JP7627615B2 - Control device for human-powered vehicles - Google Patents

Description

This disclosure relates to a control device for a human-powered vehicle.

For example, the control device for a human-powered vehicle in Patent Document 1 controls an electric device according to the output of a sensor that detects the driving state.

Patent No. 6005110

The control device for a human-powered vehicle in Patent Document 1 can execute control according to the traveling state of the human-powered vehicle, but does not take into consideration the surrounding environment, etc.
An object of the present disclosure is to provide a control device for a human-powered vehicle that can suitably control an electric device.

A control device according to a first aspect of the present disclosure is a control device for a human-powered vehicle, and includes a control unit configured to control an electric device of the human-powered vehicle, wherein the control unit is configured to estimate a driving path based on a forward image acquired from an imaging device, estimate a future pedaling state of the rider in accordance with the estimated driving path, and control the electric device in accordance with a control state of the electric device that is pre-associated with the estimated pedaling state.
According to the control device of the first aspect, the electric device is controlled according to a control state that is pre-associated with a future pedaling state of the rider according to the estimated travel route. Therefore, when the human-powered vehicle passes through the estimated travel route, the electric device is controlled according to the control state that is pre-associated with a pedaling state according to the estimated travel route, and therefore the electric device can be controlled in an optimal manner.

In the control device of a second aspect according to the first aspect of the present disclosure, the control unit includes a pedaling state estimation unit configured to estimate the travel path along which the human-powered vehicle will travel in the future based on the forward image, and to estimate the pedaling state according to the estimated travel path.
According to the control device of the second aspect, the pedaling state estimating section can appropriately estimate the road along which the human-powered vehicle will travel in the future, and the future pedaling state of the rider.

In the control device of a third aspect according to the second aspect of the present disclosure, the pedaling state estimator includes an artificial intelligence processor.
According to the control device of the third aspect, the artificial intelligence processing unit can suitably estimate the route along which the human-powered vehicle will travel in the future and the future pedaling state of the rider.

In the control device of a fourth aspect according to the third aspect of the present disclosure, the artificial intelligence processing unit is configured to estimate the driving path based on the forward image, and the pedaling state estimation unit is configured to estimate the pedaling state according to the driving path estimated by the artificial intelligence processing unit.
According to the control device of the fourth aspect, the artificial intelligence processing unit can suitably estimate the traveling path based on the forward image, and therefore the pedaling state estimating unit can suitably estimate the pedaling state from the traveling path estimated by the artificial intelligence processing unit.

In the control device of a fifth aspect according to the fourth aspect of the present disclosure, the artificial intelligence processing unit is configured to identify an object on the driving path based on the forward image, and to estimate the driving path based on the identified object.
According to the control device of the fifth aspect, the travel path can be suitably estimated based on objects on the travel path.

In the control device of a sixth aspect according to the fifth aspect of the present disclosure, the artificial intelligence processing unit is configured to identify a boundary between an area in the forward image that is the driving path and an area that is not the driving path based on the identified object, and to estimate the driving path based on the identified boundary.
According to the control device of the sixth aspect, the road can be suitably estimated by identifying the boundary between an area that is a road and an area that is not a road in the forward image.

In the control device of a seventh aspect according to the fifth or sixth aspect of the present disclosure, the artificial intelligence processing unit is configured to detect a distance between the object and the human-powered vehicle based on the forward image, and the pedaling state estimation unit is configured to estimate the pedaling state according to the object identified by the artificial intelligence processing unit and the distance estimated by the artificial intelligence processing unit.
According to the control device of the seventh aspect, the future pedaling state of the rider can be estimated according to the distance between the object and the human-powered vehicle.

In the control device of an eighth aspect according to any one of the first to seventh aspects of the present disclosure, the pedaling state includes a first pedaling state and a second pedaling state, the first pedaling state is a state in which it is more difficult for the rider to pedal than the second pedaling state, and the control unit is configured to estimate whether the pedaling state is the first pedaling state or the second pedaling state.
According to the control device of the eighth aspect, the electric device can be suitably controlled in accordance with an estimation of the difficulty of the rider pedaling.

In the control device of a ninth aspect according to the eighth aspect of the present disclosure, the second pedaling state includes a third pedaling state and a fourth pedaling state, the third pedaling state is a state in which it is more difficult for the rider to pedal than the fourth pedaling state, and the control unit is configured to estimate whether the pedaling state is the first pedaling state, the third pedaling state, or the fourth pedaling state.
According to the control device of the ninth aspect, when it is estimated that the rider is having difficulty pedaling, the electric device can be suitably controlled further in accordance with the third pedaling state and the fourth pedaling state.

In the control device of a tenth aspect according to the eighth or ninth aspect of the present disclosure, when the control unit estimates, based on the forward image, that it is difficult for the rider pedaling the human-powered vehicle to travel on the road, the control unit is configured to estimate the pedaling state to be the first pedaling state.
According to the control device of the tenth aspect, the first pedaling state can be suitably estimated from the forward image.

In the control device of an eleventh aspect according to any one of the first to seventh aspects of the present disclosure, the pedaling state includes a third pedaling state and a fourth pedaling state, the third pedaling state is a state in which it is more difficult for the rider to pedal than the fourth pedaling state, and the control unit is configured to estimate whether the pedaling state is the third pedaling state or the fourth pedaling state.
According to the control device of the eleventh aspect, the electric device can be suitably controlled in accordance with an estimation of the difficulty of the rider pedaling.

In the control device of the twelfth aspect according to the ninth or eleventh aspect of the present disclosure, the third pedaling state is a state in which the load required by the rider for pedaling is equal to or greater than a first load, and the fourth pedaling state is a state in which the load required by the rider for pedaling is less than the first load.
According to the control device of the twelfth aspect, the electric device can be suitably controlled in accordance with an estimate of the load required by the rider when pedaling.

In the control device of a thirteenth aspect according to any one of the ninth, eleventh and twelfth aspects of the present disclosure, the third pedaling state is a state in which the load required by the rider for pedaling increases to or exceeds a second load, and the fourth pedaling state is a state in which the load required by the rider for pedaling decreases to less than the second load.
According to the control device of the thirteenth aspect, the electric device can be suitably controlled in accordance with an estimation of the load that the rider will require in future pedaling.

In the control device of a fourteenth aspect according to any one of the ninth, eleventh to thirteenth aspects of the present disclosure, when the control unit estimates, based on the forward image, that it is easy for the rider pedaling the human-powered vehicle to travel along the travel path, the control unit is configured to estimate, depending on the travel path, whether the pedaling state is the third pedaling state or the fourth pedaling state.
According to the control device of the fourteenth aspect, the electric device can be suitably controlled in accordance with an estimation of the ease with which the rider can pedal from the forward image.

In the control device of the fifteenth aspect according to the tenth or fourteenth aspect of the present disclosure, when it is estimated that the road has at least one of a curve, a step, a road surface gradient, and an obstacle, the control unit estimates the pedaling state in accordance with the estimation that the road has at least one of the curve, the step, the road surface gradient, and the obstacle.
According to the control device of the fifteenth aspect, the electric device can be suitably controlled in accordance with the estimation of at least one of a curve, a step, a road surface gradient, and an obstacle on the estimated travel path.

In the control device of a sixteenth aspect according to any one of the first to fifteenth aspects of the present disclosure, the electric device includes a transmission device that changes a gear ratio of the human-powered vehicle.
According to the control device of the sixteenth aspect, the transmission can be suitably controlled by a control state that is associated in advance with a pedaling state according to the estimated traveling route.

In the control device of a seventeenth aspect in accordance with the sixteenth aspect of the present disclosure, the control unit is configured to control the transmission to change the gear ratio in accordance with a gear shift condition, and is configured to change the gear shift condition in accordance with an estimated pedaling state.
According to the control device of the seventeenth aspect, the gear shift conditions can be suitably changed in accordance with the pedaling state corresponding to the estimated traveling road.

In the control device of aspect 18 according to any one of aspects 8 to 10 of the present disclosure, the electric device includes a transmission that changes a gear ratio, and the control unit is configured to control the transmission to change the gear ratio in accordance with a gear shift condition, and is configured to change the gear shift condition in accordance with the estimated pedaling state, and is configured to change the gear shift condition when the pedaling state is the first pedaling state so that the gear ratio is changed less frequently than when the pedaling state is the second pedaling state.
According to the control device of the eighteenth aspect, the gear change conditions can be suitably changed in accordance with an estimation of the difficulty of the rider's pedaling. According to the control device of the eighteenth aspect, when it is estimated that the rider's pedaling is difficult, the frequency of changes in the gear ratio can be reduced.

In the control device of a 19th aspect according to any one of the 9th, 11th to 13th aspects of the present disclosure, the electric device includes a transmission that changes a gear ratio, and the control unit is configured to control the transmission to change the gear ratio in accordance with a gear shift condition, and to change the gear shift condition in accordance with an estimated pedaling state.
According to the control device of the nineteenth aspect, the gear shift conditions can be suitably changed in accordance with an estimate of the rider's pedaling difficulty.

In the control device of the twentieth aspect in accordance with the nineteenth aspect of the present disclosure, the control unit is configured to change the gear shift conditions when the pedaling state is the third pedaling state so that the frequency of changes to increase the gear ratio is reduced compared to when the pedaling state is the fourth pedaling state.
According to the control device of the twentieth aspect, the gear change conditions can be suitably changed in accordance with an estimation of the difficulty of the rider's pedaling. According to the control device of the twentieth aspect, when it is estimated that the rider's pedaling is difficult, the frequency of changes to increase the gear ratio can be reduced.

In the control device of aspect 21 according to aspect 19 or 20 of the present disclosure, the control unit is configured to change the gear shift conditions so that, when the pedaling state is the third pedaling state, the frequency of changes to reduce the gear ratio is greater than when the pedaling state is the fourth pedaling state.
According to the control device of the twenty-first aspect, the gear change conditions can be suitably changed in accordance with the estimation of the difficulty of the rider's pedaling. According to the control device of the twenty-first aspect, when it is estimated that the rider's pedaling is difficult, the gear ratio is likely to be small. Therefore, when the human-powered vehicle travels on a road where the rider's pedaling is difficult, the load on the rider can be reduced.

In the control device of aspect 22 according to any one of aspects 17 to 21 of the present disclosure, the gear shift condition includes a gear shift threshold, and the control unit is configured to control the transmission to change the gear ratio in response to a comparison between a parameter relating to at least one of a running state and a running environment of the human-powered vehicle and the gear shift threshold, and is configured to change the gear shift threshold in response to an estimated pedaling state.
According to the control device of the twenty-second aspect, the gear ratio can be suitably changed by changing the gear shift threshold value in response to a parameter related to at least one of the running state and running environment of the human-powered vehicle.

In the control device of the twenty-third aspect according to the twenty-second aspect of the present disclosure, the parameters include at least one of the human-powered driving force input to the human-powered vehicle, the rotational speed of a crank of the human-powered vehicle, the vehicle speed of the human-powered vehicle, and the inclination of the human-powered vehicle.
According to the control device of the twenty-third aspect, the gear ratio can be suitably changed depending on at least one of the human-powered driving force input to the human-powered vehicle, the rotational speed of the crank of the human-powered vehicle, the vehicle speed of the human-powered vehicle, and the inclination of the human-powered vehicle.

In the control device of aspect 24 according to aspect 22 or 23 of the present disclosure, the shift threshold includes an upper limit value and a lower limit value, and the control unit is configured to change the shift condition by changing at least one of the upper limit value and the lower limit value depending on the pedaling state to be estimated.
According to the control device of the twenty-fourth aspect, by changing at least one of the upper limit value and the lower limit value, it is possible to suitably change the gear shift conditions in accordance with the estimated pedaling state.

In the control device of aspect 25 according to any one of aspects 17 to 24 of the present disclosure, the control unit is configured to be able to switch the gear shift state, and the gear shift state includes a first gear shift state in which the gear ratio is changed in accordance with the gear shift conditions, and a second gear shift state in which the gear ratio is not changed in accordance with the gear shift conditions, and the control unit is configured to switch the gear shift state between the first gear shift state and the second gear shift state in accordance with the estimated pedaling state.
According to the control device of the twenty-fifth aspect, the gear shift state can be suitably switched in accordance with the estimated pedaling state.

The control device for a human-powered vehicle disclosed herein can optimally control the electric device according to the estimated driving route.

1 is a side view of a human-powered vehicle including a control device for a human-powered vehicle according to an embodiment; FIG. 2 is a block diagram showing an electrical configuration of the control device for the human-powered vehicle shown in FIG. 1 . FIG. 3 is a schematic diagram showing an implementation example of a learning model of the artificial intelligence processing unit in FIG. 2 . FIG. 3 is a schematic diagram showing a first example of a forward image processed by the artificial intelligence processing unit of FIG. 2 . FIG. 3 is a schematic diagram showing a second example of a forward image processed by the artificial intelligence processing unit in FIG. 2 . 3 is a flowchart of a process executed by the control unit in FIG. 2 to estimate a pedaling state. 3 is a flowchart of a process executed by the control unit of FIG. 2 to control the transmission in accordance with a gear shifting condition. 3 is a flowchart of a process executed by the control unit in FIG. 2 to control the transmission in accordance with a gear shift state. 10 is a flowchart of a process executed by a control unit of a modified example to control an electrically-powered device.

<Embodiment>
A control device 60 for a human-powered vehicle according to an embodiment will be described with reference to Figs. 1 to 8. The human-powered vehicle 10 is a vehicle that has at least one wheel and can be driven at least by human-powered driving force. The human-powered vehicle includes various types of bicycles, such as mountain bikes, road bikes, city bikes, cargo bikes, hand bikes, and recumbents. The number of wheels that the human-powered vehicle has is not limited. The human-powered vehicle also includes, for example, one-wheeled vehicles and vehicles with three or more wheels. The human-powered vehicle is not limited to vehicles that can be driven only by human-powered driving force. The human-powered vehicle includes not only human-powered driving force but also E-bikes that use the driving force of an electric motor for propulsion. The E-bike includes an electrically assisted bicycle whose propulsion is assisted by an electric motor. In the following embodiment, the human-powered vehicle will be described as a mountain bike.

The human-powered vehicle 10 includes at least one wheel 14 and a vehicle body 16. The at least one wheel 14 includes a rear wheel 14A and a front wheel 14B. The vehicle body 16 includes a frame 18. The vehicle body 16 is provided with an input rotating shaft 12A that can rotate relative to the frame 18. In this embodiment, the input rotating shaft 12A is a crankshaft included in the crank 12. The crank 12 includes the input rotating shaft 12A, a first crank arm 12B provided at one end in the axial direction of the input rotating shaft 12A, and a second crank arm 12C provided at the other end in the axial direction of the input rotating shaft 12A. A first pedal 20A is connected to the first crank arm 12B. A second pedal 20B is connected to the second crank arm 12C. The rear wheel 14A is driven by the rotation of the crank 12. The rear wheel 14A is supported by the frame 18. The crank 12 is connected to the rear wheel 14A by a drive mechanism 22.

The drive mechanism 22 includes a first rotating body 24 connected to the input rotating shaft 12A. The input rotating shaft 12A may be connected to the first rotating body 24 so as to rotate integrally with the first rotating body 24, or may be connected via a first one-way clutch. The first one-way clutch is configured to rotate the first rotating body 24 forward when the crank 12 rotates forward, and to allow relative rotation between the crank 12 and the first rotating body 24 when the crank 12 rotates backward. The first rotating body 24 includes a front sprocket. The first rotating body 24 may include a pulley or a bevel gear. The drive mechanism 22 further includes a second rotating body 26 and a connecting member 28. The connecting member 28 transmits the rotational force of the first rotating body 24 to the second rotating body 26. The connecting member 28 includes, for example, a chain, a belt, or a shaft.

The second rotating body 26 is connected to the rear wheel 14A. The second rotating body 26 includes a rear sprocket. The second rotating body 26 may include a pulley or a bevel gear. A second one-way clutch is preferably provided between the second rotating body 26 and the rear wheel 14A. The second one-way clutch is configured to rotate the rear wheel 14A forward when the second rotating body 26 rotates forward, and to allow relative rotation between the second rotating body 26 and the rear wheel 14A when the second rotating body 26 rotates backward.

A front wheel 14B is attached to the frame 18 via a front fork 30. A handlebar 34 is connected to the front fork 30 via a stem 32. In this embodiment, the rear wheel 14A is connected to the crank 12 by the drive mechanism 22, but at least one of the rear wheel 14A and the front wheel 14B may be connected to the crank 12 by the drive mechanism 22.

For example, the human-powered vehicle 10 further includes a battery 38. The battery 38 includes one or more battery elements. The battery elements include a rechargeable battery. The battery 38 is configured to supply power to the control device 60. The battery 38 is preferably communicatively connected to a control unit 62 of the control device 60 via an electric cable or a wireless communication device. The battery 38 can communicate with the control unit 62, for example, via power line communication (PLC), a controller area network (CAN), or a universal asynchronous receiver/transmitter (UART).

The human-powered vehicle 10 further includes an electric device 40. For example, the electric device 40 includes a transmission 42 that changes the gear ratio R of the human-powered vehicle 10. The transmission 42 is provided in the transmission path of the human-powered driving force. The gear ratio R is expressed by the ratio of the rotation speed A2 of the output section of the transmission 42 to the rotation speed A1 of the input section of the transmission 42. The gear ratio R is expressed by the formula "R = A2/A1". The larger the gear ratio R, the faster the rotation speed of the crank 12 is transmitted to the wheels 14.

In this embodiment, the transmission 42 has a derailleur 42A and a plurality of sprockets 42B that have a rotation axis and are arranged in the direction in which the rotation axis extends. If the derailleur 42A includes a rear derailleur, the plurality of sprockets 42B includes the second rotating body 26. If the derailleur 42A includes a front derailleur, the plurality of sprockets 42B includes the first rotating body 24. If the transmission 42 is a derailleur 42A, the rotational speed A2 of the output part of the transmission 42 corresponds to the rotational speed of the second rotating body 26. If the transmission 42 is a derailleur 42A, the rotational speed A1 of the input part of the transmission 42 corresponds to the rotational speed of the first rotating body 24. The transmission 42 may include an internal gearbox.

For example, the transmission 42 includes an electric actuator. When the transmission 42 has a derailleur 42A, the electric actuator is provided in the derailleur 42A to operate the derailleur 42A. The electric actuator may be provided, for example, in the frame 18, separate from the derailleur 42A and the internal gear shifter. The electric actuator includes an electric motor and a reducer. When the electric actuator is provided in the frame 18, the electric actuator is connected to the derailleur 42A or the internal gear shifter by, for example, a Bowden cable.

For example, the human-powered vehicle 10 is provided with an imaging device 44 capable of capturing a forward image F of the human-powered vehicle 10. When the imaging device 44 is provided on the human-powered vehicle 10, for example, the imaging device 44 is provided on the frame 18 or the handlebars 34. The imaging device 44 may be provided on the rider. When the imaging device 44 is provided on the rider, the imaging device 44 may be provided on a helmet worn by the rider. The imaging device 44 includes, for example, a camera. The imaging device 44 may be capable of capturing only the forward image F, or may be capable of capturing peripheral images other than the forward image F. The imaging device 44 may be capable of capturing an image of the entire circumference around the imaging device 44. For example, the imaging device 44 is configured to be able to communicate with the control device 60 by at least one of wireless communication and wired communication. The imaging device 44 transmits the captured forward image F to the control device 60.

The control device 60 includes a control unit 62. The control unit 62 includes a processing unit that executes a predetermined control program. The processing unit included in the control unit 62 includes, for example, a CPU (Central Processing Unit) or an MPU (Micro Processing Unit). The processing units included in the control unit 62 may be provided in multiple locations that are separate from each other. The control unit 62 may include one or multiple microcomputers.

For example, the control device 60 includes a memory unit 64. The memory unit 64 stores a predetermined control program and information used in the control process. The memory unit 64 includes, for example, a non-volatile memory and a volatile memory. The non-volatile memory includes, for example, at least one of a ROM (Read-Only Memory), an EPROM (Erasable Programmable Read Only Memory), an EEPROM (Electrically Erasable Programmable Read-Only Memory), and a flash memory. The volatile memory includes, for example, a RAM (Random Access Memory).

The control unit 62 is configured to control the electric device 40 of the human-powered vehicle 10. The control device 60 may be provided in the electric device 40, or may be provided in a component of the human-powered vehicle 10 other than the electric device 40. For example, the control unit 62 includes an electric device control unit 66. The electric device control unit 66 controls the electric device 40. For example, the electric device control unit 66 controls the electric device 40 in response to a signal from the operation unit 46. The operation unit 46 is provided, for example, on the handlebar 34. When the electric device 40 includes a transmission 42, the operation unit 46 is configured to be able to transmit to the control unit 62 an upshift signal that is an instruction to increase the gear ratio R and a downshift signal that is an instruction to decrease the gear ratio R. The electric device control unit 66 controls the gear ratio 42 in response to the upshift signal and the downshift signal. When the electric device 40 includes the transmission 42, in the description of the control of the electric device 40 by the control unit 62, the electric device 40 can be read as the transmission 42.

For example, the control unit 62 is configured to control the transmission 42 to change the gear ratio R in accordance with the shifting conditions. For example, the electric device control unit 66 is configured to control the transmission 42 to change the gear ratio R in accordance with the shifting conditions.

For example, the gear shift condition includes a gear shift threshold. The control unit 62 is configured to control the transmission 42 to change the gear ratio R in response to a comparison between a parameter related to at least one of the running state and running environment of the human-powered vehicle 10 and the gear shift threshold. For example, the control unit 62 controls the transmission 42 to change the gear ratio R in response to a difference between the parameter and the gear shift threshold. The gear shift threshold may change in response to the gear ratio R. For example, if the transmission 42 includes a continuously variable transmission and thus the gear ratio R can change smoothly, the gear shift threshold changes smoothly in response to the gear ratio R. For example, if the gear ratio R changes in stages, the gear shift threshold changes in stages in response to the gear ratio R.

For example, the gear shift threshold includes an upper limit value and a lower limit value. For example, the control unit 62 controls the gear shift device 42 to change the gear ratio R when the parameter is greater than the upper limit value. When the parameter is greater than the upper limit value, the control unit 62 controls the gear shift device 42 to change the gear ratio R so that the parameter is smaller than the upper limit value. When the parameter is greater than the upper limit value, the control unit 62 controls the gear shift device 42 so that the gear ratio R is increased. For example, the upper limit value may be configured to increase as the gear ratio R increases. In this case, the larger the gear ratio R, the more difficult it is to execute control to increase the gear ratio R. For example, the upper limit value may be configured to decrease as the gear ratio R increases. In this case, the larger the gear ratio R, the more easily it is to execute control to increase the gear ratio R.

For example, when the parameter is smaller than the lower limit, the control unit 62 controls the transmission 42 to change the gear ratio R. When the parameter is smaller than the lower limit, the control unit 62 controls the transmission 42 to change the gear ratio R so that the parameter is larger than the lower limit. When the parameter is smaller than the lower limit, the control unit 62 controls the transmission 42 to reduce the gear ratio R. For example, the control unit 62 may be configured so that the lower limit becomes larger as the gear ratio R becomes smaller. In this case, the smaller the gear ratio R, the easier it is to execute control to reduce the gear ratio R. For example, the control unit 62 may be configured so that the lower limit becomes smaller as the gear ratio R becomes smaller. In this case, the smaller the gear ratio R, the harder it is to execute control to reduce the gear ratio R.

Preferably, the parameters include at least one of the human-powered driving force input to the human-powered vehicle 10, the rotation speed of the crank 12 of the human-powered vehicle 10, the vehicle speed of the human-powered vehicle 10, and the inclination of the human-powered vehicle 10. For example, the control unit 62 may select at least two parameters, the human-powered driving force input to the human-powered vehicle 10, the rotation speed of the crank 12 of the human-powered vehicle 10, the vehicle speed of the human-powered vehicle 10, and the inclination of the human-powered vehicle 10, and calculate a new parameter from the relationship between the selected parameters. For example, the largest parameter among the selected parameters may be used as the new parameter. For example, the selected parameters may be combined to create the new parameter.

For example, the human-powered vehicle 10 further includes at least one of a human-powered force detection unit 48, a crank rotation sensor 50, a vehicle speed sensor 52, and an inclination sensor 54.

The human-powered driving force detection unit 48 is configured to detect information related to the human-powered driving force. The human-powered driving force detection unit 48 is provided, for example, on the frame 18, crank 12, or pedals 20A, 20B of the human-powered vehicle 10. If a drive unit is provided on the human-powered vehicle 10, the human-powered driving force detection unit 48 may be provided on the housing of the drive unit. The human-powered driving force detection unit 48 includes, for example, a torque sensor. The torque sensor is configured to output a signal corresponding to the torque applied to the crank 12 by the human-powered driving force. For example, if a first one-way clutch is provided on the power transmission path, the torque sensor is preferably provided upstream of the first one-way clutch on the power transmission path. The torque sensor includes a strain sensor, a magnetostrictive sensor, a pressure sensor, or the like. The strain sensor includes a strain gauge.

The crank rotation sensor 50 is configured to detect information related to the rotation speed of the input rotating shaft 12A. The crank rotation sensor 50 is provided, for example, on the frame 18 of the human-powered vehicle 10. If the human-powered vehicle 10 is provided with a drive unit, the crank rotation sensor 50 may be provided on the drive unit. The crank rotation sensor 50 may be provided on the housing of the drive unit. The crank rotation sensor 50 is configured to include a magnetic sensor that outputs a signal according to the strength of a magnetic field. An annular magnet with an S pole and an N pole adjacent in the circumferential direction is provided on the input rotating shaft 12A, a member that rotates in conjunction with the input rotating shaft 12A, or a power transmission path between the input rotating shaft 12A and the first rotating body 24. The member that rotates in conjunction with the input rotating shaft 12A may include the output shaft of a motor.

The vehicle speed sensor 52 is configured to detect information related to the vehicle speed of the human-powered vehicle 10. In this embodiment, the vehicle speed sensor 52 is configured to detect information related to the rotational speed of at least one wheel 14 of the human-powered vehicle 10. The vehicle speed sensor 52 is configured to detect, for example, a magnet provided on at least one wheel 14 of the human-powered vehicle 10. The vehicle speed sensor 52 is configured to output a predetermined number of detection signals, for example, during one rotation of one of the at least one wheels 14. The predetermined number is, for example, 1. The vehicle speed sensor 52 outputs a signal corresponding to the rotational speed of the wheel 14. The control unit 62 can calculate the vehicle speed of the human-powered vehicle 10 based on the signal corresponding to the rotational speed of the wheel 14 and information related to the circumference of the wheel 14.

The vehicle speed sensor 52 includes, for example, a magnetic reed constituting a reed switch, or a magnetic sensor such as a Hall element. The vehicle speed sensor 52 may be attached to a chain stay of the frame 18 of the human-powered vehicle 10 and configured to detect a magnet attached to the rear wheel 14A, or may be provided on the front fork 30 and configured to detect a magnet attached to the front wheel 14B.

The vehicle speed sensor 52 may have any configuration as long as it can acquire information related to the vehicle speed of the human-powered vehicle 10. The vehicle speed sensor 52 may be configured to detect slits provided in a disc brake, for example. The vehicle speed sensor 52 may be configured to include an optical sensor, for example. The vehicle speed sensor 52 may be configured to include a GPS (Global Positioning System) receiver, for example. When the vehicle speed sensor 52 includes a GPS receiver, the control unit 62 can calculate the vehicle speed according to the time and the traveled distance. The vehicle speed sensor 52 is connected to the control unit 62 via a wireless communication device or an electric cable.

The tilt sensor 54 is configured to detect information related to the tilt of the human-powered vehicle 10. The tilt sensor 54 includes, for example, a gyro sensor or an acceleration sensor. The tilt sensor 54 includes a GPS receiver. The control unit 62 may calculate the tilt angle of the road surface on which the human-powered vehicle 10 is traveling based on the GPS information acquired by the GPS receiver and the road surface gradient included in the pre-recorded map information.

The control unit 62 estimates the travel path based on the forward image F acquired from the imaging device 44. The control unit 62 estimates the rider's future pedaling state P according to the estimated travel path. The control unit 62 is configured to control the electric device 40 according to the control state of the electric device 40 that is pre-associated with the estimated pedaling state P. For example, the control unit 62 is configured to control the transmission 42 according to the control state of the transmission 42 that is pre-associated with the estimated pedaling state P. The control state of the electric device 40 that is pre-associated with the estimated pedaling state P is stored in advance in the memory unit 64, for example.

For example, the control unit 62 includes a pedaling state estimation unit 68. The pedaling state estimation unit 68 is configured to estimate the road along which the human-powered vehicle 10 will travel in the future based on the forward image F, and to estimate the pedaling state P according to the estimated road.

For example, the pedaling state estimation unit 68 includes an artificial intelligence processing unit 70. The artificial intelligence processing unit 70 includes, for example, a storage device 72 that stores software and an arithmetic processing unit 74 that executes the software stored in the storage device 72. The arithmetic processing unit 74 includes, for example, a CPU or an MPU. The arithmetic processing unit 74 preferably includes a GPU (Graphics Processing Unit) in addition to the CPU or MPU. The arithmetic processing unit 74 may include an FPGA (Field-Programmable Gate Array). The artificial intelligence processing unit 70 may include one or more arithmetic processing units 74. The artificial intelligence processing unit 70 may include multiple arithmetic processing units 74 that are arranged at multiple locations. The storage device 72 includes, for example, a non-volatile memory and a volatile memory. The storage device 72 stores a control program 76, a learning program 78, and a learning model 80. The learning model 80 may be a learned model that has been learned by a predetermined learning algorithm, or may be configured to be updated by a learning algorithm. The learning algorithm includes machine learning, deep learning, or deep reinforcement learning. The learning algorithm includes, for example, at least one of supervised learning, unsupervised learning, and reinforcement learning. As the learning algorithm, a method other than the method described in this specification may be used as long as it is configured to update the learning model 80 using a method belonging to the field of artificial intelligence. The learning process for updating the learning model 80 is preferably performed by a GPU. The learning algorithm may use a neural network (NN). The learning algorithm may use a recurrent neural network (RNN).

An example of the learning model 80 of the artificial intelligence processing unit 70 shown in FIG. 3 includes an input layer 82, an intermediate layer 84, and an output layer 86. A forward image F is input to the input layer 82. The learning model 80 is trained in advance to output a road when a forward image F is input. The intermediate layer 84 uses training data to learn the relationship between the information input to the input layer 82 and the road output by the output layer 86. The training data used by the intermediate layer 84 includes information about the image of the road and information about the boundary B between the area that is the road and the area that is not the road. The intermediate layer 84 includes a convolution layer 84A, a pooling layer 84B, and a fully connected layer 84C. A plurality of intermediate layers 84 may be provided. The output layer 86 outputs the road.

For example, the learning model 80 is configured to be updatable by an external device 90. The external device 90 is, for example, a smartphone or a personal computer. For example, the learning model 80 can update the number of convolutional layers 84A, pooling layers 84B, and fully connected layers 84C of the intermediate layer 84 by the external device 90. For example, the learning model 80 can update the relationship between the information input to the input layer 82 on which the intermediate layer 84 learns and the driving path output by the output layer 86 by the external device 90.

For example, the artificial intelligence processing unit 70 is configured to estimate the driving path based on the forward image F. For example, the artificial intelligence processing unit 70 estimates the driving path on which the human-powered vehicle 10 can travel, or the driving path that is likely to be traveled based on the rider's intention, based on the forward image F. For example, the artificial intelligence processing unit 70 estimates the state of the driving path based on the forward image F. The artificial intelligence processing unit 70 estimates the state of whether or not the driving path has at least one of a curve, a step, a road surface gradient, and an obstacle, based on the forward image F. When the forward image F transmitted from the imaging device 44 is input, the artificial intelligence processing unit 70 is configured to output the estimated driving path based on the feature amount of the forward image F. For example, the artificial intelligence processing unit 70 estimates the driving path by performing image processing on the forward image F. For example, the artificial intelligence processing unit 70 estimates the driving path by detecting edges in the forward image F.

For example, the artificial intelligence processing unit 70 identifies an object X on the road based on the forward image F. The artificial intelligence processing unit 70 is configured to estimate the road based on the identified object X. For example, the artificial intelligence processing unit 70 detects edges from the forward image F and identifies the detected edges as object X. For example, object X is a characteristic part in the forward image F. For example, object X includes at least one outline of a characteristic ground surface that is different from the surrounding ground surface, a tree W, a rock, and an artificial object.

For example, the artificial intelligence processing unit 70 identifies a boundary B between an area in the forward image F that is a driving path and an area that is not a driving path based on the identified object X. The artificial intelligence processing unit 70 is configured to estimate the driving path based on the identified boundary B. For example, the area in the forward image F that is a driving path includes an area of the front image F that is the ground surface and is suitable for driving. For example, the area suitable for driving includes flat soil and flat paved roads. For example, the area in the forward image F that is not a driving path includes an area in the forward image F that is not the ground surface and an area in the forward image F that is the ground surface and is not suitable for driving. For example, the area that is not the ground surface includes water surfaces, sky, and space. For example, the area that is not suitable for driving includes green spaces, slopes that cannot be traversed, and cliffs. For example, the area that is not suitable for driving includes uneven soil and uneven paved roads. For example, the driving path estimated by the artificial intelligence processing unit 70 has a width in a direction perpendicular to the driving direction of the human-powered vehicle 10. For example, if there is an obstacle that the human-powered vehicle 10 cannot pass through, the artificial intelligence processing unit 70 will not identify that part as part of the driving path.

With reference to FIG. 4 and FIG. 5, an example of the boundary B and the road of the forward image F identified by the artificial intelligence processing unit 70 will be described. For example, FIG. 4 includes an area that is a road and an area that is not a road. The area that is a road in FIG. 4 is, for example, an area with a flat earth surface. The area that is not a road in FIG. 4 is, for example, an area with a green space on the ground. The artificial intelligence processing unit 70 judges the area with a flat earth surface and the area with a green space on the ground by, for example, color. When the forward image F in FIG. 4 is input, the artificial intelligence processing unit 70 identifies the part where the area with a flat earth surface and the area with a green space on the ground surface are adjacent as the object X. The artificial intelligence processing unit 70 connects the identified object X to identify the boundary B. For example, the artificial intelligence processing unit 70 identifies two boundaries B extending forward from the forward image F and estimates the part between the two boundaries B as the road.

For example, the artificial intelligence processing unit 70 estimates that there is a change in the road based on the boundary B heading forward. For example, if the boundary B heading forward is curved, the artificial intelligence processing unit 70 determines that there is a curve in the road. For example, if the boundary B heading forward is interrupted in the traveling direction of the human-powered vehicle 10, and if the boundary B extends laterally in the front and the two boundaries B overlap, the artificial intelligence processing unit 70 determines that there is a step in the estimated road. For example, if the boundary B is distorted, the artificial intelligence processing unit 70 estimates that there is a road surface gradient on the road. For example, the artificial intelligence processing unit 70 estimates that there is a road surface gradient on the road based on the position on the front image F of the boundary B that contacts the part corresponding to the sky. For example, the artificial intelligence processing unit 70 estimates that there is a change in the road surface gradient on the road in accordance with the distortion of the boundary B. For example, the artificial intelligence processing unit 70 estimates that there is a change in the road surface gradient on the road based on the position on the front image F of the boundary B that contacts the part corresponding to the sky. For example, if boundary B surrounds a certain area, the artificial intelligence processing unit 70 estimates that there is an obstacle on the road.

For example, FIG. 5 shows a forward image F in which a tree W, which is an obstacle on the travel path, is included. In this case, it is estimated that there is an obstacle on the travel path in the forward image F in FIG. 5 based on a boundary B that surrounds the part corresponding to the tree W.

For example, the pedaling state estimation unit 68 is configured to estimate a pedaling state P according to the traveling path estimated by the artificial intelligence processing unit 70. For example, the pedaling state estimation unit 68 estimates a pedaling state P according to an object X on the traveling path estimated by the artificial intelligence processing unit 70. For example, the pedaling state estimation unit 68 estimates a pedaling state P according to a boundary B on the traveling path estimated by the artificial intelligence processing unit 70. For example, the pedaling state estimation unit 68 estimates a pedaling state P according to a change in the traveling path estimated by the artificial intelligence processing unit 70. For example, the pedaling state estimation unit 68 estimates a pedaling state P according to whether or not the traveling path estimated by the artificial intelligence processing unit 70 has at least one of a curve, a step, a road surface gradient, and an obstacle.

For example, when the traveling path is estimated by the boundary B estimated by the artificial intelligence processing unit 70, the pedaling state estimation unit 68 estimates the rider's future pedaling state P according to the estimated traveling path. For example, the memory unit 64 stores correspondence information that associates the estimated traveling path with the pedaling state P. The pedaling state estimation unit 68 estimates the pedaling state P using the correspondence information stored in the memory unit 64. The artificial intelligence processing unit 70 may be configured to estimate the traveling path and the pedaling state P from the forward image F. The control unit 62 may include an additional artificial intelligence processing unit, separate from the artificial intelligence processing unit 70, that estimates the pedaling state P from the traveling path estimated by the artificial intelligence processing unit 70.

When the control unit 62 includes an additional artificial intelligence processing unit, the additional artificial intelligence processing unit includes a learning model that outputs information about the pedaling state P in response to input of information about the estimated driving path. For example, the learning model includes an input layer, an intermediate layer, and an output layer, and is configured to be able to update the relationship between the information about the estimated driving path input to the input layer and the information about the pedaling state P output by the output layer in response to the characteristics of the rider of the human-powered vehicle 10. The pedaling state estimation unit 68 may store information about at least one of the curves, steps, road surface gradient, and obstacles of the driving path, and information about the load required by the rider to pedal when the human-powered vehicle 10 actually passes through the driving path. The load required by the rider to pedal when the human-powered vehicle 10 actually passes through the driving path is represented, for example, by the human-powered driving force. The additional artificial intelligence processing unit may learn, as training data, information on at least one of the curves, steps, road surface gradient, and obstacles on the road, and information on the load applied by the rider in pedaling when the human-powered vehicle 10 actually passes through the road.

For example, the control unit 62 is configured to estimate whether the pedaling state P is a first pedaling state P1 or a second pedaling state P2. For example, when it is estimated that there is a change in the road, the control unit 62 estimates the pedaling state P in accordance with the manner of the change in the road. The manner of the change in the road includes, for example, an increase in the curvature of the curve of the road, a decrease in the curvature of the curve, an increase in the height of a step, a decrease in the height of a step, an increase in the magnitude of the road surface gradient, a decrease in the magnitude of the road surface gradient, a change in the shape of an obstacle, a change in the type of obstacle, and a change in the shape of an obstacle. For example, when it is estimated that there is at least one of a curve, a step, a road surface gradient, and an obstacle on the road, the control unit 62 estimates the pedaling state P in accordance with the estimation that there is at least one of a curve, a step, a road surface gradient, and an obstacle on the road.

For example, the pedaling state P includes a first pedaling state P1 and a second pedaling state P2. The first pedaling state P1 is a state in which it is more difficult for the rider to pedal than the second pedaling state P2. For example, if the estimated pedaling state P is the first pedaling state P1, it is estimated that the rider will be in a state in which he or she will be unable to pedal in the future. The state in which the rider will be unable to pedal includes a state in which there is an obstacle in the traveling direction of the human-powered vehicle 10. For example, if the estimated pedaling state P is the first pedaling state P1, it is estimated that the rider will be in a state in which it is preferable for the rider not to pedal. The state in which it is preferable for the rider not to pedal includes at least one of a state in which the human-powered vehicle 10 is passing through a curve, a state in which it is passing through a step, and a state in which it is passing through an obstacle. For example, the control unit 62 may estimate the first pedaling state P1 when it is estimated that there is at least one of a curve, a step, a road gradient, and an obstacle on the traveling road.

For example, the control unit 62 is configured to estimate the pedaling state P as the first pedaling state P1 when the control unit 62 estimates based on the forward image F that it is difficult for the rider pedaling the human-powered vehicle 10 to travel along the road. For example, a state in which it is difficult for the rider pedaling the human-powered vehicle 10 to travel along the road is a state in which it is difficult for the rider of the human-powered vehicle 10 to pedal. For example, a state in which it is difficult for the rider of the human-powered vehicle 10 to pedal includes a state in which the rider cannot pedal and a state in which it is desirable for the rider not to pedal. When the control unit 62 does not estimate the pedaling state P as the first pedaling state P1, it may estimate it to be the second pedaling state P2.

For example, the control unit 62 estimates whether the pedaling state P is the first pedaling state P1 or the second pedaling state P2 depending on at least one of the number, size, and position of obstacles present on the travel path. When an obstacle that makes it difficult for the human-powered vehicle 10 to pass is present on the travel path, the control unit 62 estimates that the pedaling state P is the first pedaling state P1 . For example, the control unit 62 estimates whether the pedaling state P is the first pedaling state P1 or the second pedaling state P2 depending on at least one of the curvature and the number of curves included in the estimated travel path. For example, the control unit 62 estimates that the pedaling state P is the first pedaling state P1 when the curvature of a curve included in the estimated travel path is large. For example, the control unit 62 estimates whether the pedaling state P is the first pedaling state P1 or the second pedaling state P2 depending on at least one of the size of the step included in the estimated traveling path and the number of steps. For example, when the step included in the estimated traveling path is large, the control unit 62 estimates that the pedaling state P is the first pedaling state P1.

For example, the pedaling state P includes the third pedaling state P3 and the fourth pedaling state P4. For example, the second pedaling state P2 includes the third pedaling state P3 and the fourth pedaling state P4. For example, when the estimated pedaling state P is the second pedaling state P2, it is estimated that the rider will be in a state where he or she can pedal. For example, when the estimated pedaling state P is the second pedaling state P2, it is estimated that the rider will be in a state where it is desirable for the rider to pedal. For example, the third pedaling state P3 is a state in which it is more difficult for the rider to pedal than the fourth pedaling state P4.

For example, the control unit 62 is configured to estimate whether the pedaling state P is the first pedaling state P1, the third pedaling state P3, or the fourth pedaling state P4. For example, the control unit 62 is configured to estimate whether the pedaling state P is the third pedaling state P3 or the fourth pedaling state P4. For example, when the control unit 62 estimates that the pedaling state P is the second pedaling state P2, it estimates whether the pedaling state P is the third pedaling state P3 or the fourth pedaling state P4. The control unit 62 may estimate whether the pedaling state P is the first pedaling state P1, the third pedaling state P3, or the fourth pedaling state P4 from the estimated running path.

For example, the control unit 62 is configured to estimate whether the pedaling state P is a first pedaling state P1, a third pedaling state P3, or a fourth pedaling state P4, depending on the characteristics of the curves, steps, road surface gradient, and obstacles estimated to be on the road. For example, the control unit 62 determines the difficulty of the rider's pedaling based on the characteristics of the curves, steps, road surface gradient, and obstacles estimated to be on the road. The characteristics of the curve include, for example, curvature. When the control unit 62 estimates that the road includes a curve, it estimates the pedaling state P depending on the curvature of the curve. The control unit 62 determines that the greater the curvature of the curve, the greater the load required by the rider to pedal. The characteristics of the step include, for example, height. When the control unit 62 estimates that the road includes a step, it estimates the pedaling state P depending on the height of the step. The control unit 62 determines that the greater the height of the step, the greater the load required by the rider to pedal. The characteristics of the road surface gradient include, for example, the magnitude of the gradient. When the control unit 62 estimates that the road includes a road surface gradient, it estimates the pedaling state P according to the road surface gradient. The control unit 62 determines that the greater the road surface gradient, the greater the load required by the rider to pedal. For example, the characteristics of the road surface gradient is an increase in the magnitude of the road surface gradient. For example, the increase in the magnitude of the road surface gradient includes an increase rate of the magnitude of the road surface gradient. For example, the control unit 62 determines that the greater the increase rate of the magnitude of the road surface gradient, the greater the load required by the rider to pedal. The characteristics of the obstacle include, for example, the type of the obstacle. The types of obstacles include obstacles that the human-powered vehicle 10 cannot pass through and obstacles that the human-powered vehicle 10 can pass through. When the control unit 62 estimates that the road includes an obstacle, it estimates the pedaling state P according to the type of obstacle. When the type of obstacle is difficult to pass through, the control unit 62 determines that the greater the load required by the rider to pedal.

The difficulty of pedaling corresponds to at least one of curves, steps, road gradient, and obstacles. The difficulty of pedaling may include at least one of the type of road and the environment of the road. At least one of the type of road and the environment of the road includes, for example, tunnels and jumps. The difficulty of pedaling corresponds to at least one of the difficulty of passing through the road and the load on the rider. The difficulty of passing through the road corresponds, for example, to the operation technique of the rider of the human-powered vehicle 10. The load on the rider corresponds, for example, to the human-powered driving force required to pass through the road.

When the pedaling difficulty corresponds to the passage difficulty of the road, for example, in the case of the road shown in FIG. 4, the control unit 62 estimates the future pedaling state P to be the second pedaling state P2, and in the case of the road shown in FIG. 5, the control unit 62 estimates the future pedaling state P to be the first pedaling state P1. When the pedaling difficulty corresponds to the passage difficulty of the road, for example, when there are fewer trees W than in FIG. 5, the control unit 62 estimates the pedaling state P to be the third pedaling state P3, and when there are even fewer trees W than when the pedaling state P is estimated to be the third pedaling state P3, and when there are no trees W, the control unit 62 estimates the pedaling state P to be the fourth pedaling state P4.

When the difficulty of pedaling corresponds to the degree of difficulty of passage of the road, the control unit 62 estimates the pedaling state P to be a first pedaling state P1, for example, when there is an obstacle above the road and the height from the ground surface to the obstacle above is less than a first height. When the difficulty of pedaling corresponds to the degree of difficulty of passage of the road, the control unit 62 estimates the pedaling state P to be a second pedaling state P2, for example, when there is an obstacle above the road and the height from the ground surface to the obstacle above is equal to or greater than a first height. When the difficulty of pedaling corresponds to the degree of difficulty of passage of the road, the control unit 62 estimates the pedaling state P to be a third pedaling state P3, for example, when there is an obstacle above the road and the height from the ground surface to the obstacle above is equal to or greater than a first height and less than a second height. When the pedaling difficulty corresponds to the passage difficulty of the road, the control unit 62 estimates the pedaling state P to be the fourth pedaling state P4, for example, when there is an obstacle above the road and the height of the obstacle above from the ground surface is equal to or greater than the second height. For example, when the control unit 62 estimates the pedaling state P to be the second pedaling state P2, the control unit 62 may estimate whether the pedaling state P is the third pedaling state P3 or the fourth pedaling state P4. For example, when the control unit 62 is configured to estimate whether the pedaling state P is the first pedaling state P1, the third pedaling state P3, or the fourth pedaling state P4, the control unit 62 may be configured not to estimate the pedaling state P to be the second pedaling state P2.

When the difficulty of pedaling corresponds to the rider's load, for example, the third pedaling state P3 is a state in which the load required by the rider to pedal is equal to or greater than the first load, and the fourth pedaling state P4 is a state in which the load required by the rider to pedal is less than the first load. For example, the load required by the rider to pedal is the load required by the rider to pedal the pedal 20. For example, the load required by the rider to pedal is the load required by the rider to pedal in the future. For example, the load required by the rider to pedal is associated with at least one of the curve, step, road gradient, and obstacle of the road. The load required by the rider to pedal changes depending on the curvature of the curve. When the load required by the rider to pedal is determined in relation to the curvature of the curve, if the curvature of the curve of the road is equal to or greater than a predetermined curvature, the control unit 62 estimates that the load is equal to or greater than the first load. The load required by the rider to pedal varies depending on the height of the step. In a case where the load required by the rider to pedal is determined in relation to the height of the step, and the height of the step on the road is equal to or greater than a predetermined height, the control unit 62 estimates that the load is equal to or greater than a first load. In a case where the load required by the rider to pedal is determined in relation to the road gradient, and the road gradient of the road is equal to or greater than a predetermined gradient, the control unit 62 estimates that the load is equal to or greater than a first load. In a case where the load required by the rider to pedal varies depending on the type of obstacle. In a case where the load required by the rider to pedal is determined in relation to the type of obstacle, and the type of obstacle on the road is equal to or greater than a predetermined type, the control unit 62 estimates that the load is equal to or greater than a first load. The control unit 62 may, for example, calculate a score for each of the curves, steps, road gradient, and obstacles on the road, and estimate the load required by the rider to pedal based on the total value of the scores.

When the difficulty of pedaling corresponds to the rider's load, for example, the third pedaling state P3 is a state in which the rider's pedaling load increases to the second load or more, and the fourth pedaling state P4 is a state in which the rider's pedaling load decreases to less than the second load. In this case, the control unit 62 estimates a future change in the pedaling state P. For example, when the current load is less than the second load, if the rider's pedaling load changes to the second load or more when the human-powered vehicle 10 passes through the estimated driving path in the future, the control unit 62 estimates the pedaling state P to be the third pedaling state P3. For example, when the current load is the second load or more, if the rider's pedaling load changes to less than the second load when the human-powered vehicle 10 passes through the estimated driving path in the future, the control unit 62 estimates the pedaling state P to be the fourth pedaling state P4. The second load may be the current load. The current load is, for example, the driving force detected by the manual driving force detection unit 48.

Whether the load required by the rider to pedal increases to the second load or more may be determined according to the running state of the human-powered vehicle 10 and the estimated running path. For example, if the magnitude of the road surface gradient of the running path estimated by the control unit 62 becomes larger than the current inclination of the human-powered vehicle 10, the control unit 62 estimates that the inclination will become larger in the future. In this case, since it is estimated that the load required by the rider to pedal increases to the second load or more, the control unit 62 estimates the pedaling state P to be the third pedaling state P3. For example, if the curvature of the curve of the running path estimated by the control unit 62 becomes smaller than the curvature of the curve of the running path on which the human-powered vehicle 10 is currently running, it is estimated that the curve will end. In this case, since it is estimated that the load required by the rider to pedal decreases to less than the second load, the control unit 62 estimates the pedaling state P to be the fourth pedaling state P4. If the magnitude of the road surface gradient of the road estimated by the control unit 62 becomes smaller than the current pitch angle of the human-powered vehicle 10, it is estimated that the road surface gradient will become smaller in the future. In this case, since it is estimated that the load required by the rider for pedaling will decrease to less than the second load, the control unit 62 estimates the pedaling state P to be the fourth pedaling state P4. For example, the control unit 62 may calculate a score for each of the curves, steps, road surface gradient, and obstacles on the road, and estimate the change in the load required by the rider for pedaling based on the total score value.

For example, when the control unit 62 estimates that it is easy for the rider pedaling the human-powered vehicle 10 to travel along the road based on the forward image F, the control unit 62 estimates the pedaling state P to be the third pedaling state P3 or the fourth pedaling state P4. The control unit 62 is configured to estimate whether the pedaling state P is the third pedaling state P3 or the fourth pedaling state P4 depending on the road. A state in which it is easy for the rider pedaling the human-powered vehicle 10 to travel along the road is, for example, a state in which it is easy for the rider of the human-powered vehicle 10 to pedal. A state in which it is easy for the rider of the human-powered vehicle 10 to pedal includes, for example, at least one of a state in which the rider can pedal and a state in which it is desirable for the rider to pedal. For example, a state in which it is easy for the rider pedaling the human-powered vehicle 10 to travel along the road is the opposite state to a state in which it is difficult for the rider to pedal. Therefore, the state in which it is easy for the rider pedaling the human-powered vehicle 10 to travel along the roadway is estimated using a configuration opposite to the method for determining the state in which it is difficult for the rider to pedal.

For example, the pedaling state estimation unit 68 is configured to detect the distance D between the object X and the human-powered vehicle 10 based on the forward image F. For example, the pedaling state estimation unit 68 detects the distance D between the object X and the human-powered vehicle 10 from the forward image F by color aperture imaging technology. The pedaling state estimation unit 68 estimates the distance D between a specific position of the boundary B and the human-powered vehicle 10 based on the forward image F. For example, the pedaling state estimation unit 68 detects the distance D between the human-powered vehicle 10 and at least one of the curves, steps, road surface gradients, and obstacles of the travel path estimated by the artificial intelligence processing unit 70. For example, the pedaling state estimation unit 68 detects the distance D between the human-powered vehicle 10 and the point where at least one of the curves, steps, road surface gradients, and obstacles of the travel path estimated by the artificial intelligence processing unit 70 begins. For example, the pedaling state estimation unit 68 detects the distance D between the human-powered vehicle 10 and a point where at least one of the curves, steps, road surface gradients, and obstacles on the road estimated by the artificial intelligence processing unit 70 ends. The pedaling state estimation unit 68 may detect the distance D between the object X and the human-powered vehicle 10 by means other than color aperture imaging technology.

For example, the pedaling state estimation unit 68 is configured to estimate a pedaling state P according to the object X identified by the artificial intelligence processing unit 70 and the distance D detected by the pedaling state estimation unit 68. For example, the pedaling state estimation unit 68 is configured to estimate a pedaling state P according to the boundary B identified by the artificial intelligence processing unit 70 and the distance D between the boundary B and the boundary B detected by the pedaling state estimation unit 68. For example, the pedaling state estimation unit 68 is configured to estimate a pedaling state P according to at least one of the curves, steps, road surface gradient, and obstacles of the traveling path identified by the artificial intelligence processing unit 70, and the distance D to at least one of the curves, steps, road surface gradient, and obstacles of the traveling path estimated by the artificial intelligence processing unit 70. For example, the pedaling state estimation unit 68 is configured to estimate a pedaling state P according to at least one of a curve, a step, a road surface gradient, and an obstacle of the traveling path specified by the artificial intelligence processing unit 70, and a distance D to a point where at least one of the curve, the step, the road surface gradient, and the obstacle of the traveling path estimated by the artificial intelligence processing unit 70 starts. For example, the pedaling state estimation unit 68 is configured to estimate a pedaling state P according to at least one of a curve, the step, the road surface gradient, and an obstacle of the traveling path specified by the artificial intelligence processing unit 70, and a distance D to a point where at least one of the curve, the step, the road surface gradient, and the obstacle of the traveling path estimated by the artificial intelligence processing unit 70 ends. For example, when it is estimated that the road includes a curve, a step, a road gradient, or an obstacle, and the distance D to the curve, step, road gradient, or obstacle is within a predetermined range DX, the pedaling state estimation unit 68 estimates a pedaling state P corresponding to the curve, step, road gradient, or obstacle. For example, the predetermined range DX is set based on the timing to change the control state of the electric device 40. For example, the predetermined range DX is set in advance according to the distance D required to change the control state of the electric device 40 at the timing when the human-powered vehicle 10 enters the predetermined range DX by traveling forward.

The control unit 62 may change the distance D according to information about the current human-powered vehicle 10. For example, the control unit 62 may change the distance D according to the vehicle speed. The control unit 62 may change the distance D according to information about driving, for example, human-powered driving force. The control unit 62 may change the distance D according to information about braking, for example, braking operation. For example, the control unit 62 is configured to estimate the pedaling state P according to the vehicle speed and the distance over which the road gradient estimated from the forward image F continues. The control unit 62 may be configured to determine that the future pedaling state P is the fourth pedaling state P4 even when the pedaling state estimation unit 68 estimates that the future pedaling state P is the third pedaling state P3 from the forward image F. In this case, for example, the control unit 62 determines that the future pedaling state P is the fourth pedaling state P4 based on the vehicle speed and the distance over which the uphill road gradient estimated from the forward image F continues. For example, when the control unit 62 determines that the vehicle can coast up an upward road gradient at the current vehicle speed, it determines that the future pedaling state P is the fourth pedaling state P4 based on the vehicle speed and the distance the upward road gradient will continue estimated from the forward image F. In this case, for example, the artificial intelligence processing unit 70 may be configured to estimate whether the vehicle can coast up an upward road gradient at the current vehicle speed.

The process by which the control unit 62 estimates the pedaling state P will be described with reference to the flowchart in FIG. 6. For example, when power is supplied to the control unit 62, the control unit 62 starts the process and proceeds to step S11 of the flowchart shown in FIG. 6. When the flowchart in FIG. 6 ends, the control unit 62 repeats the process from step S11 after a predetermined period, for example, until the supply of power is stopped.

In step S11, the control unit 62 determines whether or not a forward image F has been input. If a forward image F has been input, the control unit 62 proceeds to step S12. If a forward image F has not been input, the control unit 62 ends the process.

In step S12, the control unit 62 estimates the driving path and proceeds to step S13. For example, the control unit 62 inputs the input front image F to the artificial intelligence processing unit 70 and estimates the driving path. For example, the processing of step S12 is executed by the artificial intelligence processing unit 70.

In step S13, the control unit 62 estimates the pedaling state P and ends the process. In step S13, the control unit 62 estimates the pedaling state P based on the traveling path estimated by the process of step S12. For example, the process of step S13 is executed by the pedaling state estimation unit 68.

When the electric device 40 includes the transmission 42, for example, the control unit 62 is configured to change the gear shift conditions according to the estimated pedaling state P. The control state of the transmission 42 according to the gear shift conditions corresponds to the control state of the electric device 40. For example, the control unit 62 changes the gear shift conditions to gear shift conditions suitable for the estimated pedaling state P. For example, the storage unit 64 may be configured to store gear shift conditions associated with each of the first pedaling state P1, the second pedaling state P2, the third pedaling state P3, and the fourth pedaling state P4. When the control unit 62 estimates the pedaling state P, it changes the gear shift conditions to gear shift conditions associated with the estimated pedaling state P.

For example, the control unit 62 is configured to change the gear shift threshold in accordance with the estimated pedaling state P. For example, gear shift thresholds corresponding to each of the first pedaling state P1, the second pedaling state P2, the third pedaling state P3, and the fourth pedaling state P4 are stored in the memory unit 64. For example, the control unit 62 selects the gear shift threshold corresponding to the current pedaling state P and changes the gear shift threshold. Preferably, the control unit 62 is configured to change the gear shift conditions by changing at least one of the upper limit value and the lower limit value in accordance with the estimated pedaling state P.

The control unit 62 is configured to change the gear change conditions so that when the pedaling state P is the first pedaling state P1, the gear ratio R is changed less frequently than when the pedaling state P is the second pedaling state P2. For example, the upper limit value corresponding to the first pedaling state P1 is greater than the upper limit value corresponding to the second pedaling state P2. For example, the lower limit value corresponding to the first pedaling state P1 is less than the lower limit value corresponding to the second pedaling state P2.

For example, the control unit 62 is configured to change the gear shift conditions when the pedaling state P is the third pedaling state P3 so that the gear ratio R is changed less frequently than when the pedaling state P is the fourth pedaling state P4. For example, the upper limit value corresponding to the third pedaling state P3 is greater than the upper limit value corresponding to the fourth pedaling state P4. For example, the lower limit value corresponding to the third pedaling state P3 is less than the lower limit value corresponding to the fourth pedaling state P4. The gear shift threshold value corresponding to the second pedaling state P2 may be the same as the gear shift threshold value corresponding to the third pedaling state P3.

For example, the control unit 62 is configured to change the gear shift conditions so that, when the pedaling state P is the third pedaling state P3, the gear ratio R is changed less frequently than when the pedaling state P is the fourth pedaling state P4. For example, the upper limit value corresponding to the third pedaling state P3 is greater than the upper limit value corresponding to the fourth pedaling state P4. For example, the lower limit value corresponding to the third pedaling state P3 is greater than the lower limit value corresponding to the fourth pedaling state P4. The lower limit value corresponding to the third pedaling state P3 may be less than or equal to the lower limit value corresponding to the fourth pedaling state P4. For example, the control unit 62 is configured to change the gear shift conditions so that, when the future pedaling state P based on the forward image F is the third pedaling state, even if the current pedaling state P is the fourth pedaling state, the gear shift conditions are changed so that the gear ratio R is changed less frequently. For example, the control unit 62 is configured to change the gear change conditions so that the frequency of changes to increase the gear ratio R is reduced when the road along which the human-powered vehicle 10 will travel in the future based on the forward image F is an uphill road, even if the road along which the human-powered vehicle 10 is currently traveling is downhill or flat.

For example, the control unit 62 is configured to change the gear shift conditions so that when the pedaling state P is the third pedaling state P3, the gear ratio R is changed to a smaller value more frequently than when the pedaling state P is the fourth pedaling state P4. For example, the lower limit value corresponding to the third pedaling state P3 is greater than the lower limit value corresponding to the fourth pedaling state P4. For example, the upper limit value corresponding to the third pedaling state P3 is less than the upper limit value corresponding to the fourth pedaling state P4. The upper limit value corresponding to the third pedaling state P3 may be greater than or the same as the upper limit value corresponding to the fourth pedaling state P4.

The process in which the control unit 62 controls the transmission 42 will be described with reference to the flowchart in FIG. 7. For example, when power is supplied to the control unit 62, the control unit 62 starts the process and proceeds to step S21 in the flowchart shown in FIG. 7. When the flowchart in FIG. 7 ends, the control unit 62 repeats the process from step S21 after a predetermined period, for example, until the supply of power is stopped.

In step S21, the control unit 62 acquires the estimated pedaling state P and proceeds to step S22. For example, the control unit 62 acquires the pedaling state P estimated in step S13 of FIG. 6.

In step S22, the control unit 62 determines whether the future pedaling state P will change. For example, if the estimated pedaling state P differs from the current pedaling state P, the control device 60 determines that the future pedaling state P will change. The current pedaling state P may be acquired, for example, by at least one of the human-powered driving force detection unit 48 and the crank rotation sensor 50. The current pedaling state P may be, for example, an estimated pedaling state P acquired a predetermined period ago. If the future pedaling state P will not change, the control unit 62 proceeds to step S24. If the future pedaling state P will change, the control unit 62 proceeds to step S23.

In step S23, the control unit 62 changes the gear shifting conditions according to the future pedaling state P, and proceeds to step S24. In step S24, the control unit 62 controls the transmission 42 according to the gear shifting conditions, and ends the process. If the control unit 62 changes the gear shifting conditions in step S23, it controls the transmission 42 under the changed gear shifting conditions in step S24. Therefore, when the gear shifting conditions are changed in step S23, the control state of the transmission 42 is changed.

For example, the control unit 62 is configured to be able to switch the gear shift state. The gear shift state includes a first gear shift state in which the gear ratio R is changed according to the gear shift conditions, and a second gear shift state in which the gear ratio R is not changed according to the gear shift conditions. The control unit 62 is configured to switch the gear shift state between the first gear shift state and the second gear shift state according to the estimated pedaling state P. The gear shift state corresponds to the control state of the electric device 40.

For example, the control unit 62 selects the second gear shifting state when the estimated pedaling state P is a state in which the rider is unable to pedal and a state in which it is desirable for the rider not to pedal in the future. For example, the control unit 62 switches the gear shifting state to the first gear shifting state when the pedaling state P is the second pedaling state P2. For example, the control unit 62 switches the gear shifting state to the second gear shifting state when the pedaling state P is the first pedaling state P1.

The process in which the control unit 62 controls the transmission 42 will be described with reference to the flowchart in FIG. 8. For example, when power is supplied to the control unit 62, the control unit 62 starts the process and proceeds to step S31 in the flowchart shown in FIG. 8. When the flowchart in FIG. 8 ends, the control unit 62 repeats the process from step S31 after a predetermined period, for example, until the supply of power is stopped.

In step S31, the control unit 62 acquires the estimated pedaling state P and proceeds to step S32. For example, the control unit 62 acquires the pedaling state P estimated in step S13 of FIG. 6.

In step S32, the control unit 62 determines whether or not to switch the gear shift state. For example, when the current pedaling state P is the first pedaling state P1 and the estimated pedaling state P is the second pedaling state P2, the control unit 62 determines to switch the gear shift state. For example, when the current pedaling state P is the second pedaling state P2 and the estimated pedaling state P is the first pedaling state P1, the control unit 62 determines to switch the gear shift state. If the control unit 62 does not want to switch the gear shift state, it ends the process. If the control unit 62 wants to switch the gear shift state, it proceeds to step S33.

The control unit 62 switches the gear shift state in step S33 and ends the process. For example, when the current pedaling state P is the first pedaling state P1 and the estimated pedaling state P is the second pedaling state P2, the control unit 62 switches the gear shift state to the first gear shift state. For example, when the current pedaling state P is the second pedaling state P2 and the estimated pedaling state P is the first pedaling state P1, the control unit 62 switches the gear shift state to the second gear shift state. When the gear shift state is changed in step S33, the control state of the gear shift device 42 is changed.

<Modification>
The explanations of the embodiments are examples of forms that a control device for a human-powered vehicle according to the present disclosure can take, and are not intended to limit the forms. A control device for a human-powered vehicle according to the present disclosure can take forms, for example, modified versions of the embodiments shown below, or a combination of at least two mutually consistent modified versions. In the following modified versions, parts that are common to the embodiments are given the same reference numerals as in the embodiments, and descriptions thereof are omitted.

Instead of or in addition to the gear shift device 42, the electric device 40 may include at least one of a motor that assists in propulsion of the human-powered vehicle 10, an electric suspension, an electric adjustable seat post, a display device, an alarm device, and a braking device. When the electric device 40 includes a motor, the control unit 62 may be configured to change the propulsive force of the motor and the corresponding control state in accordance with the load required for pedaling by the rider corresponding to the estimated pedaling state P. When the electric device 40 includes an electric suspension, the control unit 62 may be configured to change the repulsive force of the suspension and the corresponding control state in accordance with the load required for pedaling by the rider corresponding to the estimated pedaling state P. When the electric device 40 includes an electric adjustable seat post, the control unit 62 may be configured to change the height of the seat post and the corresponding control state in accordance with the load required for pedaling by the rider corresponding to the estimated pedaling state P. When the electric device 40 includes a display device, the control unit 62 may be configured to change a control state related to the display in accordance with the load required for pedaling by the rider corresponding to the estimated pedaling state P. When the electric device 40 includes a notification device, the control unit 62 may be configured to change a control state related to the notification in accordance with the load required for pedaling by the rider corresponding to the estimated pedaling state P.
The process of the control unit 62 controlling the electrically-driven device 40 will be described with reference to the flowchart of Fig. 9. For example, when power is supplied to the control unit 62, the control unit 62 starts the process and proceeds to step S41 of the flowchart shown in Fig. 9. When the flowchart of Fig. 9 ends, the control unit 62 repeats the process from step S41 after a predetermined period, for example, until the supply of power is stopped.
In step S41, the control unit 62 acquires the estimated pedaling state P and proceeds to step S42. The control unit 62 acquires the pedaling state P estimated in the process of FIG. 6, for example. The control unit 62 determines whether or not to switch the control state in step S42. For example, when the current pedaling state P is the first pedaling state P1 and the estimated pedaling state P is the second pedaling state P2, the control unit 62 determines to switch the control state. For example, when the current pedaling state P is the second pedaling state P2 and the estimated pedaling state P is the first pedaling state P1, the control unit 62 determines to switch the control state. If the control unit 62 does not switch the control state, it ends the process. If the control unit 62 switches the control state, it moves to step S43. The control unit 62 switches the control state in step S43 and ends the process.

- The control unit 62 may be configured to execute only one of the processes in the flowchart of FIG. 7 and the processes in the flowchart of FIG. 8.

- The pedaling state estimation unit 68 may estimate the pedaling state P based on information about the front of the human-powered vehicle 10 other than the front image F, in addition to the front image F. For example, the human-powered vehicle 10 is provided with a laser device. The control unit 62 estimates the travel path from the front image F and information about the front of the human-powered vehicle 10 acquired by the laser device. The pedaling state estimation unit 68 corrects the distance D detected by the artificial intelligence processing unit 70 based on the distance to the object X acquired by the laser device, and estimates the pedaling state P.

- The second pedaling state P2 may include a fifth pedaling state P5 in addition to the third pedaling state P3 and the fourth pedaling state P4. The fifth pedaling state P5 is a state in which it is more difficult for the rider to pedal than the third pedaling state P3. The second pedaling state P2 may include four or more pedaling states P.

When the gear shift state is the second gear shift state, the control unit 62 may control the gear shift device 42 to change the gear ratio R in response to the rider's operation of the gear shift operating unit.

The term "at least one" as used herein means "one or more" of the desired options. As an example, the term "at least one" as used herein means "only one option" or "both of two options" if the number of options is two. As another example, the term "at least one" as used herein means "only one option" or "any combination of two or more options" if the number of options is three or more.

10... human-powered vehicle, 40... electric device, 42... transmission, 44... imaging device, 60... control device, 62... control unit, 68... pedaling state estimation unit, 70... artificial intelligence processing unit.

Claims (26)

A control device for a human-powered vehicle,
A control unit configured to control an electric device of the human-powered vehicle,
the control unit is configured to estimate a travel path based on a forward image acquired from an imaging device, estimate a future pedaling state of the rider based on a pedaling difficulty of the rider on the estimated travel path, and control the electric device in accordance with a control state of the electric device that is associated in advance with the estimated pedaling state ,
The electric device includes at least one of a transmission that changes a gear ratio, a motor that assists the human-powered vehicle with propulsive force, an electric suspension, and an electric adjustable seat . 2. The control device according to claim 1, wherein the difficulty of the rider's pedaling corresponds to at least one of a degree of difficulty of passing the road regarding whether the rider can pedal or not, and a load required for the rider to pedal . 3. The control device according to claim 1, wherein the control unit includes a pedaling state estimation unit configured to estimate the travel route along which the human-powered vehicle will travel in the future based on the forward image, and to estimate the pedaling state according to the estimated travel route. The control device according to claim 3 , wherein the pedaling state estimator comprises an artificial intelligence processor. The artificial intelligence processing unit is
The vehicle is configured to estimate the driving path based on the forward image,
The control device according to claim 4 , wherein the pedaling state estimating unit is configured to estimate the pedaling state according to the traveling path estimated by the artificial intelligence processing unit. The artificial intelligence processing unit is
Identifying an object on the road based on the forward image;
The control device according to claim 5 , configured to estimate the travel path based on the identified object. The artificial intelligence processing unit is
Identifying a boundary between an area that is the road and an area that is not the road in the forward image based on the identified object;
The control device of claim 6 , configured to estimate the path based on the identified boundary. The pedaling state estimation unit,
A device configured to detect a distance between the object and the human-powered vehicle based on the forward image,
The control device according to claim 6 or 7 , configured to estimate the pedaling state according to the object identified by the artificial intelligence processing unit and the distance detected by the artificial intelligence processing unit. the pedaling state includes a first pedaling state and a second pedaling state,
the first pedaling state is a state in which it is more difficult for the rider to pedal than the second pedaling state,
The control device according to claim 1 , wherein the control unit is configured to estimate whether the pedaling state is the first pedaling state or the second pedaling state. the second pedaling state includes a third pedaling state and a fourth pedaling state,
the third pedaling state is a state in which it is more difficult for the rider to pedal than the fourth pedaling state,
The control device according to claim 9 , wherein the control unit is configured to estimate whether the pedaling state is the first pedaling state, the third pedaling state, or the fourth pedaling state. 11. The control device according to claim 9 or 10, wherein the control unit is configured to estimate the pedaling state to be the first pedaling state when the control unit estimates, based on the forward image, that it is difficult for the rider pedaling the human - powered vehicle to travel on the travel path. the pedaling states include a third pedaling state and a fourth pedaling state,
the third pedaling state is a state in which it is more difficult for the rider to pedal than the fourth pedaling state,
The control device according to claim 1 , wherein the control unit is configured to estimate whether the pedaling state is the third pedaling state or the fourth pedaling state. the third pedaling state is a state in which a load required by the rider for pedaling is equal to or greater than a first load,
13. The control device according to claim 10 or 12 , wherein the fourth pedaling state is a state in which a pedaling load required by the rider is less than the first load. the third pedaling state is a state in which a load required by the rider for pedaling increases to be equal to or greater than the second load,
14. The control device according to claim 10, 12 or 13 , wherein the fourth pedaling state is a state in which a pedaling load required by the rider decreases to less than the second load. 15. The control device according to any one of claims 10 and 12 to 14, wherein the control unit is configured to, when the control unit estimates, based on the forward image, that it is easy for the rider pedaling the human-powered vehicle to travel on the travel road, estimate whether the pedaling state is the third pedaling state or the fourth pedaling state depending on the travel road. 16. The control device according to claim 11, wherein, when it is estimated that the travel path has at least one of a curve, a step, a road surface gradient, and an obstacle, the control unit estimates the pedaling state in accordance with an estimation that the travel path has at least one of the curve, the step, the road surface gradient, and the obstacle . The control device according to claim 1 , wherein the electric device includes at least the transmission. The control unit is
The transmission is configured to control the transmission so as to change the gear ratio in accordance with a gear change condition,
The control device according to claim 17 , configured to change the gear shift condition depending on the estimated pedaling state. the electric device includes at least the transmission;
The control unit is
The transmission is configured to control the transmission so as to change the gear ratio in accordance with a gear change condition,
The shift condition is changed in response to the estimated pedaling state,
12. The control device according to claim 9, wherein the control device is configured to change the gear change condition when the pedaling state is the first pedaling state such that the gear ratio is changed less frequently than when the pedaling state is the second pedaling state. the electric device includes at least the transmission;
The control unit is
The transmission is configured to control the transmission so as to change the gear ratio in accordance with a gear change condition,
The control device according to claim 10 , configured to change the gear shift condition in response to the estimated pedaling state. 21. The control device according to claim 20, wherein the control unit is configured to change the gear change condition such that, when the pedaling state is the third pedaling state, the gear ratio is changed to an increased gear less frequently than when the pedaling state is the fourth pedaling state. 22. The control device according to claim 20, wherein the control unit is configured to change the gear change condition such that, when the pedaling state is the third pedaling state, the gear ratio is changed to a smaller value more frequently than when the pedaling state is the fourth pedaling state. The shift condition includes a shift threshold value,
The control unit is
a control unit configured to control the transmission to change the gear ratio in response to a comparison between a parameter related to at least one of a running state and a running environment of the human-powered vehicle and the gear shift threshold;
23. A control device as claimed in any one of claims 18 to 22 , configured to vary the shift threshold depending on the estimated pedalling state. 24. The control device according to claim 23, wherein the parameters include at least one of a human-powered driving force input to the human-powered vehicle, a rotational speed of a crank of the human-powered vehicle, a vehicle speed of the human-powered vehicle, and an inclination of the human-powered vehicle . The shift threshold value includes an upper limit value and a lower limit value,
25. The control device according to claim 23 , wherein the control unit is configured to change the gear shift condition by changing at least one of the upper limit value and the lower limit value in accordance with the estimated pedaling state. The control unit is configured to be able to switch a gear shift state,
the gear shift state includes a first gear shift state in which the gear ratio is changed in accordance with the gear shift condition, and a second gear shift state in which the gear ratio is not changed in accordance with the gear shift condition,
26. The control device according to claim 18 , wherein the control unit is configured to switch the gear shift state between the first gear shift state and the second gear shift state depending on the estimated pedaling state.