedure for proposing an installation state. Proposal of the installation state is started when the terminal device 2 used by the user executes the maintenance application program 2P and starts maintenance.

Based on the maintenance application program 20P, the processing unit 20 of the terminal device 2 displays a screen that prompts the user to capture a still image or a moving image of the part to be judged as appropriate on the screen for inputting information about the human-powered vehicle. It is displayed on the unit 26 (step S221). The processing unit 20 accepts an operation to start imaging with the operation unit 28, and images a still image or a moving image of the manpowered vehicle with the imaging unit 29 (step S223). In step S<b>223 , the processing unit 20 may read data of a still image or a moving image captured in advance from the storage unit 12 .

Based on the input to the input screen, the processing unit 20 receives the identification information of the parts of the manpowered vehicle that is imaged or to be assembled from the manpowered vehicle and the vehicle body information (step S225).

The processing unit 20 transmits a judgment request including the captured still image or moving image data, the input identification information of the parts, and the vehicle body information from the communication unit 24 to the output device 1 (step S227). ). In step S227, the processing unit 20 designates information identifying the type of part to be imaged, and whether it is a front derailleur, a rear derailleur, or a disc rotor.

The output device 1 receives the determination request through the communication unit 14 (step S171), and the processing unit 10 selects the learning model 12M corresponding to the component type specified in the determination request (step S173).

The processing unit 10 acquires at least one of still image data and moving image data of the manpowered vehicle with the parts attached from the received determination request (step S175). The processing unit 10 acquires the identification information of the parts of the manpowered vehicle shown in the still image data or moving image data of step S173 from the determination request (step S177). The processing unit 10 acquires information about the body of the human-powered vehicle from the determination request (step S179).

The processing unit 10 provides the still image data or moving image data acquired in step S175, the identification information of the parts, and the information on the vehicle body to the learning model 12M selected in step S173 (step S181). The processing unit 10 identifies information about the propriety of the attachment state of the component output from the learning model 12M, for example, which one of the appropriate state, type A inappropriate state, and type B inappropriate state ( step S183).

Based on the specified state, the processing unit 10 reads out proposal information for establishing an appropriate state from the storage unit 22 (step S185), and acquires similar image data (step S187). In step S185, the proposal information includes, for each type of part, a message when an appropriate state is specified, a proposal content for making an appropriate state when an inappropriate state of type A is specified, and an appropriate state of type B. It is stored in the storage unit 12 as the content of the proposal when the inappropriate state is specified. The similar image data may be still image data or moving image data.

The processing unit 20 transmits the suitability information identified in step S183, the proposal information read out in step S185, and similar image data from the communication unit 14 to the terminal device 2 that requested the determination (step S189).

Acquisition of similar image data is not essential, and only the proposal information read in step S185 may be obtained. The proposal information may be a message that specifically indicates the angle for mounting the component, the distance from the reference point, and the like. When the proposal information is transmitted together with the image data, the proposal information may be only numerical values, and the numerical values indicating the distance or angle are preferably displayed together with the image data.

The terminal device 2 receives the suitability information, the proposal information, and the image data from the communication unit 24 (step S229), and the processing unit 20 displays the received suitability information on the display unit 26 (step S231). The processing unit 20 determines suitability based on the received information (step S233), and if it is determined to be suitable (S233: YES), the process ends.

If it is determined to be inappropriate in step S233 (S233: NO), the processing unit 20 superimposes the image of the similar image data received in step S229 on the still image captured in step S223, and arranges the received proposal information. It is displayed on the display unit 26 (step S235), and the process ends. If a moving image is captured in step S223, the processing unit 20 extracts any frame image from the moving image data and uses it as still image data.

Regardless of the determination in step S233, the processing unit 20 may display, on the display unit 26, an image based on the image data extracted as being similar and the proposal information as a sample of how to attach the part.

13 and 14 are diagrams showing examples of screens displayed on the terminal device 2 of the second embodiment. When the user selects and activates the maintenance application program 20P using the terminal device 2, the processing unit 20 determines the appropriateness of the mounting state of the component based on the maintenance application program 20P. An input screen 281 prompting the user to take an image is displayed on the display unit 26 (S221).

The input screen 261 includes a component selection screen 283 and an imaging button 285, as shown in FIG. The input screen 281 also includes a selection screen 287 for body information. The input screen 281 includes a send button 289 for sending a request for judgment of suitability. When the send button 289 is selected, the processing unit 20 sends a determination request including the information received on the selection screens 263, 265, 267 from the communication unit 24 (S227). Any item instead of the selection screens 283 and 287 may include a free entry field.

FIG. 14 shows an example of a proposal screen 291 that displays information on suitability corresponding to a judgment request. As shown in FIG. 14, on the proposal screen 291, along with the result of suitability, proposal information regarding the installation state of parts in the human-powered vehicle based on the output information output from the learning model 12M is displayed. Similar image data are displayed side by side, superimposed on the image.

In this way, regarding the mounting state based on the learning model 12M, suitability corresponding to the image data obtained by picking up the mounting state of the component is output, and it is possible to propose an appropriate mounting state of the component according to the observation result of the component. be.

(Third Embodiment)
FIG. 15 is a block diagram showing the configuration of the terminal device 2 according to the third embodiment. In both the first embodiment and the second embodiment, the information on the adequacy of the mounting state based on the learning model 1M or the learning model 12M is output from the output device 1 managed by the parts manufacturer and transmitted to the terminal device 2. Ta. In contrast, in the third embodiment, the maintenance application program 22P of the terminal device 2 used by the user implements processing based on the learning model 2M. Processing can be performed only by the terminal device 2 without communicating with the output device 1 . In the terminal device 2 according to the third embodiment, the learning model 2M is stored together with the maintenance application program 22P in the storage unit 22. Except for the processing based on the maintenance application program 22P, the terminal device 2 according to the first or second embodiment is the same as the terminal device according to the first or second embodiment. 2. Therefore, the same components are denoted by the same reference numerals, and detailed description thereof will be omitted.

The maintenance application program 22P and the learning model 2M stored in the storage unit 22 may be obtained by reading out the maintenance application program 52P and the learning model 50M stored in the storage medium 5 and duplicating them in the storage unit 22.

The processing by the terminal device 2 is the same as the processing executed by the cooperation of the output device 1 and the terminal device 2 in the first embodiment, so detailed description thereof will be omitted.

In this way, the terminal device 2 alone can make a proposal based on the adequacy of the mounting state based on the learning model 2M.

(Fourth embodiment)
In 4th Embodiment, the output device 4 is applied as a control unit of the component mounted in a human-powered vehicle. FIG. 16 is a side view of a human-powered vehicle A to which the output device 4 of the fourth embodiment is applied. The human-powered vehicle A is a road bike including an assist mechanism C that assists the propulsion of the human-powered vehicle A using electrical energy. The configuration of the manpowered vehicle A can be changed arbitrarily. In the first example, the manpowered vehicle A does not include the assist mechanism C. In the second example, the type of manpowered vehicle A is a city cycle, mountain bike, or cross bike. In the third example, manpowered vehicle A includes features of the first and second examples.

A manpowered vehicle A includes a body A1, a handlebar A2, a front wheel A3, a rear wheel A4, a front fork A5, and a saddle A6. A human-powered vehicle A includes a drive mechanism B, an assist mechanism C, an operation device D, a transmission E, a disk rotor G, a display H, and an output device 4 . The manpowered vehicle A includes an acceleration sensor S1 and a sound collector S2. The main body A1 has a frame A12.

The drive mechanism B transmits human power to the rear wheel A4 by chain drive, belt drive, or shaft drive. FIG. 1 illustrates a drive mechanism B of a chain drive. The drive mechanism B includes a crank B1, a first sprocket assembly B2, a second sprocket assembly B3, a chain B4 and a pair of pedals B5.

The first sprocket assembly B2 has a first rotation center axis and is connected to the crankshaft B11 so as to rotate integrally therewith. The first sprocket assembly B2 includes one or more sprockets B22. The rotation center axis of the crankshaft B11 and the rotation center axis of the first sprocket assembly B2 are coaxial.

The second sprocket assembly B3 has a second rotation center axis and is rotatably supported by a hub (not shown) of the rear wheel A4. The second sprocket assembly B3 includes one or more sprockets B31.

The chain B4 is wound around one of the sprockets B22 of the first sprocket assembly B2 and one of the sprockets B31 of the second sprocket assembly B3. When the crank B1 rotates forward by the human power applied to the pair of pedals B5, the first sprocket assembly B2 rotates forward together with the crank B1, and the rotation of the first sprocket assembly B2 rotates through the chain B4 to the second sprocket assembly. The rear wheel A4 rolls forward by being transmitted to the solid B3.

The assist mechanism C assists the human-powered vehicle A in propulsion. In one example, assist mechanism C assists in propulsion of human powered vehicle A by transmitting torque to first sprocket assembly B2. Assist mechanism C includes an electric motor. The assist mechanism C may include a speed reducer. The assist mechanism C causes the chain B4 that transmits the driving force to the rear wheel A4 of the manpowered vehicle A to run. The assist mechanism C is part of a signal controllable component for assisting the running of the chain B4.

The operation device D includes an operation unit D1 operated by a user. An example of the operation unit D1 is one or more buttons. In another example, the operating part D1 may be a brake lever. In another example, the operating unit D1 not only accepts a brake operation, but also accepts an operation related to the transmission E at the operating unit D1. Each time the brake lever is tilted to the left or right, the number of gear stages or gear ratio in the transmission E can be changed. The operation device D is communicatively connected to the components so as to transmit and receive signals according to the operation of the operation unit D1. In the first example, the operation device D is connected to the component for communication by a communication line or an electric wire capable of PLC (Power Line Communication). In a second example, the operating device D is communicatively connected to the component by a wireless communication unit capable of wireless communication.

The transmission E can take various forms. In the first example, the transmission E is an exterior transmission that changes the state of connection between the second sprocket assembly B3 and the chain B4. In the second example, the transmission E is an external transmission that changes the state of connection between the first sprocket assembly B2 and the chain B4. In a third example, transmission E is an internal transmission. In a third example, the moving parts of transmission E include at least one of the sleeve and pawl of the internal transmission. In the fourth example, the transmission E is a continuously variable transmission. In a fourth example, the movable portion of the transmission E includes ball planetaries (planetary rolling elements) of a continuously variable transmission. The transmission E is part of a component controllable by signal control for changing the number of gears.

The transmission E includes a front derailleur E1 and a rear derailleur E2. The front derailleur E1 is attached to the seat tube A121 of the frame A12. The front derailleur E1 has a chain guide, a drive motor E11, and a step number sensor. The front derailleur E1 drives the chain guide of the chain B4 hanging on the sprocket B22 of the first sprocket assembly B2 in the upshift and downshift directions by the drive motor E11. The front derailleur E1 positions the chain guide by the drive motor E11 to any shift position based on a shift command from the operation device D based on the operation part D1 or an automatic shift command from a control unit (not shown).

The chain guide of the front derailleur E1 is positioned during installation by the assembly engineer or the user himself. The front derailleur E1 in the fourth embodiment can set the chain guide to an appropriate position by the drive motor E11.

The rear derailleur E2 is attached via a derailleur hanger to a chain stay A17 that is part of the frame A12 that supports the end of the crankshaft of the rear wheel A4. The rear derailleur E2 has a chain guide, a drive motor E21, and a step number sensor. The rear derailleur E2 drives the chain guide of the chain B4 hanging on the sprocket B31 of the second sprocket assembly B3 in shift-up and shift-down directions by the drive motor E21. The rear derailleur E2 positions the chain guide by the drive motor E21 to any shift position based on a shift instruction from the operation device D based on the operation unit D1 or an automatic shift instruction from a control unit (not shown).

The chain guide of the rear derailleur E2 is positioned during installation by the assembly engineer or by the user himself. The rear derailleur E2 in the fourth embodiment can set the chain guide to an appropriate position by the drive motor E21.

The disc rotor G is a component that constitutes a disc brake, and is provided around the shafts of the front wheel A3 and the rear wheel A4. The disc brake operates based on a braking instruction from the operating device D based on the operating section D1 or an automatic braking instruction from a control unit (not shown). The disk rotor G is positioned during installation by an assembly engineer or by the user himself/herself. The position of the disk rotor G can change due to overturning, contact, wear or wear.

The display H is attached to the handlebar A2 for the user who is the rider. The display H is, in one example, a liquid crystal display. Another example of the display H is an organic EL display. The display H can display the number of gears acquired through communication with the transmission E. FIG. The display H is connected to the output device 4 and displays information output from the output device 1 .

The acceleration sensor S1 is built in the front derailleur E1 attached to the seat tube A121 of the frame A12. The acceleration sensor S1 outputs a signal indicating acceleration corresponding to the vibration of the frame A12. The acceleration sensor S1 may be incorporated not only in the front derailleur E1 but also in the rear derailleur E2. The acceleration sensor S1 may be attached near the disk rotor G, for example, on the front fork A5.

The sound collector S2 is built in the front derailleur E1. The sound collector S2 collects and outputs surrounding sounds. Sound collector S2 may include a microphone. A sound collector S2 built in the front derailleur E1 can collect the driving sound of the chain guide of the front derailleur E1, the contact sound between the chain B4 and the chain guide, the contact sound between the chain B4 and the sprocket B22, and the like.

The sound collector S2 may be incorporated not only in the front derailleur E1 but also in the rear derailleur E2. The sound collector S2 built into the rear derailleur E2 can collect the driving sound of the chain guide of the rear derailleur E2, the contact sound between the chain B4 and the chain guide, the contact sound between the chain B4 and the sprocket B31, and the like.

The output device 4 in the fourth embodiment is attached to the seat tube A121 together with the front derailleur E1. The output device 4 may also serve as a drive control section for the front derailleur E1. FIG. 17 is a block diagram showing the configuration of the output device 4 in the fourth embodiment. The output device 4 is connected to a front derailleur E1, a rear derailleur E2, a disc brake system including a disc rotor G, a display H and an operation device D by wire or wirelessly. The output device 4 includes a processing section 40 , a storage section 42 and an input/output section 46 .

The processing unit 40 is a processor using a CPU or GPU, and uses built-in memories such as ROM and RAM to control learning algorithms and components provided in the human-powered vehicle A to execute processing.

The storage unit 42 includes, for example, nonvolatile memory such as flash memory. Storage unit 42 stores control program 40P and learning model 40M. The control program 40</b>P and the learning model 40</b>M may be the control program 9</b>P and the learning model 90</b>M stored in the storage medium 9 that are read out and duplicated in the storage unit 42 .

The input/output unit 44 is connected to the sensor S1, the sound collector S2, the operation device D, the drive motor E11 of the front derailleur E1 of the transmission E to be controlled, and the drive motor E21 of the rear derailleur E2 provided in the human-powered vehicle A. be done. The processing unit 40 uses the input/output unit 44 to acquire a signal indicating the speed from the acceleration sensor S1. The processing unit 40 acquires the audio signal from the sound collector S2 through the input/output unit 44 . The input/output unit 44 outputs a display signal to the display H. FIG.

The output device 1 attached to such a manpowered vehicle A uses the learning model 10M to determine the mounting state according to the input information that can be obtained from the sensor S1 and the sound collector S2 regarding the mounting state of the parts of the manpowered vehicle. Appropriateness is acquired, and an adjustment instruction signal is output to the drive motor E11, or an adjustment instruction is output to the display H toward the user. As for the disk rotor G, the user is suggested to replace it on the display H as to whether it is installed properly or not, and whether it is worn out.

A learning model 40M in the fourth embodiment will be described. The learning model 40M is generated in advance by the component manufacturer. FIG. 18 is a flow chart showing an example of a method of generating the learning model 40M in the fourth embodiment. The learning model 40M is generated by a supervised deep learning algorithm using a neural network to output output information in response to input information regarding the mounting state of parts of the human powered vehicle.

The teacher data is input information obtained in advance from the acceleration sensor S1 and the sound collector S2 under a test environment or while a model user is actually driving the human-powered vehicle A, and corresponds to the input information. and information on suitability for the mounting state of the parts in the manpowered vehicle A. The teaching data is divided into, for example, teaching data about the front derailleur E1, teaching data about the rear derailleur E2, and teaching data about the disk rotor G, for each part to be judged as appropriate. The input information obtained from the acceleration sensor S1 is data regarding vibration based on the signal output from the acceleration sensor S1. Amplitude, frequency, or power values for the vibration are collected every predetermined sampling period. The input information obtained from the sound collector S2 includes audio data of sounds collected by the sound collector S2. For audio data, information on audio waveform, amplitude, or frequency is collected for each predetermined sampling period. Assembling engineers, manufacturer engineers, etc. are responsible for judging suitability. The subject of determination may be the user. The input information may also include part identification information, vehicle body information, and user physical information.

A generating device managed by a component manufacturer acquires data related to vibrations acquired while the manpower-driven vehicle is being driven, relating to the mounting state of parts of the manpower-driven vehicle that have already been collected (step S301). The generator acquires voice data collected during the same sampling period from the same human-powered vehicle as the vibration-related data acquired in step S301 (step S303).

The generation device creates teacher data by adding a label indicating the result of determining whether the attachment state of the target component is appropriate or not, given in advance, to the input information acquired in steps S301 and S303 (step S305). The generation device may acquire, as input information, any one of identification information of an attached target part, vehicle body information, and physical information of the user, in addition to vibration data and sound data.

The generation device determines whether or not teacher data has been created from all collected information groups (step S307). If it is determined in step S307 that teacher data has not been created from all information groups (S307: NO), the generating device returns the process to step S301.

If it is determined that the teacher data has been created from all the information groups (S307: YES), the generator gives the created teacher data to the NN (step S309). The generator learns parameters including at least one of weight and bias in the hidden layer of the NN from the given teacher data (step S311), and ends the process.

FIG. 19 is a diagram showing an overview of the learning model 40M in the fourth embodiment. The learning model 40M includes an input layer 31 for inputting input information about parts of the manpowered vehicle A, and an output layer 32 for outputting output information about the suitability of the mounting state of the parts of the manpowered vehicle A. The learning model 40M includes an intermediate layer 33 having parameters that have been learned by teacher data so as to output suitability when input information is input. When input information about a part is input to the input layer 31 , computation is performed in the intermediate layer 33 using learned parameters, and output information about the propriety of the input part is output from the output layer 32 .

In the specific example of FIG. 19, the input layer 31 is provided with data on vibrations obtained while the human-powered vehicle A is driving, and audio data on whether or not the mounting state of the target part is appropriate. The input layer 31 may be provided with any or all of the input information shown in the first embodiment, such as part identification information, vehicle body information, and user physical information.

Vibration and sound may be input as an image graphing the variation during the sample period rather than the data itself, as shown in FIG.

In FIG. 19, as output information from the learning model 40M, respective probabilities are output as to whether it is appropriate or, if inappropriate, what type of inappropriate state it is. The processing unit 40 identifies the appropriate state, the type A inappropriate state, and the type B inappropriate state based on the output probabilities of the appropriate state.

The learning model 40M that has been learned in this way is used to install parts in the human-powered vehicle A that is being driven when newly inputting vibration data obtained from the acceleration sensor S1 and voice data obtained from the sound collector S2. Outputs the probabilities of the states regarding whether the states pass or fail.

FIG. 20 is a flow chart showing an example of a processing procedure for judging whether the mounting state is appropriate in the fourth embodiment. The output device 4 uses the learning model 40M stored in the storage unit 42 in advance, and sequentially executes the following processing during each sampling period while the human-powered vehicle A is driving.

The processing unit 40 of the output device 4 acquires, from the acceleration sensor S1, data relating to vibrations generated from the manpower-driven vehicle A being driven (step S401). In step S401, the processing unit 40 may acquire the vibration amplitude and frequency, or may calculate the power value.

The processing unit 40 acquires sound data collected by the manpowered vehicle A in operation from the sound collector S2 (step S403). In step S403, the processing unit 40 may perform predetermined filtering on the collected sound signal obtained from the sound collector S2, and acquire the signal after A/D conversion.

The processing unit 40 gives the input information acquired in steps S401 and S403 to the learning model 40M (step S405). The processing unit 40 specifies the information about the propriety of the attachment state output from the learning model 40M when the input information is given to the learning model 40M (step S407).

The processing unit 40 outputs information for adjusting the mounting state of the target component based on the information on the suitability specified in step S407 (step S409), and ends the process.

The output in step S409 may be an adjustment signal output to the drive motor E11 or drive motor E21, which is an actuator, or may be a display output to the display H.

In this way, in the fourth embodiment, the vibration data or voice data acquired while the human-powered vehicle A is being driven is input to the learning model 40M, and an appropriate mounting state of parts according to the running state is proposed. be possible.

The output device 4 in the fourth embodiment displays and outputs information on the propriety of the attachment state to the display H for the user. The output device 4 includes a communication unit that communicates with a communication terminal device such as the terminal device 2 in the third embodiment owned by the user. The configuration may be such that a proposal or a proposal for exchange can be made.

In the fourth embodiment, mainly, the output device 4 provided near the front derailleur E1 inputs the vibration and sound of the front derailleur E1 to the learning model 40M to output the suitability state. Of course, the object for specifying the appropriateness of the mounting state is not limited to the front derailleur E1. The target may be the distance related to the rear derailleur E2 and the gear, or the stem attached to the head tube of the frame A12 and the handlebar A2 attached to the stem. You may specify the suitability of The object may be the position and angle of the seat post of saddle A6 and the position of the seat portion of saddle A6. The object may be a brake pad, and the output device 4 may specify the propriety of the shape from the vibration and sound of the brake. The target may be the bottom bracket and the mounting position of the pedals. Objects may also include part bends, defects. Sensors for acquiring information to be input to the learning model 4M to specify suitability for these targets are appropriately selected and attached to appropriate locations for each sensor.

Reference Signs List 1, 4 output device 10, 40 processing unit 11, 42 storage unit 1P learning program 1M, 12M, 30M, 32M, 40M, 50M, 90M learning model 2 terminal device 20 Processing unit 22 Storage unit 24 Communication unit 26 Display unit 28 Operation unit 2P Maintenance application program 3, 5, 9 Storage medium A12 Frame A5 Front fork E Shift gear machine, E1... front derailleur, E2... rear derailleur, E11, E21... drive motor, G... disk rotor, H... display, S1... acceleration sensor, S2... sound collector

Claims (14)

Using a learning model learned by a learning algorithm, the mounting including propriety information indicating propriety of the mounting state as a combination of the plurality of parts according to input information regarding the mounting state of the plurality of parts of the human-powered vehicle. A processing unit that outputs output information about the state ,
The output device , wherein the input information includes product number data for each of the plurality of components . 2. The output device according to claim 1 , wherein said input information includes information relating to the vehicle body of said manpowered vehicle. 3. The output device according to claim 2, wherein said input information further includes data on the product number or size of the body of said manpowered vehicle. 4. The output device according to any one of claims 1 to 3, wherein said input information includes physical information of a user of said manpowered vehicle. 5. The output device according to claim 1, wherein said input information includes at least one of still image data of said manpowered vehicle and moving image data of said manpowered vehicle. 6. The output device according to any one of claims 1 to 5, wherein said input information includes data relating to vibrations emitted from said manpowered vehicle. 7. The output device according to claim 6, wherein said vibration-related data includes audio data. 8. The output device according to any one of claims 1 to 7, wherein said output information includes proposal information regarding proposals for said mounting state. 9. The output device according to claim 8, wherein said proposal information includes at least one of information on how to install said part, information on adjustment of said part, and information on replacement of said part. configured to output output information regarding the mounting state including suitability information indicating suitability of the mounting state as a combination of the plurality of parts in response to input information regarding the mounting state of the plurality of parts of the manpowered vehicle. A learning model generation method for generating a learning model using a neural network,
the input information includes product number data for each of the plurality of parts;
A method of generating a learning model, wherein a computer adjusts parameters in an intermediate layer of the neural network in accordance with teacher data labeled with evaluations of the input information. 11. The method of generating a learning model according to claim 10, wherein said learning model is updated by a computer based on additional teacher data labeled with an evaluation of output information outputted by giving said input information to said neural network. configured to output output information regarding the mounting state including suitability information indicating suitability of the mounting state as a combination of the plurality of parts in response to input information regarding the mounting state of the plurality of parts of the manpowered vehicle. A learning model generation method for generating a learning model using a neural network,
the input information includes product number data for each of the plurality of parts;
A method of generating a learning model, wherein parameters in an intermediate layer of the neural network are adjusted by a computer according to teacher data labeled with evaluations of output information outputted by giving the input information to the neural network. A learning algorithm for outputting output information regarding the mounting state including propriety information indicating the propriety of the mounting state as a combination of the plurality of parts according to input information regarding the mounting state of the plurality of parts of the manpowered vehicle. Giving the acquired input information to the trained learning model,
the input information includes product number data for each of the plurality of parts;
Based on the output information output from the learning model, information relating to the mounting state including propriety information indicating the propriety of the mounting state of the parts in the manpowered vehicle is superimposed on the image on a display unit for displaying the image of the manpowered vehicle. and a computer program that causes a computer to perform the process of displaying side by side. 14. A storage medium storing the computer program according to claim 13.

Images