EVALUATION DEVICE, CONTROL SYSTEM, NOTIFICATION SYSTEM, METHOD FOR GENERATING LEARNING MODEL, COMPUTER PROGRAM, AND STORAGE MEDIUM

Abstract

To provide an evaluation device, a control system, a notification system, a generation method for learning model, computer program and storage medium, used for a bicycle including an electrically-driven component.SOLUTION: The evaluation device evaluates the favorability of control for components of a human-power drive vehicle using a learning model according to travel information concerning travel of the human-power drive vehicle. The evaluation device highly evaluates such control that the traction of the human-power drive vehicle becomes better and evaluates low such control that traction loss is estimated.SELECTED DRAWING: Figure 4

Description

[Technical field]

[0001] The present invention relates to an evaluation device, a control system, a notification system, a method for generating a learning model, a computer program, and a storage medium. [Background technology]

[0002] Bicycle control systems that include electric components are known. For example, Patent Document 1 discloses a control system in which an operating device, a control device, and an electric component are wirelessly connected. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2016-165991 A Summary of the Invention [Problem to be solved by the invention]

[0004] Currently, there is a demand for evaluating the control of bicycle control systems.

[0005] The present invention aims to provide an evaluation device that evaluates the desirability of control of components of a human-powered vehicle, a control system, a notification system, a method for generating a learning model, a computer program, and a storage medium. [Means for solving the problem]

[0006] An evaluation device according to a first aspect of the present invention evaluates desirability of control of components of a human-powered vehicle by using a learning model, in accordance with travel information relating to the travel of the human-powered vehicle.

[0007] This evaluation device can use a learning model to evaluate the desirability of control of components in accordance with driving information of the human-powered vehicle. Here, the desirability of control is information that suggests the quality of the state of the human-powered vehicle when the component is controlled. The desirability of control is evaluated, for example, based on the quality of traction of the human-powered vehicle. Therefore, the evaluation device highly evaluates control that improves traction of the human-powered vehicle and low evaluates control that is estimated to result in a loss of traction.

[0008] The evaluation device according to the second aspect related to the first aspect of the present invention re-learns the learning model in response to the evaluation result of the desirability and the control result of the component in response to the evaluation result.

[0009] According to this evaluation device, the learning model is re-trained in accordance with the evaluation result of the evaluation device and the control result of the component, so that a more preferable control state can be realized as the learning progresses.

[0010] In the evaluation device of a third aspect related to the first or second aspect of the present invention, the driving information includes information on at least one of speed, acceleration, posture, load, driving force, cadence, gear ratio, operation input to an operation device, history of the operation input and control results of the components, regeneration amount, inclination, road surface conditions, position, weather, and vehicle information.

[0011] This evaluation device can evaluate the desirability of control over components according to at least one of speed, acceleration, posture, load, driving force, cadence, gear ratio, operation input to an operating device, history of operation input and control results of components, regeneration amount, inclination, road surface conditions, position, weather, and vehicle information.

[0012] In the evaluation device according to a fourth aspect relating to the first or second aspect of the present invention, the travel information of the human-powered vehicle includes information relating to traction of the human-powered vehicle.

[0013] This evaluation device makes it possible to evaluate the desirability of control over the components according to information about the traction of the human-powered vehicle.

[0014] A control system according to a fifth aspect of the present invention includes the evaluation device according to any one of the first to fourth aspects, and a control device that controls the component based on an evaluation result of the evaluation device.

[0015] According to this control system, the control device controls the components based on the evaluation results from the evaluation device, so that a more preferable control state can be achieved without depending on the operation of the rider.

[0016] In a control system according to a sixth aspect relating to the fifth aspect of the present invention, the component includes an assist device having an electric drive unit that outputs auxiliary drive force.

[0017] According to this control system, the control device controls the assist device that outputs the auxiliary drive force, so that a more preferable control state can be realized with respect to the auxiliary drive control.

[0018] A seventh aspect of the control system according to the fifth or sixth aspect of the present invention, wherein the component includes a brake device having an electric drive unit.

[0019] According to this control system, the control device controls the brake device, so that a more preferable control state can be realized with respect to the brake control.

[0020] In an eighth aspect of the control system according to any one of the fifth to seventh aspects of the present invention, the component includes an anti-lock braking system having an electric drive unit.

[0021] According to this control system, the control device controls the antilock brake device, so that a more preferable control state can be achieved with respect to the antilock brake control.

[0022] In a ninth aspect of the control system according to any one of the fifth to eighth aspects of the present invention, the component includes a suspension device having an electric drive unit.

[0023] According to this control system, the control device controls the suspension device, so that a more preferable control state can be achieved with respect to the suspension control.

[0024] A notification system according to a tenth aspect of the present invention includes the evaluation device according to any one of the first to fourth aspects, and a notification device that notifies information relating to an evaluation result of the evaluation device.

[0025] According to this notification system, since it is provided with a notification device that notifies information related to the evaluation result, the rider can know whether the control state of the human-powered vehicle is good or bad.

[0026] In an information system according to an eleventh aspect related to the tenth aspect of the present invention, the information on the evaluation result includes information on a control state of the component.

[0027] According to this notification system, the information on the evaluation result includes information on the control state of the component, so that the rider can know whether the control state of the component is good or bad.

[0028] A method for generating a learning model according to a twelfth aspect of the present invention uses a computer to evaluate the desirability of control of components of a human-powered vehicle using a function in accordance with driving information regarding the driving of the human-powered vehicle, and generates a learning model by updating the function in accordance with the evaluation result of the desirability and the control result of the component in accordance with the evaluation result.

[0029] According to this method for generating a learning model, a learning model for evaluating the desirability of component control can be generated by collecting driving information of a human-powered vehicle.

[0030] A computer program according to a thirteenth aspect of the present invention is a computer program for causing a computer to execute a process of evaluating, using a learning model, the desirability of control of components of a human-powered vehicle in accordance with driving information related to the driving of the human-powered vehicle.

[0031] According to this computer program, an execution environment for evaluating the desirability of control of components in accordance with driving information of a human-powered vehicle can be realized by a computer.

[0032] A storage medium according to a fourteenth aspect of the present invention stores the computer program according to the thirteenth aspect.

[0033] According to this storage medium, by installing the computer program stored in the storage medium in a computer, an execution environment for evaluating the desirability of control of components in accordance with driving information of a human-powered vehicle can be realized by the computer. Effect of the Invention

[0034] According to the present application, the desirability of control over components of a human-powered vehicle can be evaluated. [Brief description of the drawings]

[0035] [Figure 1] 1 is a side view of a human-powered vehicle to which the evaluation device of the first embodiment is applied. [Diagram 2] 1 is a block diagram showing an internal configuration of an evaluation device according to a first embodiment. [Diagram 3] FIG. 1 is a schematic diagram showing an implementation example of a learning model. [Figure 4] 11 is a flowchart illustrating a learning procedure of a learning model. [Diagram 5] FIG. 11 is a block diagram illustrating a control system according to a second embodiment. [Figure 6] 4 is a flowchart illustrating a procedure of a process executed by a control device. [Figure 7]11 is a flowchart illustrating a procedure for generating a learning model. [Figure 8] 11 is a flowchart illustrating a procedure for re-learning a learning model. [Figure 9] FIG. 13 is a block diagram illustrating a notification system according to a fifth embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0036] Hereinafter, the present invention will be described in detail with reference to the drawings showing embodiments thereof. (First embodiment) FIG 1 is a side view of a human-powered vehicle 1 to which an evaluation 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.

[0037] The human-powered vehicle 1 comprises a vehicle body 10, a front wheel 12, a rear wheel 14, a handlebar 16, a seat 18, and a drive mechanism 20. In the following description, terms indicating the front-rear, left-right, and up-down directions are used based on the directions when a rider is seated on the seat 18 of the human-powered vehicle 1.

[0038] The vehicle body 10 includes a frame 10A and a front fork 10B. A front wheel 12 is rotatably supported by the front fork 10B. A rear wheel 14 is rotatably supported by the frame 10A. A handlebar 16 is supported by the frame 10A so that the traveling direction of the front wheel 12 can be changed.

[0039] The drive mechanism 20 transmits human-powered driving force to the rear wheel 14. The drive mechanism 20 includes a crank 22, a front sprocket assembly 24A, a rear sprocket assembly 24B, a chain 26, and a pair of pedals 28, 28. The drive mechanism 20 may further include a chain device for preventing the chain 26 from coming off.

[0040] 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 a pair of pedals 28, 28 is rotatably supported by the right crank 22A, and the other is rotatably supported by the left crank 22B.

[0041] The front sprocket assembly 24A is connected to the crankshaft 22C and rotates integrally with the crankshaft 22C. In one example, the front sprocket assembly 24A includes a plurality of front sprockets with different outer diameters. The outer diameters of the plurality of front sprockets increase from the center plane of the vehicle body 10 toward the outside in a direction parallel to the rotational axis of the crankshaft 22.

[0042] The rear sprocket assembly 24B is rotatably supported by a hub (not shown) of the rear wheel 14. In one example, the rear sprocket assembly 24B includes multiple rear sprockets with different outer diameters. The outer diameters of the multiple rear sprockets decrease from the center plane of the vehicle body 10 toward the outside in a direction parallel to the rotational axis of the hub assembly.

[0043] The chain 26 is wound around the front sprocket assembly 24A and the rear sprocket assembly 24B. When the crank 22 rotates forward due to human driving force applied to the pedals 28, 28, the front sprocket assembly 24A rotates forward together with the crank 22. The rotation of the front sprocket assembly 24A is transmitted to the rear sprocket assembly 24B via the chain 26, causing the rear wheel 14 to rotate. A belt or a shaft may be used instead of the chain 26.

[0044] The human-powered vehicle 1 further includes an operating device 30, transmissions 32 and 33, an assist device 34, a braking device 36, an anti-lock braking device 38, a suspension device 40, and a battery unit 42.

[0045] The operating device 30 is provided, for example, on the handlebar 16. The operating device 30 includes operating switches 30A and 30B that are operated by the rider's fingers. The operating switches 30A and 30B are switches for controlling at least one of the transmission device 32, the assist device 34, the brake device 36, the anti-lock brake device 38, and the suspension device 40.

[0046] The number of operation devices 30 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 operation device 30 is configured to include two operation switches 30A, 30B, but it may also be configured to include one or three or more operation switches. The operation device 30 is not limited to a configuration including a switch, and may also be configured to include an operation lever, an operation dial, or the like.

[0047] The operation device 30 is connected to each component so as to transmit a control signal corresponding to the operation of the operation switch 30A or the operation switch 30B to the component to be controlled. In one example, the operation device 30 is connected to the component to be controlled by a communication line or an electric wire capable of PLC (Power Line Communication). The component to be controlled is at least one of the transmission device 32, the assist device 34, the brake device 36, the antilock brake device 38, and the suspension device 40. In another example, the operation device 30 is connected to the component to be controlled by a wireless communication unit capable of wireless communication. The operation device 30 may further output information regarding the operation input to the evaluation device 100.

[0048] The transmission 32 is one of the components of the human-powered vehicle 1 whose operation is controlled in response to a control signal transmitted from the operation device 30. The transmission 32 includes an electric actuator 32A and a chain guide 32B. An example of the electric actuator 32A is an electric motor. The transmission 32 drives the electric actuator 32A to operate the chain guide 32, thereby changing the rear sprocket around which the chain 26 is wound, and switching the gear ratio of the human-powered vehicle 1.

[0049] An example of the transmission 32 is an external transmission. When the transmission 32 is an external transmission, the rear sprocket assembly 24B includes a plurality of rear sprockets with different outer diameters. When the transmission 32 receives a control signal from the operation device 30 instructing an upshift, the transmission 32 operates the chain guide 32B by the electric actuator 32A so as to change the rear sprocket around which the chain 26 is wound to a rear sprocket with a smaller outer diameter than the current rear sprocket. When the transmission 32 receives a control signal from the operation device 30 instructing a downshift, the transmission 32 operates the chain guide 32B by the electric actuator 32A so as to change the rear sprocket around which the chain 26 is wound to a rear sprocket with a larger outer diameter than the current rear sprocket.

[0050] The transmission 33 is one of the components of the human-powered vehicle 1 whose operation is controlled in response to a control signal transmitted from the operation device 30. The transmission 33 includes an electric actuator 33A and a chain guide 33B. An example of the electric actuator 33A is an electric motor. The transmission 33 operates the chain guide 33B by driving the electric actuator 33A, thereby changing the front sprocket around which the chain 26 is wound, and switching the gear ratio of the human-powered vehicle 1.

[0051] An example of the transmission 33 is an external transmission. When the transmission 33 is an external transmission, the front sprocket assembly 24A includes a plurality of front sprockets with different outer diameters. In this embodiment, when the transmission 33 receives a control signal from the operation device 30 instructing an upshift, the transmission 33 operates the chain guide 33B by the electric actuator 33A so as to change the front sprocket around which the chain 26 is wound to a sprocket with a larger outer diameter than the current sprocket. When the transmission 33 receives a control signal from the operation device 30 instructing a downshift, the transmission 33 operates the chain guide 33B by the electric actuator 33A so as to change the front sprocket around which the chain 26 is wound to a front sprocket with a smaller outer diameter than the current front sprocket.

[0052] The transmissions 32 and 33 may be internal transmissions. For example, when the transmission 32 is an internal transmission, the transmission 32 is provided, for example, in the hub of the rear wheel 14, and transmits the rotation input to the rear sprocket assembly 24B to the rear wheel 14 after changing the speed. In this case, the rear sprocket assembly 24B includes a single rear sprocket. At least one of the transmissions 32 and 33 may be a continuously variable transmission.

[0053] The assist device 34 is provided for the drive mechanism 20. The assist device 34 has an electric drive unit 34A. The electric drive unit 34A includes a drive circuit, a memory, an electric motor, etc., and the drive circuit operates the electric motor to output an auxiliary drive force, thereby assisting the running of the human-powered vehicle 1. In this embodiment, the assist device 34 is one of the components automatically controlled by a control device 200 described later. The control device 200 is shown in FIG. 5. The assist device 34 may be controlled according to a control signal output from the operation device 30. The control state of the assist device 34 includes a plurality of control states with different auxiliary drive forces. The control state in the assist device 34 is also called an assist level. The assist device 34 changes the assist level according to, for example, the human-powered drive force acting on the crank 22. 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, the motor unit 34 is not driven and no assist is performed.

[0054] The assist device 34 may be configured to function as a regenerative braking device that brakes the human-powered vehicle 1. The control result of the assist device 34, including the amount of regeneration, may be output to the evaluation device 100 and stored as history in the storage unit 106 of the evaluation device 100. The storage unit 106 of the evaluation device 100 is shown in FIG. 2.

[0055] The brake device 36 is provided for the front wheel 12 and the rear wheel 14. The brake device 36 provided for the front wheel 12 is a rim brake including an electric drive unit 36A and a caliper 36B including brake pads that can contact the wheel of the front wheel 12. The electric drive unit 36A of the brake device 36 includes a drive circuit and an electric motor. The brake device 36 is one of the components that are automatically controlled by the control device 200. The brake device 36 may be connected to a brake operation device 36C provided on the handlebar 16. In response to a control signal from the control device 200 or an operation of the brake operation device 36C, the drive circuit of the electric drive unit 36A operates the caliper 36B by the electric motor and brings the brake pads into contact with the wheel of the front wheel 12, thereby attenuating the rotational force of the front wheel 12. The brake device 36 can operate in a plurality of control states with different braking forces. The control result of the braking device 36 may be output to the evaluation device 100 and stored as a history in a storage unit 106 of the evaluation device 100. The braking device 36 may output information related to an operation input to the brake operation device 36C to the evaluation device 100. The information related to the operation input includes, for example, a method of braking, an operation amount, and information indicating whether the left or right braking device 36 is being operated.

[0056] The brake device 36 provided on the rear wheel 14 is a rim brake including an electric drive unit 36A and a caliper 36B including brake pads that can contact the wheel of the rear wheel 14. The electric drive unit 36A of the brake device 36 includes a drive circuit and an electric motor. The brake device 36 is one of the components that are automatically controlled by the control device 200. The brake device 36 may be connected to a brake operation device 36C provided on the handlebar 16. In response to a control signal from the control device 200 or an operation of the brake operation device 36C, the drive circuit of the electric drive unit 36A operates the caliper 36B by the electric motor, and the brake pads are brought into contact with the wheel of the rear wheel 14, thereby attenuating the rotational force of the rear wheel 14. The brake device 36 can operate in a plurality of control states with different braking forces. The control result of the brake device 36 may be output to the evaluation device 100 and stored as history in the memory unit 106 of the evaluation device 100. The braking device 36 may output, to the evaluation device 100, information regarding an operation input to the brake operating device 36C.

[0057] The brake device 36 may be a disc brake including brake pads that can contact a rotating body attached to the hub of the front wheel 12 and the hub of the rear wheel 14. In the disc brake, for example, an electric motor controls fluid to move the brake pads and press the brake pads against a rotating body (a disc brake rotor), thereby attenuating the rotational force of the front wheel 12 or the rear wheel 14. Alternatively, the disc brake attenuates the rotational force of the front wheel 12 or the rear wheel 14 by directly moving the brake pads with an electric motor and pressing the brake pads against the rotating body.

[0058] The antilock brake device 38 is provided for the brake device 36. The antilock brake device 38 includes an electric drive unit 38A including a drive circuit, a reservoir for storing hydraulic oil, a pump for supplying hydraulic oil and applying hydraulic pressure to the caliper 36B, and a valve for adjusting the hydraulic pressure applied to the caliper 36B of the brake device 36. The electric drive unit 38A controls the opening and closing of the valve and adjusts the hydraulic pressure applied to the caliper 36B, thereby controlling the braking force applied from the caliper 36B to the front wheel 12 or the rear wheel 14. The antilock brake device 38 is one of the components automatically controlled by the control device 200. The antilock brake device 38 can operate in a plurality of control states in which the magnitude of the hydraulic pressure applied to the caliper 36B is different. The control result of the antilock brake device 38 is output to the evaluation device 100 and stored as history in the memory unit 106 of the evaluation device 100.

[0059] In one example, the suspension device 40 is a front suspension provided on the front fork 10B, which attenuates an impact applied to the front wheel 12. In another example, the suspension device 40 is a rear suspension which attenuates an impact applied to the rear wheel 14. The suspension device 40 includes an electric drive unit 40A. The electric drive unit 40A includes a drive circuit and an electric motor. The suspension device 40 is controlled by setting, as operation parameters, for example, a damping rate, a repulsive force, a stroke amount, and a lockout state. The suspension device 40 is one of the components automatically controlled by the control device 200. The suspension device 40 can operate in a plurality of control states with different cushioning properties. The control result of the suspension device 40 is output to the evaluation device 100 and stored as history in the memory unit 106 of the evaluation device 100.

[0060] The battery unit 42 includes a battery 42A and a battery holder 42B. The battery 42A is a storage battery including one or more battery cells. The battery holder 42B is fixed to the frame 10A of the human-powered vehicle 1. The battery 42A is detachably attached to the battery holder 42B. The battery 42A is electrically connected to the transmission 32, the assist device 34, the brake device 36, the anti-lock brake device 38, the suspension device 40, and the like.

[0061] The human-powered vehicle 1 may further include at least one of a vehicle speed sensor S1, an acceleration sensor S2, an attitude sensor S3, a load sensor S4, a torque sensor S5, a cadence sensor S6, an inclination sensor S7, an imaging sensor S8, a position sensor S9, and a weather sensor S10. In the following description, when there is no need to describe each sensor individually, they will also be referred to as sensors S1 to S10.

[0062] The vehicle speed sensor S1 is a sensor that outputs a signal corresponding to the speed of the human-powered vehicle 1. The vehicle speed sensor S1 is attached to the frame 10A or the front fork 10B. The vehicle speed sensor S1 includes, for example, a Hall element, and measures the rotation speed of the front wheel 12 or the rear wheel 14 by detecting a magnet (not shown) provided on the front wheel 12 or the rear wheel 14. In one example, the vehicle speed sensor S1 outputs a signal indicating the rotation speed of the front wheel 12 or the rear wheel 14. In another example, the vehicle speed sensor S1 calculates the speed of the human-powered vehicle 1 based on the measured rotation speed of the front wheel 12 or the rear wheel 14, and outputs a signal indicating the calculated speed.

[0063] The acceleration sensor S2 is a sensor that outputs a signal corresponding to the acceleration in the forward / rearward direction of the human-powered vehicle 1. The acceleration sensor S2 is provided, for example, on the frame 10A, and measures acceleration by detecting the displacement of a weight as a change in capacitance or resistance value when acceleration in the forward / rearward direction is applied. The acceleration sensor S2 may be a motion sensor that measures acceleration in the yaw, roll, and pitch directions of the human-powered vehicle 1.

[0064] 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 grip of the handlebar 16, the surface of the seat 18, the pedals 28, etc. The posture sensor S3 may be a motion sensor built into a wearable terminal attached to the rider.

[0065] The load sensor S4 is a sensor that outputs a signal according to the load applied to the human-powered vehicle 1. An example of the load sensor S4 is a strain gauge type load cell. The load sensor S4 is provided at one or more locations on the human-powered vehicle 1, such as the axle of the front wheel 12 or the rear wheel 14, the handlebar 16, the seat 18, the pedals 28, etc. The load sensor S4 may be a sensor that measures the ground load of the front wheel 12 and the rear wheel 14, or may be a sensor that measures the load balance of the human-powered vehicle 1. At least a part of the attitude sensor S3 and the load sensor S4 may be shared.

[0066] The torque sensor S5 is a sensor that outputs a signal corresponding to the torque applied to the crank 22. The torque sensor S5 is provided in any of the right crank 22A, the left crank 22B, the crankshaft 22C, the front sprocket assembly 24A, or the drive path from the crankshaft 22C to the front sprocket assembly 24A. The torque sensor S5 includes a strain sensor, a magnetostrictive sensor, an optical sensor, a pressure sensor, or the like, and detects the torque applied to the crank 22.

[0067] The cadence sensor S6 is a sensor that outputs a signal indicating the cadence. The cadence sensor S6 is attached to the frame 10A. The cadence sensor S6 includes, for example, a Hall element, and measures the number of rotations per unit time of the crankshaft 22C by detecting a magnet (not shown) provided on the right crank 22A or the left crank 22B.

[0068] The evaluation device 100 can calculate the power applied to the human-powered vehicle 100 based on the torque applied to the crank 22 detected by the torque sensor S5 and the number of rotations per unit time of the crankshaft 22C detected by the cadence sensor S6.

[0069] The inclination sensor S7 is a sensor that outputs a signal according to the inclination angle of the human-powered vehicle 1. The inclination angle detected by the inclination sensor S7 is, for example, the rotation angle around a pitch axis along the left-right direction of the human-powered vehicle 1. As an example, the inclination sensor S7 includes a sensor that detects the angular velocity of the pitch angle, and calculates the value obtained by integrating the angular velocity around the pitch axis as the pitch angle. The inclination sensor S7 may be configured to measure both the rotation angle around a roll axis along the front-rear direction of the human-powered vehicle 1 and the rotation speed around a yaw axis along the up-down direction of the human-powered vehicle 1.

[0070] The imaging sensor S8 is a sensor that outputs a video signal obtained by capturing an image of the surroundings of the human-powered vehicle 1. The imaging sensor S8 is provided, for example, at an appropriate location on the frame 10A or the handlebar 16 so that the road surface is included in the imaging range. The imaging sensor S8 may be a wearable camera attached to the rider.

[0071] The position sensor S9 is a sensor that outputs a signal indicating the current position of the human-powered vehicle 1. The position sensor S9 is attached to the frame 10A or the handlebar 16. The position sensor S9 includes, for example, a GPS (Global Positioning System) communication device, and measures the current position of the human-powered vehicle 1 by receiving radio waves from a GPS satellite. The position sensor S9 may be equipped with a memory that successively records the measured position information, and may be configured to output information about the route that the human-powered vehicle 1 has traveled based on the position information recorded in the memory.

[0072] The weather sensor S10 is a sensor that measures the weather around the human-powered vehicle 1 and outputs a signal according to the measurement result. The weather sensor S10 is attached to the frame 10A and measures at least one of wind direction, wind speed, outside air temperature, and outside air humidity. The weather sensor S10 may be configured to communicate with an external weather server (not shown) and obtain information on at least one of wind direction, wind speed, outside air temperature, and outside air humidity around the human-powered vehicle 1 from the weather server.

[0073] The evaluation device 100 is a type of computer. In one example, the evaluation device 100 is a dedicated terminal such as a cycle computer provided in the human-powered vehicle 1. In another example, the evaluation 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.

[0074] The evaluation device 100 acquires travel information related to the travel of the human-powered vehicle 1. The travel information acquired by the evaluation device 100 includes at least one of information on speed, acceleration, attitude, load, driving force, cadence, gear ratio, operation input to the operation device 30, history of operation input and component control results, regeneration amount, inclination, road surface condition, position, weather, and vehicle information. The travel information may include information on traction of the human-powered vehicle 1. The evaluation device 100 can acquire the travel information from sensors S1 to S10 or components mounted on the human-powered vehicle 1. The travel information acquired by the evaluation device 100 may include not only information directly obtained from the sensors S1 to S10 or components, but also information derived by calculation, and may include information stored in the storage unit 106 of the evaluation device 100. For example, the travel information stored in the storage unit 106 may include information on the braking distance in response to an operation input to the brake operation device 36C, and may include vehicle information such as frame dimensions, the weight of the human-powered vehicle 1, and tire diameter.

[0075] The evaluation device 100 evaluates the desirability of control over the components of the human-powered vehicle 1 using a learning model 120 according to travel information related to the travel of the human-powered vehicle 1. Here, the desirability of control is information that suggests the good or bad state of the human-powered vehicle 1 when control over the components is executed. The evaluation device 100 can evaluate the desirability of control based on, for example, the good or bad traction of the human-powered vehicle 1. The evaluation device 100 may highly evaluate control that improves the traction of the human-powered vehicle 1, and may lowly evaluate control that is estimated to reduce traction. The learning model 120 used by the evaluation device 100 when evaluating the desirability of control over the components will be described in detail with reference to FIG. 3.

[0076] 2 is a block diagram showing the internal configuration of the evaluation device 100 according to the first embodiment. The evaluation device 100 includes an input unit 102, a calculation processing unit 104, a storage unit 106, and an output unit .

[0077] The input unit 102 includes an interface that connects at least one of the operation device 30, the gear shift device 32, the assist device 34, the brake device 36, the anti-lock brake system (ABS) device 38, the suspension device 40, and the sensors S1 to S10. The interface included in the input unit 102 is, for example, a wired interface, and connects the operation device 30 and the like 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 included in the input unit 102 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.

[0078] The calculation processing unit 104 includes a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), and the like. 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 evaluation device 100. The CPU in the calculation processing unit 104 executes computer programs stored in the ROM or the storage unit 106 and controls the operation of each part of the hardware, thereby implementing a process of evaluating the desirability of control over the components of the human-powered vehicle 1 using a learning model 120 in accordance with travel information related to the travel of the human-powered vehicle 1. The RAM included in the calculation processing unit 104 temporarily stores data used during the execution of the calculations.

[0079] The arithmetic processing unit 104 is configured to include a CPU, a ROM, and a 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, a volatile or non-volatile memory, etc.

[0080] 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 evaluation processing program 110 executed by the arithmetic processing unit 104, a learning model 120 (described later), and the like.

[0081] The evaluation processing program 110 is a computer program for causing a computer to execute processing for evaluating the desirability of control over the components of the human-powered vehicle 1 using a learning model 120 in accordance with travel information related to the travel of the human-powered vehicle 1. The calculation processing unit 104 executes the evaluation processing program 110 to evaluate the desirability of control over the components of the human-powered vehicle 1 using the learning model 120 in accordance with the travel information of the human-powered vehicle 1.

[0082] Various computer programs including the evaluation processing program 110 may be provided by a storage medium M in 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 various computer programs including the evaluation processing program 110. The calculation processing unit 104 reads the various computer programs from the storage medium M using a reading device, and installs the various computer programs that have been read into the storage unit 106.

[0083] The learning model 120 stored in the storage 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. The learning model 120 in this embodiment is used to evaluate the desirability of control over components of the human-powered vehicle 1 according to travel information related to the travel of the human-powered vehicle 1. In one example, the learning model 120 is a learning model constructed using DQN (Deep Q-Network). The learning model 120 is not limited to DQN, and may be a learning model constructed using DDQN (Double DQN), CDQN (Continuous DQN), or the like.

[0084] The output unit 108 includes an output interface that outputs the evaluation result by the arithmetic processing unit 104. The output format of the evaluation 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 evaluation result regarding the desirability of control 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 evaluation result regarding the desirability of control 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 evaluation result regarding the desirability of control to an external device connected to the communication interface.

[0085] 3 is a schematic diagram showing an implementation example of the learning model 120. The learning model 120 is a learning model constructed by, for example, DQN. In DQN, a multilayer neural network is used that takes a state s as input and outputs a number of action value function Q(s, a) values ​​corresponding to selectable actions a. In the following description, the value of the action value function Q(s, a) is also referred to as a Q value. The multilayer neural network that constitutes the learning model 120 includes, for example, an input layer 122, an intermediate layer 124, and an output layer 126.

[0086] The input layer 122 includes a plurality of nodes. The number of nodes in the input layer 122 is equal to the number of states s to be input. In this embodiment, the state s is driving information related to the traveling of the human-powered vehicle 1. The driving information is a state that is distinguished by at least one of the speed, acceleration, attitude, load, driving force, cadence, gear ratio, operation input to the operation device 30, history of operation input and control results of components, regeneration amount, inclination, road surface conditions, position, weather, and vehicle information.

[0087] The intermediate layer 124 is composed of multiple layers. Each layer of the intermediate layer 124 has multiple nodes. The nodes of each layer in the intermediate layer 124 are connected in one direction to nodes in the previous and next layers by desired weights and biases. The weights and biases that characterize the connection state between nodes are updated by learning. The intermediate layer 124 performs calculations using each node and outputs the calculation results to the output layer 126.

[0088] The output layer 126 includes a plurality of nodes. The number of nodes in the output layer 126 is equal to the number of actions a to be distinguished. In this embodiment, the actions a are distinguished according to the control state of one or more components of the human-powered vehicle 1. For example, the actions a are distinguished according to whether the brake device 36 of the human-powered vehicle 1 is in an activated state or not. Furthermore, the actions a are distinguished according to whether the left or right brake device 36 has been activated, how much force or speed the brake operation device 36C has been operated, and the like. The actions a may be distinguished not only according to the control state of the brake device 36, but also according to the control state of the assist device 34, the antilock brake device 38, or the suspension device 40.

[0089] Each node of the output layer 126 outputs a Q value for each action a. The Q value is the expected future profit when action a is taken in the state indicated by state s. The learning model 120 is trained so that the Q value is high in a control state where the traction of the human-powered vehicle 1 is good, and the Q value is low in a control state where the traction of the human-powered vehicle 1 is lost.

[0090] Next, an evaluation procedure using the learning model 120 will be described. FIG. 4 is a flowchart for explaining the learning procedure of the learning model 120. In step S101, the calculation processing unit 104 of the evaluation device 100 acquires driving information. Specifically, the calculation processing unit 104 acquires driving information including at least one of speed, acceleration, attitude, load, driving force, cadence, gear ratio, operation input to the operation device 30, history of operation input and control results of components, regeneration amount, inclination, road surface condition, position, weather, and vehicle information through the input unit 102. The driving information acquired by the calculation processing unit 104 may be information about the traction of the human-powered vehicle 1. The driving information may be not only information obtained by various sensors S1 to S10, but also information calculated by the calculation processing unit 104, or information stored in the storage unit 106.

[0091] In step S102, the calculation processing unit 104 executes calculations by the learning model 120 by inputting the traveling information acquired from the input unit 102 etc. to the input layer 122 of the learning model 120. The traveling information input to the input layer 122 represents a state s in the learning model 120 constructed by DQN. The traveling information input to the input layer 122 is provided to the intermediate layer 124. In the intermediate layer 124, at each node, a weight is multiplied for the input from the previous layer, a bias is added, and a calculation using an activation function is performed, and the calculation result is output to the output layer 126. The output layer 126 outputs a Q value through each node.

[0092] In step S103, the calculation processing unit 104 acquires the Q value obtained as the calculation result of the learning model 120 from the output layer 126.

[0093] In step S104, the calculation processing unit 104 evaluates the desirability of control over the components of the human-powered vehicle 1. The learning model 120 outputs a Q value for each of the actions a corresponding to the control states of the components from each node included in the output layer 126. The calculation processing unit 104 evaluates the desirability of control over the components by referring to the Q values ​​output from each node of the output layer 126. For example, the calculation processing unit 104 can evaluate the control state corresponding to the action a showing the highest value among the Q values ​​output from each node of the output layer 126 as the most desirable control.

[0094] As described above, the evaluation device 100 in the first embodiment can use the learning model 120 to evaluate the desirability of control over the components of the human-powered vehicle 1 in accordance with travel information related to the travel of the human-powered vehicle 1.

[0095] Second embodiment In the second embodiment, a control system that controls components of the human-powered vehicle 1 based on the evaluation results by the evaluation device 100 will be described.

[0096] Fig. 5 is a block diagram illustrating a control system according to the second embodiment. The control system according to the second embodiment includes an evaluation device 100 and a control device 200 that controls components of the human-powered vehicle 1 based on the evaluation results by the evaluation 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 Fig. 5, the evaluation device 100 and the control device 200 are described as separate entities, but the evaluation device 100 and the control device 200 may also be configured as an integrated unit.

[0097] 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 a control program for controlling the operation of each hardware unit included in the control device 200. The CPU in the control unit 202 executes the control program 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 a control signal for controlling a component of the human-powered vehicle 1 based on the judgment result from the evaluation device 100 input through the input unit 206, and outputs the generated control signal from the output unit 208.

[0098] 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 a measurement start instruction is given to when a measurement end instruction is given, and a counter that counts numbers.

[0099] The storage unit 204 includes memories such as an EEPROM, an SRAM, etc. The storage unit 204 stores computer programs executed by the control unit 202, data used by these computer programs, and the like.

[0100] The input unit 206 includes an interface that connects the evaluation device 100 via a cable. The evaluation result output from the evaluation device 100 is input to the input unit 206. The evaluation result input to the input unit 206 indicates the desirability of control over the components of the human-powered vehicle 1. The input unit 206 outputs the input evaluation result to the control unit 202. In this embodiment, the evaluation device 100 is connected via a cable, but the evaluation device 100 and the control unit 200 may be configured to transmit and receive data between them via a wireless communication interface. For wireless communication, a wireless communication method conforming to a communication standard such as Bluetooth, WiFi, ZigBee, LTE, or other wireless LAN can be used.

[0101] The output unit 208 includes an interface that connects the components to be controlled via a cable. The components may include the assist device 34 having an electric drive unit 34A that outputs an auxiliary drive force. The components may include the brake device 36 having an electric drive unit 36A. The components may include the anti-lock brake device 38 having an electric drive unit 38A. The components may include the suspension device 40 having an electric drive unit 40A. In this embodiment, the assist device 34, the brake device 36, the anti-lock brake device 38, and the suspension device 40 are connected to the control device 200. 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 a cable, but the control device 200 may be configured to transmit and receive data between the components of the human-powered vehicle 1 via a wireless communication interface. For wireless communication, a wireless communication method conforming to the various communication standards described above can be used.

[0102] Hereinafter, a control procedure in which the control device 200 controls the components of the human-powered vehicle 1 based on the evaluation results of the evaluation device 100 will be described.

[0103] 6 is a flowchart illustrating the procedure of the process executed by the control device 200. In step S201, the control unit 202 of the control device 200 acquires the evaluation result by the evaluation device 100. The evaluation result includes information on the behavior with the highest Q value.

[0104] In step S202, the control unit 202 generates a random number. The control unit 202 can generate a random number using a known method. The random number generated by the control unit 202 is a random number greater than or equal to 0 and less than or equal to 1.

[0105] In step S203, the control unit 202 compares the random number generated in step S202 with ε and determines whether the random number is greater than ε. Here, ε is a parameter for determining whether to use the Q value or to randomly select an action. ε is a value between 0 and 1, and may be a fixed value that is set in advance, or may be a variable value that decreases from 1 to 0 according to the elapsed time.

[0106] If the control unit 202 determines that the random number is greater than ε, then in step S204, it selects the action with the highest Q value.

[0107] If the control unit 202 determines that the random number is equal to or smaller than ε, then in step S205, the control unit 202 randomly selects an action.

[0108] In step S206, the control unit 202 controls the components of the human-powered vehicle 1 in accordance with the action selected in step S204 or step S205. The components to be controlled may be selected in accordance with the evaluation result of the evaluation device 100.

[0109] When the component to be controlled is the assist device 34, the control unit 202 outputs a control signal to the assist device 34. For example, when the assist level before the control is a weak assist level and the assist level corresponding to the action selected in step S204 or step S205 is a strong assist level, the control unit 202 outputs a control signal to the assist device 34 to transition the assist level from the weak assist level to the strong assist level. When the assist level before the control and the assist level corresponding to the action selected in step S204 or step S205 are the same, the control unit 202 does not control the assist device 34.

[0110] When the component to be controlled is the braking device 36, the control unit 202 outputs a control signal to the braking device 36. For example, when the control state before the control corresponds to a first braking force and the action selected in step S204 or step S205 corresponds to a second braking force, the control unit 202 outputs a control signal to transition the braking force of the braking device 36 from the first braking force to the second braking force. When the braking force before the control and the braking force corresponding to the action selected in step S204 or step S205 are the same, the control unit 202 does not control the braking device 36.

[0111] When the component to be controlled is the antilock brake device 38, the control unit 202 outputs a control signal to the antilock brake device 38. For example, when the control state before the control corresponds to the first hydraulic pressure and the action selected in step S204 or step S205 corresponds to the second hydraulic pressure, the control unit 202 outputs a control signal for transitioning the hydraulic pressure of the antilock brake device 38 from the first hydraulic pressure to the second hydraulic pressure. When the hydraulic pressure before the control and the hydraulic pressure corresponding to the action selected in step S204 or step S205 are the same, the control unit 202 does not control the antilock brake device 38.

[0112] When the component to be controlled is the suspension device 40, the control unit 202 outputs a control signal to the suspension device 40. For example, when the control state before the control indicates the first damping property and the action selected in step S204 or step S205 indicates the second damping property, the control unit 202 outputs a control signal to transition the damping property of the suspension device 40 from the first damping property to the second damping property. When the damping property before the control and the damping property corresponding to the action selected in step S204 or step S205 are the same, the control unit 202 does not control the suspension device 40.

[0113] In this embodiment, an action is selected using the ε-greedy method, which uses a parameter ε to determine whether to use a Q value or to select an action randomly, but an action may be selected using an algorithm other than the ε-greedy method.

[0114] Third embodiment In the third embodiment, a method for generating a learning model 120 will be described.

[0115] The evaluation device 100 uses a computer to evaluate the desirability of control of the components of the human-powered vehicle 1 using a function in accordance with driving information related to the driving of the human-powered vehicle 1, and updates the function in accordance with the evaluation result of the desirability and the control result of the components in accordance with the evaluation result, thereby generating the learning model 120. More specifically, the evaluation device 100 updates the action value function Q(s, a) based on the following equation, and generates the learning model 120 by improving the parameters that characterize the learning model 120. Q(s t ,a t )←Q(s t ,a t ) + α(r t+1 +γmax a' Q(s t+1 , a')-Q(s t ,a t ))…Formula (1)

[0116] The action value function Q(s, a) is the expected future profit when action a is taken in state s. r is the reward. The subscript t of state s, action a, and reward r is a number indicating one step in the thought process that repeats chronologically. The subscript t is also called the trial number. When the state changes after the action decision, the trial number is incremented. Therefore, the reward r in formula (1) t+1 is the state s t Action a t is selected and the state is s t+1 α is the learning rate, and γ is the discount rate. Action a' is the reward obtained when the state s t+1 Possible actions in a t+1 Among them, the action value function Q(s t+1 ,a t+1 ) is the action that maximizes max a' Q(s t+1 , a') is the action value function that is maximized by choosing action a'.

[0117] 7 is a flowchart explaining the procedure for generating the learning model 120. In step S301, the calculation processing unit 104 of the evaluation device 100 initializes definition information of the learning model 120. 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. Hereinafter, various parameters included in the definition information of the learning model 120 are denoted as parameter θ for convenience.

[0118] In step S302, the arithmetic processing unit 104 of the evaluation device 100 acquires driving information. The driving information is observed by sensors S1 to S10 and the like. The arithmetic processing unit 104 acquires at least one of the following as the driving information: speed, acceleration, attitude, load, driving force, cadence, gear ratio, operation input to the operation device 30, history of operation input and control results of components, regeneration amount, inclination, road surface conditions, position, weather, and vehicle information. The driving information acquired by the arithmetic processing unit 104 may be information related to the traction of the human-powered vehicle 1. The driving information may be not only information observed by the various sensors S1 to S10, but also information calculated by the arithmetic processing unit 104, or information previously stored in the storage unit 106.

[0119] In step S303, the calculation processing unit 104 inputs the driving information acquired from the input unit 102 etc. to the input layer 122 of the learning model 120, and calculates the Q value by executing a calculation by the learning model 120. The driving information input to the input layer 122 is provided to the intermediate layer 124. In the intermediate layer 124, at each node, a weight is multiplied for the input from the previous layer, a bias is added, and a calculation using an activation function is performed, and the calculation result is output to the output layer 126. The output layer 126 outputs the Q value through each node. The calculation processing unit 104 outputs the calculated Q value of each action to the control device 200.

[0120] In step S304, the control unit 202 of the control device 200 selects an action to be taken by the human-powered vehicle 1. The control unit 202 uses, for example, the ε-greedy method to determine whether to select an action using a Q value or to select an action randomly.

[0121] In step S305, the control unit 202 controls the components of the human-powered vehicle 1 to cause the human-powered vehicle 1 to take the selected action.

[0122] In step S306, the calculation processing unit 104 of the evaluation device 100 acquires the driving information after the components are controlled. The driving information acquired by the calculation processing unit 104 is similar to the driving information acquired in step S302.

[0123] In step S307, the calculation processing unit 104 evaluates the reward. In this embodiment, the reward is determined based on the running state of the human-powered vehicle 1. For example, if the human-powered vehicle 1 can run without losing traction, a positive reward is awarded, and if the human-powered vehicle 1 loses traction or falls over, a negative reward is awarded. When the current trial number is t, the reward awarded in step S307 is calculated by subtracting r in formula (1). t+1 It is.

[0124] In step S308, the calculation processing unit 104 calculates the value of the objective function. In order to properly update the action value function Q(s, a), it is necessary to optimize the learning model 120 that outputs the action value function Q(s, a). At a stage where learning is not completed, there is no knowledge of the action value function Q(s, a), and it is difficult to set target teacher data. Therefore, in this embodiment, the parameter θ that characterizes the learning model 120 is improved by the objective function that minimizes the TD (Temporal Difference) error shown in the second term of formula (1).

[0125] In step S309, the calculation processing unit 104 judges whether the learning is completed. In this embodiment, a threshold value for judging whether the TD error is sufficiently small is determined in advance, and when the value of the objective function is equal to or smaller than the threshold value, the calculation processing unit 104 judges that the learning is completed.

[0126] If the calculation processing unit 104 determines in step S309 that the learning is not completed, it updates the action value function Q(s, a) in step S310. For example, the calculation processing unit 104 specifies the change direction of the parameter θ that reduces the objective function based on the partial differentiation of the TD error with respect to the parameter θ, and changes the parameter θ. The specification of the change direction of the parameter θ is not limited to the partial differentiation of the TD error with respect to the parameter θ, and various gradient descent methods can be used. By the above processing, the parameter θ can be changed so that the action value function Q(s, a) approaches the target. After updating the action value function Q(s, a), the calculation processing unit 104 returns the processing to step S303.

[0127] If the calculation processing unit 104 determines in step S309 that the learning has ended, then in step S311 it updates the definition information of the learning model 120. The calculation processing unit 104 stores θ obtained by learning in the storage unit 106 as definition information of the learning model 120 to be referenced when evaluating the desirability of control of the human-powered vehicle 1.

[0128] As described above, in this embodiment, the process of selecting an action a that maximizes the value of the action value function (s, a) and the process of controlling the human-powered vehicle 1 based on the selected action a are repeated to optimize the action value function (s, a), thereby generating the learning model 120.

[0129] In this embodiment, the learning model 120 is generated in the evaluation device 100 of the human-powered vehicle 1, but the learning model 120 may be generated in an external server. In this case, a communication device (not shown) capable of communicating with the external server is connected to the evaluation device 100. Alternatively, a communication device capable of communicating with the external server is built into the evaluation device 100. The evaluation device 100 uploads the driving information acquired from the sensors S1 to S10, etc. to the external server through the communication device. The external server selects an action a that maximizes the value of the action value function (s, a) based on the action value function Q(s, a) and notifies the human-powered vehicle 1 of information on the selected action a. The control device 200 of the human-powered vehicle 1 controls the components based on the information on the action a selected by the external server. In this way, the action value function (s, a) is optimized in the external server and the learning model 120 is generated while repeating the process of the external server that selects the action a that maximizes the value of the action value function (s, a) and the process of the control device 200 that controls the human-powered vehicle 1 based on the selected action a.

[0130] (Fourth embodiment) In the fourth embodiment, a configuration for re-learning the learning model 120 will be described.

[0131] 8 is a flowchart illustrating the re-learning procedure of the learning model 120. In step S401, the calculation processing unit 104 of the evaluation device 100 reads out definition information of the learning model 120 stored in the storage unit .

[0132] In step S402, the calculation processing unit 104 of the evaluation device 100 acquires the driving information. The driving information acquired by the calculation processing unit 104 is the same as that in the third embodiment.

[0133] In step S403, the calculation processing unit 104 inputs the driving information acquired from the input unit 102 etc. to the input layer 122 of the learning model 120, and executes a calculation by the learning model 120 to calculate the Q value.

[0134] In step S404, the control unit 202 of the control device 200 selects an action to be taken by the human-powered vehicle 1. The control unit 202 uses, for example, the ε-greedy method to determine whether to select an action using a Q value or to select an action randomly.

[0135] In step S405, the control unit 202 controls the components of the human-powered vehicle 1 to cause the human-powered vehicle 1 to take the selected action.

[0136] In step S406, the calculation processing unit 104 of the evaluation device 100 acquires the driving information after the components are controlled. The driving information acquired by the calculation processing unit 104 is the same as the driving information acquired in step S402.

[0137] In step S407, the calculation processing unit 104 evaluates the reward. In this embodiment, the reward is determined based on the running state of the human-powered vehicle 1. For example, if the human-powered vehicle 1 can run without losing traction, a positive reward is given, and if the human-powered vehicle 1 loses traction or falls over, a negative reward is given.

[0138] In step S408, the calculation processing unit 104 calculates a value of the objective function. In this embodiment, the parameter θ that characterizes the learning model 120 is improved by the objective function that minimizes the TD error, which is shown in the second term of equation (1).

[0139] In step S409, the calculation processing unit 104 judges whether the learning is completed. In this embodiment, a threshold value for judging whether the TD error is sufficiently small is determined in advance, and when the value of the objective function is equal to or smaller than the threshold value, the calculation processing unit 104 judges that the learning is completed.

[0140] If the calculation processing unit 104 determines in step S409 that the learning is not completed, it updates the action value function Q(s, a) in step S410. For example, the calculation processing unit 104 specifies the change direction of the parameter θ that reduces the objective function based on the partial differentiation of the TD error with respect to the parameter θ, and changes the parameter θ. The specification of the change direction of the parameter θ is not limited to the partial differentiation of the TD error with respect to the parameter θ, and various gradient descent methods can be used. By the above processing, the parameter θ can be changed so that the action value function Q(s, a) approaches the target. After updating the action value function Q(s, a), the calculation processing unit 104 returns the processing to step S403.

[0141] When it is determined in step S409 that the learning is completed, in step S411, the calculation processing unit 104 updates the definition information of the learning model 120. The updated definition information of the learning model 120 is stored in the storage unit .

[0142] As described above, in this embodiment, the evaluation device 100 re-learns the learning model 120 in accordance with the evaluation result of desirability and the control result of the component in accordance with the evaluation result. That is, in this embodiment, the learning model 120 can be re-learned by optimizing the action value function (s, a) while repeating the process of selecting the action a that maximizes the value of the action value function (s, a) and the process of controlling the human-powered vehicle 1 based on the selected action a.

[0143] Fifth embodiment In the fifth embodiment, a notification system that notifies the evaluation result of the evaluation device 100 will be described.

[0144] Fig. 9 is a block diagram illustrating a notification system according to a fifth embodiment. The notification system according to the fifth embodiment includes an evaluation device 100 and a notification device 300 that notifies the evaluation result by the evaluation device 100. The evaluation result notified by the notification device 300 may include information on the control state of the component. In the example shown in Fig. 9, the evaluation device 100 and the notification device 300 are described as separate entities, but the evaluation device 100 and the notification device 300 may be configured as an integrated unit.

[0145] The notification device 300 is, for example, a display device such as a liquid crystal display, an organic electroluminescence display, etc. The notification device 300 can notify the rider of information related to the evaluation result by displaying text information or image information on the display device according to the evaluation result of the evaluation device 100.

[0146] The notification device 300 may be a light-emitting element such as an LED. The notification device 300 can notify the rider of information related to the evaluation result by turning on, blinking, or turning off the light-emitting element depending on the evaluation result of the evaluation device 100.

[0147] The notification device 300 may be an audio output device such as a speaker. The notification device 300 can notify the rider of information related to the evaluation result by outputting a sound from the audio output device in accordance with the evaluation result of the evaluation device 100.

[0148] The notification device 300 may be a communication device. The notification device 300 can notify the rider of information related to the evaluation result by transmitting text information or image information to an external communication device according to the evaluation result of the evaluation device 100. The external communication device is, for example, a smartphone, tablet, or wearable device carried by the rider.

[0149] The notification device 300 may be a vibration device such as a vibrator. The notification device 300 vibrates the vibration device in accordance with the evaluation result of the evaluation device 100, thereby notifying the rider of information related to the evaluation result.

[0150] The notification device 300 does not need to constantly notify the evaluation result of the evaluation device 100, and may notify information related to the evaluation result under specific conditions. For example, the notification device 300 may notify information related to the evaluation result when the deceleration of the human-powered vehicle 1 becomes lower than a predetermined value compared to the operation input of the brake operation device 36C.

[0151] As described above, in this embodiment, the evaluation result of the evaluation device 100 can be notified to the rider by the notification device 300.

[0152] The embodiments disclosed herein are illustrative in all respects and should not be considered as limiting. The scope of the present invention is defined by the claims, not by the above meaning, and is intended to include all modifications within the scope and meaning equivalent to the claims. [Explanation of symbols]

[0153] 1... human-powered vehicle, 30... operation device, 32, 33... transmission, 34... assist device, 36... brake device, 38... anti-lock brake device, 40... suspension device, 42... battery unit, 100... evaluation device, 102... input unit, 104... calculation processing unit, 106... storage unit, 108... output unit, 110... evaluation processing program, 120... learning model, 200... control device, 202... control unit, 204... storage unit, 206... input unit, 208... output unit, 300... notification device

Claims

1. An evaluation device that uses a learning model to evaluate the desirability of control when control is executed on components of a human-powered vehicle in accordance with driving information including information regarding traction related to the driving of the human-powered vehicle, from the perspective of whether the traction of the human-powered vehicle is good or bad.

2. The evaluation device according to claim 1 , further comprising: a step of retraining the learning model in response to the evaluation result of the desirability and a control result of the component in response to the evaluation result.

3. 3. The evaluation device according to claim 1, wherein the driving information includes at least one of information regarding speed, acceleration, posture, load, driving force, cadence, gear ratio, operation input to an operation device, history of the operation input and control results of the component, regeneration amount, inclination, road surface conditions, position, weather, and vehicle information.

4. An evaluation device according to any one of claims 1 to 3, A control device that controls the component based on an evaluation result of the evaluation device.

5. The control system according to claim 4 , wherein the component includes an assist device having an electric drive unit that outputs an auxiliary drive force.

6. 6. A control system according to claim 4 or 5, wherein the component comprises a brake device having an electric drive unit.

7. 7. A control system according to claim 4, wherein the component comprises an anti-lock braking system having an electrically powered drive unit.

8. 8. A control system according to claim 4, wherein the component comprises a suspension arrangement having an electric drive unit.

9. An evaluation device according to any one of claims 1 to 3, a notification device that notifies information regarding an evaluation result of the evaluation device.

10. The notification system according to claim 9 , wherein the information relating to the evaluation result includes information relating to a control state of the component.

11. The computer evaluating, using a function, the desirability of control when control is executed on a component of the human-powered vehicle in accordance with driving information including information related to traction related to the driving of the human-powered vehicle, from the viewpoint of the traction of the human-powered vehicle; A method for generating a learning model, which generates a learning model by updating the function according to the evaluation result of the desirability and the control result of the component according to the evaluation result.

12. On the computer, According to driving information including information related to traction related to the driving of the human-powered vehicle, the desirability of control when the control is executed on a component of the human-powered vehicle is evaluated from the viewpoint of the traction of the human-powered vehicle using a learning model. A computer program for executing a process.

13. A storage medium on which the computer program according to claim 12 is stored.

Images