JP7626352B2 - Estimation device, control system, learning model, learning model generation method, computer program, and storage medium - Google Patents

Description

The present invention relates to an estimation device, a control system, a learning model, a method for generating a learning model, a computer program, and a storage medium.

Non-patent document 1 discloses a method for calculating the center of gravity of a bicycle.

"Bicycle Practical Handbook," edited by the Bicycle Industry Promotion Association, 5th edition, Automobile Industry Promotion Association, 1993

In Non-Patent Document 1, the center of gravity of the bicycle is calculated when no rider is riding the bicycle, and it is not possible to estimate the center of gravity of the bicycle when a rider is riding the bicycle.

The present invention aims to provide an estimation device, a control system, a learning model, a method for generating a learning model, a computer program, and a storage medium that can estimate the center of gravity position of a human-powered vehicle when a rider is on board.

The estimation device according to the first aspect of the present invention estimates the center of gravity position of the human-powered vehicle when a rider is riding on the human-powered vehicle based on vehicle body information about the vehicle body of the human-powered vehicle and load information about the load acting on the human-powered vehicle.

The estimation device according to the first aspect of the present invention can estimate the center of gravity position of a human-powered vehicle with a rider on board based on vehicle body information and load information of the human-powered vehicle.

The estimation device according to the second aspect of the present invention estimates the center of gravity position using a learning model configured to output an estimation result regarding the center of gravity position of the human-powered vehicle in response to input of the vehicle body information and the load information.

According to the estimation device according to the second aspect of the present invention, the position of the center of gravity of a human-powered vehicle with a rider on board can be estimated by inputting vehicle body information and load information of the human-powered vehicle into a learning model.

In the estimation device of the third aspect according to the first or second aspect of the present invention, the vehicle body information includes information on the vehicle body dimensions, vehicle body weight, and tilt angle of the human-powered vehicle.

The estimation device according to the third aspect of the present invention can estimate the center of gravity position of a human-powered vehicle with a rider on board based on information related to the vehicle dimensions, weight, and tilt angle of the human-powered vehicle.

In the estimation device of the fourth aspect according to the first or second aspect of the present invention, the human-powered vehicle includes a front axle, a rear axle, a saddle, a handlebar, a crank, and pedals, and the load information includes information regarding a load acting on at least one of the front axle, the rear axle, the saddle, the handlebar, the crank, and the pedals.

The estimation device according to the fourth aspect of the present invention can estimate the center of gravity position of a human-powered vehicle with a rider on it, based on the load acting on at least one of the front axle, rear axle, saddle, handlebars, crank, and pedals.

In the estimation device of the fifth aspect according to the fourth aspect of the present invention, the load information includes at least one of the rider's weight, height, posture, and gender.

The estimation device according to the fifth aspect of the present invention can estimate the position of the center of gravity of a human-powered vehicle when a rider is riding on the vehicle, based on the rider's weight, height, posture, and gender.

A control system according to a sixth aspect of the present invention includes an estimation device according to any one of the first to fifth aspects of the present invention, and a control device that controls components of the human-powered vehicle based on the estimation result.

The control system according to the sixth aspect of the present invention controls the components of a human-powered vehicle based on the results of estimating the center of gravity position, so that, for example, the components of the human-powered vehicle can be controlled in a manner that is comfortable for the rider.

In the seventh aspect of the control system according to the sixth aspect of the present invention, the component includes an alarm device.

According to the control system according to the seventh aspect of the present invention, the controlled component includes an alarm device, so that the estimated center of gravity position can be notified to the rider.

In the control system of the eighth aspect according to the seventh aspect of the present invention, the control device controls the notification device to notify the estimation result.

The control system according to the eighth aspect of the present invention can notify the rider of the estimated center of gravity position.

In the control system of the ninth aspect according to the seventh aspect of the present invention, the control device controls the notification device to notify information regarding the recommended posture of the rider.

The control system according to the ninth aspect of the present invention can provide information regarding the recommended rider posture so that the center of gravity is positioned appropriately.

In the control system of aspect 10 according to any one of aspects 6 to 9 of the present invention, the components include at least one of a lighting device, a brake device, a gear shifting device, a suspension device, a drive assist device, and an adjustable seat post.

The control system according to the tenth aspect of the present invention controls the components of the human-powered vehicle, including at least one of the lighting device, the brake device, the transmission, the suspension device, the drive assist device, and the adjustable seat post, in accordance with the result of estimating the center of gravity position, so that, for example, the components of the human-powered vehicle can be controlled in a comfortable manner for the rider.

The learning model according to the eleventh aspect of the present invention comprises an input layer to which vehicle body information on the vehicle body of a human-powered vehicle and load information on the load acting on the human-powered vehicle are input, an output layer to which an estimation result on the center of gravity position of the human-powered vehicle with a rider on board is output, and an intermediate layer that uses the vehicle body information, the load information, and center of gravity information indicating the center of gravity position as training data to learn the relationship between the vehicle body information and the load information input to the input layer and the estimation result output from the output layer, and is configured such that a computer executes arithmetic processing in the intermediate layer to output the estimation result from the output layer in response to the input of the vehicle body information and the load information to the input layer.

The learning model according to the eleventh aspect of the present invention allows a computer to realize an execution environment for estimating the center of gravity position of a human-powered vehicle when a rider is riding on it.

A method for generating a learning model according to a twelfth aspect of the present invention uses vehicle body information about the body of a human-powered vehicle, load information about the load acting on the human-powered vehicle, and center of gravity information indicating the center of gravity position of the human-powered vehicle as training data, and uses a computer to generate a learning model that outputs an estimation result about the center of gravity position in response to input of the vehicle body information and the load information.

According to the method for generating a learning model according to the twelfth aspect of the present invention, by collecting vehicle body information and load information of a human-powered vehicle, as well as center of gravity information indicating the center of gravity position of the human-powered vehicle, it is possible to generate a learning model for estimating the center of gravity position of a human-powered vehicle with a rider on board.

A computer program according to a thirteenth aspect of the present invention is a computer program for causing a computer to execute a process for estimating the position of the center of gravity of a human-powered vehicle with a rider on board, based on vehicle body information relating to the vehicle body of the human-powered vehicle and load information relating to the load acting on the human-powered vehicle.

The computer program according to the thirteenth aspect of the present invention can provide an execution environment for estimating the center of gravity position of a human-powered vehicle when a rider is riding on the vehicle, using a computer.

A storage medium according to the fourteenth aspect of the present invention stores a computer program according to the thirteenth aspect of the present invention.

According to the storage medium according to the fourteenth aspect of the present invention, by installing the computer program stored in the storage medium into a computer, an execution environment for estimating the position of the center of gravity of a human-powered vehicle when a rider is riding on the vehicle can be realized by the computer.

This application makes it possible to estimate the center of gravity of a human-powered vehicle when a rider is on board.

1 is a side view of a human-powered vehicle to which an estimation device according to a first embodiment is applied. 1 is a block diagram showing an internal configuration of an estimation device according to a first embodiment. FIG. FIG. 1 is a schematic diagram showing an implementation example of a learning model. FIG. 13 is a schematic diagram showing an example of setting a coordinate system. 11 is a flowchart illustrating a procedure for estimating a center of gravity position. FIG. 11 is a block diagram illustrating a control system according to a second embodiment. 10 is a flowchart showing a control procedure in a second embodiment. FIG. 13 is a schematic diagram showing an example of a notification example. FIG. 11 is a schematic diagram showing another example of the notification example. FIG. 2 is a block diagram showing an internal configuration of a server device. FIG. 13 is a conceptual diagram illustrating an example of teacher data. 13 is a flowchart illustrating a procedure for generating a learning model by a server device.

Hereinafter, the present invention will be described in detail with reference to the drawings showing embodiments thereof.
First Embodiment
1 is a side view of a human-powered vehicle 1 to which an estimation device 100 of the first embodiment is applied. The human-powered vehicle 1 is a vehicle that uses human power at least in part as a driving force for traveling. Vehicles that use only an internal combustion engine or an electric motor as a driving force are excluded from the human-powered vehicle 1 of this embodiment. The human-powered vehicle 1 is, for example, a bicycle, including a mountain bike, a road bike, a cross bike, a city cycle, etc.

The human-powered vehicle 1 comprises a vehicle body 10, a front wheel 12, a rear wheel 14, a handlebar 16, a saddle 18, and a drive mechanism 20. In the following description, terms expressing the front-rear, left-right, and up-down directions are used based on the directions when the rider is seated on the saddle 18 of the human-powered vehicle 1.

The vehicle body 10 comprises a frame 10A and a front fork 10B. The front wheel 12 is rotatably supported at the end of the front fork 10B via a front wheel axle 12A. The rear wheel 14 is rotatably supported at the rear end of the frame 10A via a rear wheel axle 14A. The handlebars 16 are supported by the frame 10A so that the direction of travel of the front wheel 12 can be changed.

The drive mechanism 20 transmits the human-powered driving force to the rear wheel 14 by a chain drive, a belt drive, or a shaft drive. FIG. 1 illustrates a chain drive drive mechanism 20. The drive mechanism 20 includes a crank 22, a front sprocket 24A, a rear sprocket 24B, a chain 26, and a pair of pedals 28, 28. The drive mechanism 20 may further include a chain device or a chain tensioner for stably holding the chain 26.

The crank 22 includes a right crank 22A, a left crank 22B, and a crankshaft 22C. The crankshaft 22C is rotatably supported by the frame 10A. The right crank 22A and the left crank 22B are each connected to the crankshaft 22C. One of the pair of pedals 28, 28 is rotatably supported by the right crank 22A, and the other is rotatably supported by the left crank 22B.

The front sprocket 24A is connected to the crankshaft 22C and rotates integrally with the crankshaft 22C. In one example, the front sprocket 24A is a sprocket assembly made up of multiple sprockets with different outer diameters. When the front sprocket 24A is equipped with multiple sprockets, the outer diameters of these sprockets increase, for example, as they move away from the vehicle body 10.

The rear sprocket 24B is rotatably supported on the hub of the rear wheel 14. In one example, the rear sprocket 24B is a sprocket assembly made up of multiple sprockets with different outer diameters. When the rear sprocket 24B is equipped with multiple sprockets, the outer diameters of these sprockets become smaller, for example, as they move away from the vehicle body 10.

The chain 26 is wound around the front sprocket 24A and the rear sprocket 24B. When the crank 22 rotates forward due to the human driving force applied to the pedals 28, 28, the front sprocket 24A rotates forward together with the crank 22. The rotation of the front sprocket 24A is transmitted to the rear sprocket 24B via the chain 26, causing the rear wheel 14 to rotate.

The human-powered vehicle 1 includes at least one of the following components: an alarm device 30, a lighting device 32, a brake device 34, a gear change device 36, a suspension device 38, a drive assist device 40, and an adjustable seat post 42. The human-powered vehicle 1 further includes an operating device 44 that switches the control state of at least one of the components, and a battery unit 46 that supplies power to the components.

The notification device 30 is a device that notifies the rider of various information. The notification device 30 is provided, for example, on the handlebar 16. In one example, the notification device 30 is a display device equipped with a liquid crystal display panel or the like, and displays information such as characters, figures, and symbols based on a control signal input from the outside. In another example, the notification device 30 is an audio output device equipped with a speaker or the like, and outputs audio based on a control signal input from the outside. The notification device 30 is not limited to audio, and may be configured to output sounds such as warning sounds or to generate vibrations. The notification device 30 does not need to be an independent device, and may be a device mounted on the estimation device 100.

The lighting device 32 is a device that illuminates the direction of travel of the human-powered vehicle 1. The lighting device 32 is provided, for example, on the front fork 10B. The lighting device 32 is configured to turn on, blink, or turn off in response to a control signal input from the outside.

In one example, the brake device 34 is a device that applies a braking force to the wheel of the front wheel 12. The brake device 34 includes an electric drive unit 34A, a caliper 34B, a brake operation device 34C, and the like. The electric drive unit 34A includes an electric motor that operates the caliper 34B and a drive circuit that drives the electric motor. The caliper 34B includes a brake pad that can contact the wheel of the front wheel 12. The brake operation device 34C is provided, for example, on the handlebar 16 and receives the operation of the rider who applies the brakes. The electric drive unit 34A of the brake device 34 drives the electric motor in response to the operation of the brake operation device 34C, and operates the caliper 34B to bring the brake pad into contact with the wheel of the front wheel 12, thereby attenuating the rotational force of the front wheel 12. The brake device 34 may be configured to operate in a plurality of control states with different braking forces.

The brake device 34 may be a disc brake. In a disc brake, for example, an electric motor controls fluid to move brake pads and press the brake pads against a rotor, thereby attenuating the rotational force of the front wheel 12 or the rear wheel 14. The disc brake may also be configured to directly move the brake pads with an electric motor and attenuate the rotational force of the front wheel 12 or the rear wheel 14 by pressing the brake pads against a rotor.

The brake device 34 may be provided for the rear wheel 14. The front and rear brake devices 34 may share the brake operation device 34C. The front and rear brake devices 34 may each be provided with a brake operation device 34C. The configuration of the brake device 34 provided for the rear wheel 14 is similar to that of the brake device 34 for the front wheel 12, so a description thereof will be omitted.

The transmission 36 is a device that changes the gear ratio of the human-powered vehicle 1. The transmission 36 is equipped with an electric drive unit 36A whose operation is controlled based on a control signal output in response to the operation of the operating device 44. The electric drive unit 36A includes a drive circuit and an electric motor. The transmission 36 drives the electric motor equipped in the electric drive unit 36A to change the sprocket around which the chain 26 is wound, thereby switching the gear ratio of the human-powered vehicle 1.

One example of the transmission 36 is an external transmission. More specifically, the transmission 36 is a rear derailleur. In this case, the rear sprocket 24B includes multiple sprockets with different outer diameters. When the drive circuit of the electric drive unit 36A receives a control signal from the operation device 44 instructing an upshift, it drives the electric motor, for example, to change the sprocket around which the chain 26 is wrapped from the top side to the low side. When the drive circuit of the electric drive unit 36A receives a control signal from the operation device 44 instructing a downshift, it drives the electric motor, for example, to change the sprocket around which the chain 26 is wrapped from the low side to the top side.

The transmission 36 may be a front derailleur. In this case, the front sprocket 24A includes multiple sprockets with different outer diameters. The transmission 36 may include both a front derailleur and a rear derailleur. The transmission 36 may be an internal transmission. If the transmission 36 is an internal transmission, the transmission 36 is provided, for example, in the hub of the rear wheel 14, and transmits the rotation input to the rear sprocket 24B to the rear wheel 14 after changing the speed. The transmission 36 is not limited to an external transmission or an internal transmission, and may also be a continuously variable transmission.

In one example, the suspension device 38 is a front suspension provided on the front fork 10B that attenuates shocks applied to the front wheel 12. In another example, the suspension device 38 is a rear suspension that attenuates shocks applied to the rear wheel 14. The suspension device 38 includes an electric drive unit 38A whose operation is controlled based on a control signal output in response to the operation of the operating device 44. The electric drive unit 38A includes a drive circuit and an electric motor. The suspension device 38 is controlled by setting a damping rate, a stroke amount, and a lockout state as operating parameters. The suspension device 38 may be configured to operate in a plurality of control states with different cushioning properties.

The drive assist device 40 is a device for assisting the manual drive and is provided for the drive mechanism 20. The drive assist device 40 includes an electric drive unit 40A whose operation is controlled based on a control signal input from the outside. The electric drive unit 40A includes a drive circuit and an electric motor, and the drive circuit drives the electric motor to assist the manual drive. The operation of the electric drive unit 40A may be controlled based on a control signal output in response to the operation of the operation device 44. The drive assist device 40 can operate in a plurality of control states with different assist forces that assist the manual drive. The control state in the drive assist device 40 is also called the assist level. The drive assist device 40 changes the assist level according to the manual drive force acting on the crank 22, for example. When four assist levels are provided, for example, they include a strong assist level, a medium assist level, a weak assist level, and a zero assist level. At the zero assist level, no assist is performed by the drive assist device 40.

The adjustable seat post 42 is attached to the frame 10A. The adjustable seat post 42 is equipped with an electric drive unit 42A whose operation is controlled based on a control signal output in response to the operation of the operating device 44. The electric drive unit 42A includes an electric actuator that raises and lowers the saddle 18 relative to the frame 10A, and a drive circuit that drives the electric actuator. The adjustable seat post 42 is controlled by setting the support position of the saddle 18 relative to the frame 10A.

The operating device 44 is provided, for example, on the handlebar 16. The operating device 44 includes operating switches 44A and 44B that are operated by the rider's fingers. The operating switches 44A and 44B are switches for switching the control state of at least one of the alarm device 30, the lighting device 32, the brake device 34, the gear shift device 36, the suspension device 38, the drive assist device 40, and the adjustable seat post 42.

The number of operating devices 44 mounted on the human-powered vehicle 1 does not need to be one, and may be multiple. In the example shown in FIG. 1, the operating device 44 is configured to include two operating switches 44A, 44B, but it may also be configured to include one or three or more operating switches. The operating device 44 is not limited to a configuration including a switch, and may also be configured to include an operating lever, an operating dial, or the like.

The operating device 44 is connected to each component so that a control signal corresponding to the operation of the operating switch 44A or the operating switch 44B can be transmitted to the controlled component. In one example, the operating device 44 is connected to the controlled component by a communication line or an electric wire capable of PLC (Power Line Communication). In another example, the operating device 44 is connected to the controlled component by a wireless communication unit capable of wireless communication.

The human-powered vehicle 1 may include a battery unit 46 that supplies power to at least one of the aforementioned components. The battery unit 46 includes a battery 46A and a battery holder 46B. The battery 46A is a storage battery including one or more battery cells. The battery holder 46B is fixed to the frame 10A of the human-powered vehicle 1. The battery 46A is detachable from the battery holder 46B. The battery 46A is electrically connected to the alarm device 30, the lighting device 32, the gear shift device 36, the brake device 34, the suspension device 38, the drive assist device 40, the adjustable seat post 42, and the like.

The human-powered vehicle 1 may further include at least one of a tilt sensor S1, a load sensor S2, and a posture sensor S3. In the following description, when there is no need to describe each of the tilt sensor S1, load sensor S2, and posture sensor S3 individually, they will also be referred to as sensors S1 to S3.

The tilt sensor S1 is a sensor that outputs a signal according to the tilt angle of the human-powered vehicle 1. The tilt angle detected by the tilt sensor S1 is, for example, the rotation angle around the pitch axis along the left-right direction of the human-powered vehicle 1. As an example, the tilt sensor S1 includes a sensor that detects the angular velocity of the pitch angle, and calculates the integrated value of the angular velocity around the pitch axis as the pitch angle. The tilt sensor S1 may also be configured to measure both the rotation angle around the roll axis along the front-rear direction of the human-powered vehicle 1, and the rotation speed around the yaw axis along the up-down direction of the human-powered vehicle 1.

The load sensor S2 is a sensor that outputs a signal corresponding to the load acting on the human-powered vehicle 1. An example of the load sensor S2 is a strain gauge type load cell. The load sensor S2 is provided on at least one of the front wheel axle 12A, the rear wheel axle 14A, the handlebar 16, the saddle 18, and the pedal 28.

The posture sensor S3 is a sensor that outputs a signal according to the rider's posture. One example of the posture sensor S3 is a piezoelectric sensor, which is provided at multiple locations on the human-powered vehicle 1 where the rider's weight is applied. For example, the posture sensor S3 is provided at one or multiple locations such as the grips on the handlebars 16, the surface of the saddle 18, the pedals 28, etc. The posture sensor S3 may be a motion sensor provided in a wearable device attached to the rider. The posture sensor S3 may be shared with at least one of the load sensors S2.

The estimation device 100 estimates the center of gravity position of the human-powered vehicle 1 when the rider is riding on the human-powered vehicle 1, based on vehicle body information related to the vehicle body of the human-powered vehicle 1 and load information related to the load acting on the human-powered vehicle 1. In one example, the estimation device 100 is a dedicated terminal such as a cycle computer provided on the human-powered vehicle 1. In another example, the estimation device 100 is a general-purpose terminal such as a smartphone, tablet terminal, or wearable terminal carried by the rider of the human-powered vehicle 1.

The vehicle body information used to estimate the center of gravity position includes information on the vehicle body dimensions, vehicle body weight, and tilt angle of the human-powered vehicle 1. For the vehicle body dimensions and vehicle body weight of the human-powered vehicle 1, design values or previously measured values may be stored in the estimation device 100 in advance. On the other hand, the tilt angle of the human-powered vehicle 1 is information obtained based on the output of the tilt sensor S1.

The load information used to estimate the center of gravity position includes information about the load acting on at least one of the front axle 12A, rear axle 14A, handlebar 16, saddle 18, crank 22, and pedal 28. This information is obtained based on the output of the load sensor S2.

The load information used to estimate the center of gravity position may include at least one of the rider's weight, height, posture, and gender. The rider's weight, height, and gender are information input by the rider using, for example, the operation device 44. These pieces of information are input using, for example, the operation device 44 and stored in the estimation device 100. The rider's posture is information obtained based on the output of the posture sensor S3.

The configuration of the estimation device 100 will be described below.
2 is a block diagram showing the internal configuration of the estimation device 100 according to the first embodiment. The estimation device 100 includes an input unit 102, a calculation processing unit 104, a storage unit 106, and an output unit 108.

The input unit 102 has an interface that connects the sensors S1 to S3. The interface that the input unit 102 has is, for example, a wired interface, and connects the sensors S1 to S3 via a communication cable. Information based on a signal input through the input unit 102 is temporarily stored in the storage unit 106. The interface that the input unit 102 has may be a wireless interface. As the wireless interface, a communication interface conforming to a communication standard including Bluetooth (registered trademark), WiFi (registered trademark), ZigBee (registered trademark), LTE (Long Term Evolution), and other wireless LANs (Local Area Networks) can be used.

The calculation processing unit 104 includes a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), etc. The ROM included in the calculation processing unit 104 stores computer programs and the like for controlling the operation of each part of the hardware included in the estimation device 100. The CPU in the calculation processing unit 104 executes computer programs stored in the ROM or the memory unit 106, and controls the operation of each part of the hardware, thereby controlling the operation of the entire device. The RAM included in the calculation processing unit 104 temporarily stores data used during the execution of calculations.

The arithmetic processing unit 104 is configured to include a CPU, ROM, and RAM, but may also be one or more arithmetic circuits including a GPU (Graphics Processing Unit), an FPGA (Field Programmable Gate Array), a DSP (Digital Signal Processor), a quantum processor, volatile or non-volatile memory, etc.

The storage unit 106 includes memories such as an EEPROM (Electronically Erasable Programmable Read Only Memory) and an SRAM (Static Random Access Memory). The storage unit 106 stores various computer programs including an estimation processing program 110 executed by the calculation processing unit 104, a learning model 120 (described later), various data used when estimating the center of gravity position of the human-powered vehicle 1, and the like.

The estimation processing program 110 is a computer program that causes the estimation processing device 100 to execute a process of estimating the position of the center of gravity of the human-powered vehicle 1 when a rider is riding on the human-powered vehicle 1, based on vehicle body information related to the vehicle body of the human-powered vehicle 1 and load information related to the load acting on the human-powered vehicle 1. By executing the estimation processing program 110, the calculation processing unit 104 estimates the position of the center of gravity of the human-powered vehicle 1 when a rider is riding on the human-powered vehicle 1, based on the vehicle body information and load information of the human-powered vehicle 1, using the learning model 120.

In one example, the various computer programs including the estimation processing program 110 may be provided by a storage medium M on which the computer programs are stored. The storage medium M is, for example, a portable memory such as a CD-ROM, a USB memory, a Secure Digital (SD) card, a micro SD card, or a Compact Flash (registered trademark). In this embodiment, the storage medium M is a non-transitory storage medium that readably stores the various computer programs including the estimation processing program 110. The calculation processing unit 104 reads the various computer programs from the storage medium M using a reading device not shown in the figure, and installs the various computer programs that have been read into the storage unit 106.

The learning model 120 stored in the memory unit 106 is described by its definition information. The definition information of the learning model 120 includes structural information of the learning model 120, various parameters such as weights and biases between nodes used in the learning model 120, and the like. In this embodiment, the learning model 120 learned by a predetermined learning model is stored in the memory unit 106 using the body information and load information of the human-powered vehicle 1, as well as center of gravity information indicating the center of gravity position of the human-powered vehicle 1, as teacher data. As an example, the learning algorithm can be a supervised learning algorithm using a neural network.

The output unit 108 includes an output interface that outputs the estimation result by the calculation processing unit 104. The output format of the estimation result is arbitrary. In one example, the output unit 108 includes a display device such as a liquid crystal display or an organic EL (Electro-Luminescence) display. In this case, the output unit 108 may display the estimation result of the center of gravity position on the display device as text information or image information. In another example, the output unit 108 includes an audio output device such as a speaker. In this case, the output unit 108 may output the estimation result of the center of gravity position as audio from the audio output device. In yet another example, the output unit 108 includes a communication interface. In this case, the output unit 108 may output the estimation result of the center of gravity position to an external device connected to the communication interface.

Figure 3 is a schematic diagram showing an example implementation of the learning model 120. The learning model 120 is, for example, a machine learning learning model including deep learning, and is configured by a neural network. The learning model 120 includes an input layer 122, intermediate layers 124A and 124B, and an output layer 126. In the example of Figure 3, two intermediate layers 124A and 124B are shown, but the number of intermediate layers is not limited to two and may be three or more.

The input layer 122, the intermediate layers 124A and 124B, and the output layer 126 each have one or more nodes, and the nodes of each layer are connected in one direction with the nodes in the previous and next layers with the desired weights and biases. Vector data having the same number of components as the number of nodes in the input layer 122 is provided as input data to the learning model 120. For example, the data provided to each node of the input layer 122 is the body information and load information of the human-powered vehicle 1. The body information includes information on the body dimensions, body weight, and tilt angle of the human-powered vehicle 1. For the body dimensions and body weight, information stored in the memory unit 106 of the estimation device 100 is used. For the tilt angle, information obtained based on the output signal of the tilt sensor S1 is used. The load information includes information on the load acting on at least one of the front wheel axle 12A, the rear wheel axle 14A, the handlebar 16, the saddle 18, the crank 22, and the pedals 28. For these load information, information obtained based on the output signal of the load sensor S2 is used. The load information may further include at least one of the rider's weight, height, posture, and gender. For the rider's weight, height, and gender, information stored in the memory unit 106 of the estimation device 100 is used. For the information regarding the rider's posture, information obtained based on the output signal of the posture sensor S3 is used.

The data given to each node of the input layer 122 is given to the first intermediate layer 124A. In the intermediate layer 124A, an output is calculated using an activation function including weights and biases, and the calculated value is given to the next intermediate layer 124B, and so on until the output of the output layer 126 is obtained. Various parameters such as weights and biases that connect the nodes are learned by a predetermined learning algorithm. For example, a deep learning learning algorithm is used as the learning algorithm for learning various parameters. In this embodiment, various parameters can be learned by a predetermined learning algorithm using the vehicle body information and load information of the human-powered vehicle 1, and center of gravity information indicating the center of gravity position of the human-powered vehicle 1 as teacher data. The learning method will be described in detail later.

The output layer 126 outputs an estimation result regarding the center of gravity position of the human-powered vehicle 1 when the rider is riding on it. The output form of the estimation result by the output layer 126 is arbitrary. For example, the output layer 126 may set a coordinate system based on the human-powered vehicle 1 and output the probability that each of the multiple positions set in this coordinate system is the center of gravity position. Figure 4 is a schematic diagram showing an example of setting the coordinate system. The example of Figure 4 shows an orthogonal coordinate system in which the front-to-rear direction of the human-powered vehicle 1 is the X-axis direction and the up-down direction is the Z-axis direction. In this coordinate system, if m points are taken in the X-axis direction and n points are taken in the Z-axis direction, the position coordinates of each point can be expressed as (X1, Z1), (X2, Z1), ..., (Xi, Zj), ..., (Xm, Zn). Here, i is an integer from 1 to m, and j is an integer from 1 to n. m and n are integers of 2 or more. The output layer 126 of the learning model 120 outputs the probability that each position indicated by these position coordinates is a center of gravity position. In this case, the output layer 126 is provided with m×n nodes, and each node outputs the probability that each position is a center of gravity position.

As described above, the learning model 120 is configured such that the estimation device 100 executes calculation processing in the intermediate layers 124A and 124B in order to output the estimation result of the center of gravity position from the output layer 126 in response to the input of vehicle body information and load information to the input layer 122.

In the example of Figure 4, for simplicity, the position coordinates are set in a two-dimensional Cartesian coordinate system that includes the forward/rearward and upward directions of the human-powered vehicle 1, but it goes without saying that the position coordinates can also be set in a three-dimensional Cartesian coordinate system that includes the left/right direction as the Y-axis direction.

Next, the procedure for estimating the center of gravity will be described.
5 is a flow chart for explaining the procedure for estimating the center of gravity position. In step S101, the calculation processing unit 104 of the estimation device 100 acquires vehicle body information and load information of the human-powered vehicle 1. The calculation processing unit 104 may acquire the vehicle body information by reading the vehicle body dimensions and vehicle body weight from the storage unit 106, or may acquire the tilt angle obtained from the output signal of the tilt sensor S1 as the vehicle body information. The calculation processing unit 104 may acquire the load information based on the output signal of the load sensor S2, and further, the calculation processing unit 104 may acquire the load information by reading the weight, height, and sex of the rider from the storage unit 106, or may acquire the load information based on the output signal of the attitude sensor S3.

In step S102, the calculation processing unit 104 inputs the vehicle body information and load information of the human-powered vehicle 1 to the learning model 120. At this time, the calculation processing unit 104 may provide the vehicle body information and load information acquired in step S101 to the input layer 122 of the learning model 120, or may provide the vehicle body information and load information that have been subjected to calculation processing to match the input form of each node to the input layer 122 of the learning model 120.

The learning model 120 performs calculations by providing vehicle body information and load information to the input layer 122. Information provided to each node of the input layer 122 is output to the adjacent node of the intermediate layer 124A. In the intermediate layer 124A, calculations are performed using an activation function including weights and biases between nodes, and the calculation results are output to the subsequent intermediate layer 124B. In the intermediate layer 124B, calculations are further performed using an activation function including weights and biases between nodes, and the calculation results are output to each node of the output layer 126. In this way, the learning model 120 outputs an estimation result regarding the center of gravity position of the human-powered vehicle 1. In other words, the learning model 120 is configured so that the estimation device 100 performs calculations in the intermediate layers 124A and 124B to output an estimation result of the center of gravity position from the output layer 126 in response to the input of vehicle body information and load information to the input layer 122.

In step S103, the calculation processing unit 104 acquires the calculation results by the learning model 120 from the output layer 126. For example, when position coordinates are set in an orthogonal coordinate system as shown in FIG. 4, the calculation processing unit 104 acquires the probability that each position indicated by the position coordinates (X1, Z1), (X2, Z1), ..., (Xi, Zj), ..., (Xm, Zn) is the center of gravity position.

In step S104, the calculation processing unit 104 estimates the center of gravity position of the human-powered vehicle 1 with the rider on board, based on the calculation results obtained from the output layer 126 of the learning model 120. In one example, the calculation processing unit 104 compares the magnitude of the probability output from each node of the output layer 126, and estimates the position with the highest probability as the center of gravity position. In another example, the calculation processing unit 104 may calculate a weighted average using the probability at each position coordinate as a weight, and estimate the calculation result as the center of gravity position.

As described above, the estimation device 100 in the first embodiment can estimate the center of gravity position of the human-powered vehicle 1 with a rider on board by inputting vehicle body information and load information of the human-powered vehicle 1 into the learning model 120.

The estimation device 100 in the first embodiment is configured to estimate the center of gravity position using the learning model 120, but it may also be configured to estimate the center of gravity position without using the learning model 120. For example, the estimation device 100 may be configured to prepare a function that outputs the center of gravity position of the human-powered vehicle 1 with a rider on board in response to input of vehicle body information and load information of the human-powered vehicle 1, and to estimate the center of gravity position using this function.

Second Embodiment
In the second embodiment, a control system that controls components of the human-powered vehicle 1 based on the estimation result by the estimation device 100 will be described.

Figure 6 is a block diagram illustrating a control system according to the second embodiment. The control system according to the second embodiment includes an estimation device 100 and a control device 200 that controls components of the human-powered vehicle 1 based on the estimation results by the estimation device 100. The control device 200 may be provided as an independent device, or may be provided within the component to be controlled. In the example shown in Figure 6, the estimation device 100 and the control device 200 are described as separate entities, but the estimation device 100 and the control device 200 may also be configured as an integrated unit.

The control device 200 includes a control unit 202, a storage unit 204, an input unit 206, and an output unit 208. The control unit 202 includes, for example, a CPU, a ROM, and a RAM. The ROM included in the control unit 202 stores control programs and the like for controlling the operation of each hardware unit included in the control device 200. The CPU in the control unit 202 executes the control programs stored in the ROM and various programs stored in the storage unit 204, and controls the operation of each hardware unit, thereby realizing the control device of the present application. Specifically, the control unit 202 generates control signals for controlling components of the human-powered vehicle 1 based on the estimation results from the estimation device 100 input through the input unit 206, and outputs the generated control signals from the output unit 208 to the components to be controlled.

The control unit 202 is not limited to the above configuration. The control unit 202 may be one or more control circuits including a single-core CPU, a multi-core CPU, an FPGA, a volatile or non-volatile memory, etc. The control unit 202 may also have functions such as a clock that outputs date and time information, a timer that measures the elapsed time from when an instruction to start measurement is given to when an instruction to end measurement is given, and a counter that counts numbers.

The storage unit 204 includes memories such as an EEPROM and an SRAM. The storage unit 204 stores computer programs executed by the control unit 202, as well as data used by these computer programs.

The input unit 206 has an interface that connects the estimation device 100 via a cable. The estimation result output from the estimation device 100 is input to the input unit 206. The estimation result input to the input unit 206 represents the center of gravity position of the human-powered vehicle 1 with a rider on board. The input unit 206 outputs the input estimation result to the control unit 202. In this embodiment, the estimation device 100 is connected via a cable, but it may also be configured so that data can be transmitted and received between the estimation device 100 and the control device 200 via a wireless communication interface. For wireless communication, a wireless communication method conforming to communication standards such as Bluetooth, WiFi, ZigBee, LTE, and other wireless LANs can be used.

The output unit 208 has an interface that connects the components to be controlled via a cable. The components connected to the control device 200 include the alarm device 30, the lighting device 32, the brake device 34, the gear shift device 36, the suspension device 38, the drive assist device 40, and the adjustable seat post 42. The output unit 208 outputs a control signal output from the control unit 202 to the components to be controlled. In this embodiment, the components of the human-powered vehicle 1 are connected via cables, but it may also be configured so that data can be sent and received between the control device 200 and the components of the human-powered vehicle 1 via a wireless communication interface. For wireless communication, a wireless communication method that complies with the various communication standards described above can be used.

The following describes the control procedure in which the control device 200 controls the components of the human-powered vehicle 1 based on the estimation results of the estimation device 100.

Figure 7 is a flowchart showing the control procedure in the second embodiment. In step S201, the control unit 202 of the control device 200 acquires the estimation result output from the estimation device 100 via the input unit 206. In step S202, the control unit 202 generates a control signal for controlling the component to be controlled based on the estimation result acquired from the estimation device 100.

In step S203, the control unit 202 outputs the control signal generated in step S202 from the output unit 208 to the component to be controlled, and controls the component to be controlled.

For example, if the component to be controlled is the notification device 30, the control unit 202 generates a control signal to cause the notification device 30 to notify the estimation result, and transmits the control signal to the notification device 30 via the output unit 208. The notification device 30 notifies the estimation result by the estimation device 100 based on the control signal output from the control device 200.

Figure 8 is a schematic diagram showing an example of a notification example. Figure 8 shows an example of the estimation result of the estimation device 100 displayed as an image. In this notification example, the center of gravity position of the estimation result is indicated by a hatched rectangular area.

Figure 9 is a schematic diagram showing another example of the notification example. Figure 9 shows an example in which the estimated center of gravity position and the recommended center of gravity position are displayed. In this notification example, the estimated center of gravity position is indicated by a hatched area, and the recommended center of gravity position is indicated by a cross. In one example, the recommended center of gravity position is calculated by the control device 200 using a function that derives the recommended center of gravity position based on the body dimensions, body weight, tilt angle, rider weight, height, etc. of the human-powered vehicle 1. In another example, the recommended center of gravity position is read by the control device 200 using a table that shows the relationship between the recommended center of gravity position and data such as the body dimensions, body weight, tilt angle, rider weight, height, etc. of the human-powered vehicle 1. The control device 200 notifies the rider of the recommended posture by notifying the rider of the recommended center of gravity position from the notification device 30.

As described above, the control device 200 of the second embodiment controls the components of the human-powered vehicle 1 based on the estimation results of the estimation device 100, and can therefore notify the rider of information including the center of gravity position of the estimation results, for example.

In the second embodiment, the component to be controlled is the alarm device 30, but the components to be controlled may include at least one of the lighting device 32, the braking device 34, the gear shift device 36, the suspension device 38, the drive assist device 40, and the adjustable seat post 42.

When the component to be controlled is the lighting device 32, the control unit 202 generates a control signal for controlling the operation of the lighting device 32 based on the estimation result of the estimation device 100. The control unit 202 outputs the generated control signal from the output unit 208 to the lighting device 32. For example, when the center of gravity position estimated by the estimation device 100 is not present in a predetermined area, the control unit 202 may warn the rider by controlling the lighting device 32 to blink.

When the component to be controlled is the brake device 34, the control unit 202 generates a control signal for controlling the operation of the brake device 34 based on the estimation result of the estimation device 100. The control unit 202 outputs the generated control signal from the output unit 208 to the brake device 34. For example, the control unit 202 performs control to change the braking force of the brake device 34 according to the longitudinal position of the center of gravity estimated by the estimation device 100.

When the component to be controlled is the transmission 36, the control unit 202 generates a control signal for controlling the operation of the transmission 36 based on the estimation result of the estimation device 100. The control unit 202 outputs the generated control signal from the output unit 208 to the transmission 36. For example, the control unit 202 performs control to change the gear ratio of the transmission 36 according to the longitudinal position of the center of gravity estimated by the estimation device 100.

When the component to be controlled is the suspension device 38, the control unit 202 generates a control signal for controlling the operation of the suspension device 38 based on the estimation result of the estimation device 100. The control unit 202 outputs the generated control signal from the output unit 208 to the suspension device 38. For example, the control unit 202 performs control to change the cushioning provided by the suspension device 38 according to the vertical position of the center of gravity estimated by the estimation device 100.

When the component to be controlled is the drive assist device 40, the control unit 202 generates a control signal for controlling the operation of the drive assist device 40 based on the estimation result of the estimation device 100. The control unit 202 outputs the generated control signal from the output unit 208 to the drive assist device 40. For example, the control unit 202 may perform control to change the assist force by the drive assist device 40 according to the fore-aft position of the center of gravity estimated by the estimation device 100.

When the component to be controlled is the adjustable seatpost 42, the control unit 202 generates a control signal to control the operation of the adjustable seatpost 42 based on the estimation result of the estimation device 100. The control unit 202 outputs the generated control signal from the output unit 208 to the adjustable seatpost 42. For example, the control unit 202 may control the operation of the adjustable seatpost 42 to change the height position of the saddle 18 according to the vertical position of the center of gravity estimated by the estimation device 100.

Third Embodiment
In the third embodiment, a method for generating a learning model 120 will be described.

The learning model 120 used in the estimation device 100 of the human-powered vehicle 1 is generated, for example, in a server device 300. FIG. 10 is a block diagram showing the internal configuration of the server device 300. The server device 300 includes a control unit 302, a storage unit 304, an input unit 306, and a communication unit 308.

The control unit 302 includes, for example, a CPU, a ROM, and a RAM. The ROM included in the control unit 302 stores control programs and the like for controlling the operation of each piece of hardware included in the server device 300. The CPU in the control unit 302 executes the control programs stored in the ROM and various programs stored in the memory unit 304, and controls the operation of each piece of hardware.

The control unit 302 is not limited to the above-mentioned configuration. The control unit 302 is not limited to a configuration including a CPU, ROM, and RAM. The control unit 302 may be, for example, one or more control circuits or arithmetic circuits including a GPU, FPGA, DSP, quantum processor, volatile or non-volatile memory, etc. The control unit 302 may also have functions such as a clock that outputs date and time information, a timer that measures the elapsed time from when a measurement start instruction is given to when a measurement end instruction is given, and a counter that counts numbers.

The storage unit 304 includes a storage device such as a flash memory or a hard disk drive. The storage unit 304 stores various computer programs executed by the control unit 302, teacher data used to generate the learning model, the learning model generated by the server device 300, and the like.

The input unit 306 has an input interface for acquiring data and programs from a storage medium that stores various data or programs. The various data and programs input through the input unit 306 are stored in the storage unit 304.

The communication unit 308 has a communication interface. For example, the communication unit 308 receives information sent from an external device addressed to the server device 300, and transmits information to be sent to the external device.

In this embodiment, for the sake of simplicity, the server device 300 is described as a single device, but it may be configured with multiple server devices, or may be configured with one or multiple virtual machines.

Figure 11 is a conceptual diagram showing an example of teacher data. The teacher data includes vehicle body information related to the vehicle body of the human-powered vehicle 1, load information related to the load acting on the human-powered vehicle 1, and center of gravity information indicating the center of gravity position of the human-powered vehicle 1. The vehicle body information includes information related to the vehicle body dimensions, vehicle body weight, and tilt angle of the human-powered vehicle 1. Design values or values measured in advance are used for the vehicle body dimensions and vehicle body weight of the human-powered vehicle 1. A value measured in advance using the tilt sensor S1 is used for the tilt angle of the human-powered vehicle 1.

The load information used to estimate the center of gravity position includes information about the load acting on at least one of the front wheel axle 12A, the rear wheel axle 14A, the handlebar 16, the saddle 18, the crank 22, and the pedal 28. This information is obtained by using values measured in advance using the load sensor S2.

The load information used to estimate the center of gravity position may include at least one of the rider's weight, height, posture, and gender. The rider's weight, height, and gender are determined by values set by an administrator or the like. The rider's posture is determined by a value measured in advance using the posture sensor S3.

A value measured in advance is used for the center of gravity information indicating the center of gravity position. A known method is used to measure the center of gravity position. As an example, a method of measuring the center of gravity position of the human-powered vehicle 1 with a rider riding on it by image analysis will be described. When measuring the center of gravity position in the X-axis direction and the Z-axis direction described in the first embodiment, a captured image captured from the Y-axis direction is used. If the position coordinates of each pixel in the captured image are P _ij and the mass of the minute part at the position coordinate P _ij is m _ij , the center of gravity position is calculated by (m _ij P _ij )/M. Here, M is the sum of the vehicle body weight of the human-powered vehicle 1 and the weight of the rider. The mass m _ij of the minute part at the position coordinate P _ij may be given by a table, or may be given depending on which part of the human-powered vehicle 1 or the rider the minute part belongs to.

In this embodiment, center of gravity information is measured when the vehicle body information and load information of the human-powered vehicle 1 are changed in various ways. The server device 300 acquires the measured center of gravity information and the vehicle body information and load information used when measuring the center of gravity information, associates these and stores them in the memory unit 304, and uses them as training data.

Figure 12 is a flowchart explaining the procedure for generating a learning model by the server device 300. In step S301, the control unit 302 of the server device 300 acquires training data for training the learning model from the memory unit 304. In the initial stage of generating the learning model, the training data may be set, for example, by an administrator of the server device 300. As learning progresses, the training data may be set using the estimation results of the learning model. In the latter case, the input to the learning model and the label data obtained as the estimation result can be set as training data.

In step S302, the control unit 302 inputs the vehicle body information and load information of the human-powered vehicle 1 to the learning model to be learned. Before learning begins, initial settings are given to the definition information describing the learning model. The learning model, like the learning model 120 described in the first embodiment, has an input layer, an intermediate layer, and an output layer. The input layer of the learning model is given vehicle body information and load information. The information given to the input layer is output to the adjacent intermediate layer nodes. In the intermediate layer, calculations are performed using activation functions including weights and biases between nodes, and the calculation results by the intermediate layer are output to each node in the output layer.

In step S303, the control unit 302 obtains the computation results of the learning model from each node in the output layer.

In step S304, the control unit 302 evaluates the calculation result obtained from the output layer of the learning model. In step S305, the control unit 302 determines whether learning of the learning model is complete. Specifically, the control unit 302 can evaluate the calculation result using an error function based on the calculation result obtained in step S303 and the teacher data. The error function is also called an objective function, a loss function, or a cost function. In the process of optimizing the error function by a gradient descent method such as the steepest descent method, the control unit 302 determines that learning of the learning model is complete when the error function becomes equal to or smaller than a threshold value or equal to or larger than a threshold value. In order to avoid the problem of overfitting, the control unit 302 may adopt a method such as cross-validation or early termination to terminate learning at an appropriate time.

If it is determined in step S305 that learning is not complete, the control unit 302 updates the weights and biases between nodes used in the intermediate layer in step S306, and returns the process to step S301. The control unit 302 can update the weights and biases between each node using the backpropagation method, which sequentially updates the weights and biases between nodes from the output layer to the input layer.

If it is determined in step S305 that the learning is complete, the control unit 302 stores the learned learning model in the memory unit 304 in step S307, and ends the processing according to this flowchart.

As described above, in the third embodiment, the server device 300 can be used to generate a learning model that uses the body information and load information of the human-powered vehicle 1, as well as center of gravity information indicating the center of gravity position of the human-powered vehicle 1, as training data, and outputs an estimation result regarding the center of gravity position in response to the input of the body information and load information.

The server device 300 may provide the trained learning model to the estimation device 100 as necessary. For example, if the estimation device 100 has a communication interface for communicating with the server device 300, the server device 300 may provide the trained learning model to the estimation device 100 by communication. The server device 300 may write the trained learning model to a storage medium M by a writing device not shown in the figure, and provide the learning model written to the storage medium M to the estimation device 100. The trained learning model provided by communication or the storage medium M is stored in the storage unit 106 of the estimation device 100. The learning model stored in the storage unit 106 of the estimation device 100 functions as the learning model 120 described in the first embodiment, and is configured to output an estimation result regarding the center of gravity position of the human-powered vehicle 1 in response to input of vehicle body information and load information of the human-powered vehicle 1.

The embodiments disclosed herein should be considered in all respects as illustrative and not restrictive. The scope of the present invention is indicated by the claims, not by the above meaning, and is intended to include all modifications within the meaning and scope of the claims.

1...human-powered vehicle, 30...alarm device, 32...lighting device, 34...brake device, 36...gearbox, 38...suspension device, 40...drive assist device, 42...adjustable seat post, 100...estimation device, 102...input unit, 104...arithmetic processing unit, 106...storage unit, 108...output unit, 110...estimation processing program, 120...learning model, 200...control device, 202...control unit, 204...storage unit, 206...input unit, 208...output unit, 300...server device, 302...control unit, 304...storage unit, 306...input unit, 308...communication unit, S1...tilt sensor, S2...load sensor, S3...posture sensor

Claims (14)

estimating a center of gravity position of the human-powered vehicle with a rider on the basis of vehicle body information relating to a vehicle body of the human-powered vehicle and load information including a load acting on the human-powered vehicle;
the vehicle body information includes information on at least one of a vehicle body dimension, a vehicle body weight, and an inclination angle of the human-powered vehicle;
The load is a measurement value measured by a load sensor mounted on the human-powered vehicle . An estimation device that estimates the center of gravity position using a learning model that is configured to output an estimation result regarding the center of gravity position of the human-powered vehicle with a rider on board, based on vehicle body information regarding the body of the human-powered vehicle and load information regarding the load acting on the human-powered vehicle, in response to input of the vehicle body information and the load information. The estimation device according to claim 2, wherein the vehicle body information includes information regarding the vehicle body dimensions, vehicle body weight, and tilt angle of the human-powered vehicle. The human-powered vehicle includes a front axle, a rear axle, a saddle, a handlebar, a crank, and pedals;
The estimation device according to claim 1 or 2, wherein the load information includes information regarding a load acting on at least one of the front wheel axle, the rear wheel axle, the saddle, the handlebars, the cranks, and the pedals. estimating a center of gravity position of the human-powered vehicle with a rider on the basis of vehicle body information relating to a vehicle body of the human-powered vehicle and load information including a load acting on the human-powered vehicle;
the load is a measurement value measured by a load sensor mounted on the human-powered vehicle,
The load information further includes at least one of the rider's weight, height, posture, and gender. The estimation device according to any one of claims 1 to 5,
and a control device that controls components of the human-powered vehicle based on the center of gravity position. The control system of claim 6, wherein the component includes an alarm device. The control system according to claim 7, wherein the control device controls the notification device to notify the center of gravity position. The control system according to claim 7, wherein the control device controls the notification device to notify information related to the recommended posture of the rider. The control system of any one of claims 6 to 9, wherein the components include at least one of a lighting device, a braking device, a transmission device, a suspension device, a drive assist device, and an adjustable seat post. an input layer to which vehicle body information relating to a vehicle body of a human-powered vehicle and load information relating to a load acting on the human-powered vehicle are input;
an output layer that outputs an estimation result regarding the center of gravity position of the human-powered vehicle with a rider on board;
an intermediate layer that uses the vehicle body information, the load information, and center of gravity information indicating the center of gravity position as teacher data to learn a relationship between the vehicle body information and the load information input to the input layer and the estimation result output from the output layer;
A learning model, wherein a computer is configured to perform calculation processing in the intermediate layer to output the estimation result from the output layer in response to input of the vehicle body information and the load information to the input layer. A method for generating a learning model using, as training data, vehicle body information relating to the body of a human-powered vehicle, load information relating to the load acting on the human-powered vehicle, and center of gravity information indicating the center of gravity position of the human-powered vehicle when a rider is riding on the vehicle, and using a computer to generate a learning model that outputs an estimation result regarding the center of gravity position in response to input of the vehicle body information and the load information. A computer program for causing a computer to execute a process of estimating the position of the center of gravity of a human-powered vehicle with a rider on board, based on vehicle body information including information regarding at least one of vehicle body dimensions, vehicle weight, and tilt angle for the body of the human-powered vehicle, and load information including measurement values obtained by measuring the load acting on the human-powered vehicle using a load sensor mounted on the human-powered vehicle. A storage medium on which the computer program according to claim 13 is stored.

Images