DE102020201956A1 - DETERMINATION DEVICE, CONTROL SYSTEM, COMMUNICATION SYSTEM, LEARNING MODEL, METHOD FOR GENERATING THE LEARNING MODEL, COMPUTER PROGRAM AND STORAGE MEDIUM - Google Patents

Abstract

The present invention relates to a determination device 100, a control system, a communication system, a learning model 160, a method for generating the learning model 160, a computer program and a storage medium M, wherein a determination device 100 determines a physical condition of a driver on the basis of driving information the driving of a human-powered vehicle 1 is determined.

Description

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

A system for collecting information about a bicycle is known. For example, the JP 2015 - 131 533 A a system for collecting information about components mounted on a bicycle and notifying the rider of the information collected.

At present, it is necessary to use the information gathered about a driver's driving with a more convenient method.

The present invention provides a determination device for determining the physical condition of a driver of a human-powered vehicle, a control system, a communication system, a learning model, a method for generating the learning model, a computer program and a storage medium.

A determination device according to a first aspect of the present invention determines the physical condition of a driver based on driving information of a human-powered vehicle.

According to this determination device, the physical condition of the driver is determined based on the driving information of the human-powered vehicle so that e.g. it can be detected how far the driver of the human-powered vehicle is still fit to drive.

The determining device of a second aspect according to the first aspect determines the physical condition by using a learning model configured to output data indicating the physical condition according to an input of the driving information. According to this determination device, the learning model that outputs a physical state in accordance with an input of the driving information is used, so that the physical state of the driver can be determined efficiently.

According to the determining device of a third aspect according to the first or second aspect, the driving information includes information about at least one of the following variables: speed, acceleration, geographical position, angle of inclination, driving distance, driving time, cadence, power, rotational speed of a wheel, driving distance, wind direction, wind speed, Outside temperature and humidity of the human-powered vehicle. According to this determination device, the physical condition of the driver can be determined from various driving information.

The determining device of a fourth aspect according to any one of the first to third aspects determines the physical condition of the driver based on the driving information and driver information about the driver of the human-powered vehicle. With this determination device, the physical condition of the driver can be determined on the basis of the driving information and the driver information, so that e.g. it can be detected how far the driver of the human-powered vehicle is still fit to drive.

The determining device of a fifth aspect according to the fourth aspect determines the physical condition by using a learning model configured to output data indicating the physical condition according to inputs of the driving information and the driver information. According to this determination device, the learning model that outputs a physical state according to the inputs of the driving information and the driver information is used, so that the physical state of the driver can be determined efficiently.

In the determining device of a sixth aspect according to the fourth or fifth aspect, the driver information includes information on at least one of the following features: gender, age, height, weight, muscle amount, fat amount, heartbeat, body temperature, sweating, blood pressure, breathing, myoelectric potential and brain waves Driver. According to this determination device, the physical condition of the driver can be determined from various driver information.

A control system according to a seventh aspect of the present invention includes the determining device according to any one of the first to sixth aspects; and a control device that controls a component of the human-powered vehicle based on the information output from the determination device. According to this control system, each of the components of the human-powered vehicle is controlled in accordance with the determination result obtained by the determining device, thereby enabling control such that the physical burden on the driver is reduced when the driver's physical condition deteriorates, for example.

In the control system of an eighth aspect according to the seventh aspect, the component includes a notification device. According to this control system, the physical condition determined by the determining device can be reported to the driver.

In the control system of a ninth aspect according to the eighth aspect, the control device controls the notification device to notify information on a recommended travel route. According to this control system, a recommended travel route that can reduce the physical burden on the driver can be reported when the driver's physical condition deteriorates, for example.

In the control system of a tenth aspect according to the eighth or ninth aspect, the control device controls the reporting device to report information on at least one of eating and / or drinking and a rest. According to this control system, the driver can be requested to have food and / or drinks or to take a break, e.g. the physical condition of the driver deteriorates.

In the control system of an eleventh aspect according to any one of the seventh to tenth aspects, the component includes a transmission device. According to this control system, the operation of the transmission device can be controlled so that the physical load on the driver is e.g. is reduced in case of deterioration in the physical strength of the driver.

A communication system according to a twelfth aspect of the present invention comprises the determination device according to any one of the first to sixth aspects; and a communication device that communicates with an external device, and the communication device transmits the information output from the determination device to the external device. According to this communication system, the information on the physical condition of the driver determined by the determining device can be reported to the external device. The external device includes a communication device mounted on another human operated vehicle. The external device can be a server device that collects information about the physical condition of the driver.

In the communication system of a thirteenth aspect according to the twelfth aspect, the information output from the determination device also includes warning information about the physical condition. According to this communication system, for example, the warning information can be output when the physical condition of the driver deteriorates or deterioration is predicted.

In the communication system of a fourteenth aspect according to the twelfth or thirteenth aspect, the communication device receives information on the physical condition of another driver transmitted from a determination device of another human-powered vehicle and from a communication device of that other human-powered vehicle. According to this communication system, the information on the physical condition of another driver transmitted from the communication device of another human-powered vehicle is received. If several drivers drive in a group and, for example, the physical condition of one driver deteriorates, the driver can be asked to eat and / or drink or to rest or to reduce the driving speed in order to reduce the physical strain.

In the communication system of a fifteenth aspect according to any one of the twelfth to fourteenth aspects, the communication device receives information on the driving environment including information on at least one traffic light and a traffic volume from the external device. According to this communication system, the driving environment information is received from the external device so that e.g. a recommended route to be reported to the driver can be generated on the basis of the received driving environment information.

The communication system of a sixteenth aspect according to the fifteenth aspect further includes a notification device that notifies information about a recommended travel route based on the travel environment information. According to this communication system, the driver can be informed of the recommended travel route based on the driving environment information so that the driver's physical burden can be reduced e.g. can be reduced when the driver's physical condition deteriorates.

A learning model according to a seventeenth aspect of the present invention includes an input plane into which driving information about driving a human-powered vehicle is input; an intermediate plane that learns a relationship between the driving information and a physical condition of a driver; and an output level from which data indicating the physical condition are output, the learning model being used in the calculation processes in the Intermediate level based on the information entered into the input level and used to output data indicating the physical condition from the output level. According to this learning model, the computer can achieve an execution environment for outputting data indicating the physical condition of the driver according to the driving information of the human-powered vehicle.

A learning model generation method according to an eighteenth aspect generates a learning model used in the processes of determining the physical condition of a driver from driving information about driving a human-powered vehicle based on driving information and data indicative of the physical condition of the driver , using a computer. According to this method of generating the learning model, by collecting the driving information of the human-powered vehicle and data indicating the physical condition of the driver, the learning model for determining the physical condition of the driver can be generated.

A computer program according to a nineteenth aspect of the present invention is a computer program that causes a computer to execute a process of determining the physical condition of a driver based on driving information about driving a human-powered vehicle. According to this computer program, the execution environment for outputting the data indicating the physical condition of the driver based on the driving information of the human-powered vehicle can be achieved by the computer.

A storage medium according to a twentieth aspect according to the nineteenth aspect, which stores the above-described computer program. According to this storage medium, by installing the computer program stored in the storage medium in the computer, the execution environment for determining the physical condition of the driver based on the driving information of the human-powered vehicle by the computer can be achieved.

According to the present application, it is possible to determine the physical condition of the driver of a human-powered vehicle.

• 1 eine Seitenansicht eines menschlich angetriebenen Fahrzeugs ist, an dem eine Bestimmungsvorrichtung gemäß der Ausführungsform 1 angebracht ist;
• 2 ein Blockdiagramm ist, das ein Beispiel für die interne Ausgestaltung der Bestimmungsvorrichtung gemäß der Ausführungsform 1 zeigt;
• 3 ein Konfigurationsdiagramm eines Lernmodells aus dem ersten Einsatzbeispiel ist;
• 4 ein Flussdiagramm ist, das ein Bestimmungsverfahren in Ausführungsform 1 veranschaulicht;
• 5 ein Blockdiagramm ist, das ein Steuerungssystem gemäß der Ausführungsform 2 darstellt;
• 6 ein Flussdiagramm ist, das einen Steuerungsvorgang in Ausführungsform 2 veranschaulicht;
• 7 eine schematische Ansicht ist, die ein Beispiel für Informationen darstellt, die auf einer Anzeigevorrichtung angezeigt werden sollen;
• 8 eine schematische Ansicht ist, die ein weiteres Beispiel für Informationen darstellt, die auf der Anzeigevorrichtung angezeigt werden sollen;
• 9 ein Blockdiagramm ist, das ein Kommunikationssystem gemäß der Ausführungsform 3 veranschaulicht;
• 10 ein Flussdiagramm ist, das einen Steuerungsvorgang in Ausführungsform 3 veranschaulicht;
• 11 ein Blockdiagramm ist, das ein Beispiel für die Konfiguration einer Servervorrichtung veranschaulicht;
• 12 ein Flussdiagramm ist, das ein Verfahren zur Erzeugung eines Lernmodells veranschaulicht;
• 13 ein Flussdiagramm ist, das ein Bestimmungsverfahren in Ausführungsform 5 veranschaulicht;
• 14 ein Konfigurationsdiagramm eines Lernmodells im zweiten Einsatzbeispiel ist; und
• 15 ein Konfigurationsdiagramm eines Lernmodells aus dem dritten Einsatzbeispiel ist.

A more complete assessment of the invention and many of its advantages will be readily achieved when better understood from the following detailed description in conjunction with the accompanying drawings, in which:

• 1 Fig. 3 is a side view of a human-powered vehicle to which a determination device according to Embodiment 1 is attached;
• 2 Fig. 13 is a block diagram showing an example of the internal configuration of the determination device according to Embodiment 1;
• 3 Fig. 13 is a configuration diagram of a learning model of the first application example;
• 4th Fig. 13 is a flowchart illustrating a determination method in Embodiment 1;
• 5 Fig. 13 is a block diagram illustrating a control system according to Embodiment 2;
• 6th Fig. 13 is a flowchart illustrating a control process in Embodiment 2;
• 7th Fig. 13 is a schematic view showing an example of information to be displayed on a display device;
• 8th Fig. 3 is a schematic view showing another example of information to be displayed on the display device;
• 9 Figure 13 is a block diagram illustrating a communication system according to Embodiment 3;
• 10 Fig. 10 is a flow chart illustrating a control process in Embodiment 3;
• 11 Fig. 13 is a block diagram illustrating an example of the configuration of a server device;
• 12 Figure 3 is a flow diagram illustrating a method for generating a learning model;
• 13 Fig. 13 is a flowchart illustrating a determination method in Embodiment 5;
• 14th Fig. 10 is a configuration diagram of a learning model in the second application example; and
• 15th Fig. 3 is a configuration diagram of a learning model from the third application example.

The present invention will be described below with reference to the drawings illustrating embodiments thereof.

1 Figure 13 is a side view of a human powered vehicle 1 , on the one Determining device 100 according to embodiment 1 is attached.

The human-powered vehicle 1 includes vehicles that at least partially use human power as a drive source for driving. Vehicles that use only an internal combustion engine or an electric motor as a power source are human-powered vehicles 1 excluded in the present embodiment. The human-powered vehicle 1 is a bicycle that includes a racing bike, a mountain bike, a hybrid bike, a utility bike, or the like. The human-powered vehicle 1 may be an electric bicycle with an electric assist mechanism that drives the human-powered vehicle 1 supported by electrical energy.

The human-powered vehicle 1 includes a vehicle main body 10 , a front wheel 12 , a rear wheel 14th , a handlebar 16 , a saddle 18th and a drive mechanism 20th . In the following description, the terms indicating the directions, such as front-back, right-left, up-down, etc., are used in relation to the directions seen when the rider is in the saddle 18th of the human-powered vehicle 1 sits.

The vehicle main body 10 includes a frame 10A and a front fork 10B . The front wheel 12 is rotatable on the front fork 10B stored. The rear wheel 14th is rotatable on the frame 10A stored. The handlebar 16 includes a handle 16A and is on the frame 10A propped up to the direction of travel of the front wheel 12 to be able to change.

The drive mechanism 20th transfers the human-generated drive force to the rear wheel 14th . The drive mechanism 20th includes a crank 22nd , a front sprocket 24A , a rear sprocket 24B , a chain 26th and a pair of pedals 28 and 28 .

The crank 22nd includes a right crank 22A , a left crank 22B and a crankshaft 22C . The crankshaft 22C is rotatable on the frame 10A stored. The right crank 22A and the left crank 22B are with the crankshaft 22C coupled. One of the pair of pedals 28 and 28 can be turned on the right crank 22A stored, while the other of the two pedals rotatable on the left crank 22B is stored.

The front sprocket 24A is with the crankshaft 22C coupled so that it becomes integral with the crankshaft 22C turns. The front sprocket assembly 24A includes, for example, multiple front sprockets with different outside diameters. The outer diameter of the plurality of front sprockets increases from the center plane of the vehicle main body 10 outward in a direction parallel to the central axis of rotation of the crankshaft 22C .

The rear sprocket assembly 24B is rotatable on a hub (not shown) of the rear wheel 14th stored. The rear sprocket assembly 24B includes, for example, multiple rear sprockets with different outside diameters. The outer diameter of the plurality of rear sprockets decreases from the median plane of the vehicle main body 10 outwards in a direction parallel to the rotational center axis of the hub assembly.

The chain 26th is about the front sprocket group 24A and the rear sprocket group 24B guided. When the crank 22nd by a human driving force exerted on the pair of pedals 28 and 28 acts, rotates forward, the front sprocket assembly rotates 24A together with the crank 22nd Forward. The rotation of the front sprocket assembly 24A is about the chain 26th onto the rear sprocket assembly 24B transferred to thereby the rear wheel 14th to turn. Instead of the chain 26th a belt or a shaft can be used.

The human-powered vehicle 1 can also be equipped with an operating device 30th , Transmission devices 32 and 33 , an adjustable seat post 34 , a suspension device 36 and a battery unit 38 be equipped.

The operating device 30th contains a control unit 30A which is operated by the driver. An example for the control unit 30A is a control button or switch. Another example of the control unit 30A are on the handlebar 16 attached brake lever.

With each lateral inclination of the brake lever, a gear stage or a transmission ratio for the transmission device 32 be changed. The operating device 30th accepted by the control unit 30A the control of various components in human-powered vehicles 1 , such as switching the adjustable seat post 34 , switching the suspension device 36 etc.

The operating device 30th can be connected for communication with each of the components to generate a signal corresponding to one on the control unit 30A performed operation to the transmission device 32 or to transfer the other components. In one example this is an operating device 30th connected to with the transmission device 32 and the other components via a communication line or an electrical line that enables powerline communication (PLC) to communicate. In another example is the operating device 30th for communication with the transmission device 32 and connected to the other components by a wireless communication unit that enables wireless communication. When operating the control unit 30A becomes a control signal for shifting the gear stage of the transmission device 32 eg to the transmission device 32 transfer. In response to the signal, the transmission device switches 32 the gear stage or the gear ratio.

The transmission device 32 is one of the components of the human-powered vehicle 1 , its operation in response to an from the operating device 30th transmitted control signal is controlled. The transmission device 32 includes an electric actuator 32A and a chain guide 32B . An example of the electric actuator 32A is an electric motor. The transmission device 32 drives the electric actuator 32A to the chain guide 32B to actuate, causing the rear sprocket to which the chain 26th is guided, changed and the gear ratio of the human-powered vehicle 1 is switched.

An example of the transmission device 32 is an external gear. If it is the transmission device 32 is an external transmission, contains the rear sprocket assembly 24B several rear sprockets with different outside diameters. Used by the control device 30th receiving a control signal instructing an upshift actuates the transmission device 32 the chain guide 32B by the electric actuator 32A so that the rear sprocket to the present the chain 26th is performed, is switched to a rear sprocket with a smaller outer diameter. Used by the control device 30th receiving a control signal instructing a downshift actuates the transmission device 32 the chain guide 32B by the electric actuator 32A so that the rear sprocket to the present the chain 26th is performed, is changed to a rear sprocket with a larger outer diameter.

The transmission device 33 is one of the components of the human-powered vehicle 1 , its operation in response to an from the operating device 30th transmitted control signal is controlled. The transmission device 33 includes an electric actuator 33A and a chain guide 33B . An example of the electric actuator 33A is an electric motor. The transmission device 33 drives the electric actuator 33A to the chain guide 33B to operate, causing the front sprocket to put the chain 26th is guided, changed and the gear ratio of the human-powered vehicle 1 is switched.

An example of the transmission device 33 is an external gear. If it is the transmission device 33 is an external transmission, contains the front sprocket assembly 24A several front sprockets with different outside diameters. In the present embodiment, the transmission device operates 33 upon receipt of a control signal from the operating device 30th that instructs an upshift, the chain guide 33B by the electric actuator 33A so that the front sprocket to the present the chain 26th is performed, is switched to a sprocket with a larger outer diameter. Used by the control device 30th receiving a control signal instructing a downshift actuates the transmission device 33 the chain guide 33B by the electric actuator 33A so that the front sprocket to the present the chain 26th is performed, is changed to a front sprocket with a smaller outer diameter.

The transmission devices 32 and 33 can be internal gears. When the transmission device 32 for example, an internal gearbox is the transmission device 32 on the hub of the rear wheel 14th attaches and guides the gears to those in the rear sprocket assembly 24B entered rotation and transfers the rotation to the rear wheel 14th . Here is a single rear sprocket in the rear sprocket assembly 24B contain. At least one of the transmission devices 32 and 33 can be a continuously variable transmission.

The adjustable seat post 34 is on the frame 10A attached.

The adjustable seat post 34 contains an electric actuator 34A . The electric actuator 34A causes the saddle 18th relative to the frame 10A rises or falls. The electric actuator 34A is for example an electric motor. The adjustable seat post 34 is a part of the components created by adjusting a support position of the saddle 18th relative to the frame 10A can be controlled as operating parameters.

The suspension device 36 is a front suspension, for example on the front fork 10B is provided and for damping the on the front wheel 12 applied shock. The suspension device 36 contains an electric actuator 36A . The electric actuator 36A is for example an electric motor. The Suspension device 36 is one of the components that can be controlled by setting, for example, a damping factor, a repulsive force, a stroke amount, and a locking state as operating parameters. The suspension device 36 can be a rear suspension to absorb the impact on the rear wheel 14th be.

The battery unit 38 includes a battery 38A and a battery holder 38B . The battery 38A is a rechargeable battery with one or more battery cells. The battery holder 38B is on the frame 10A of the human-powered vehicle 1 attached. The battery 38A can be attached to the battery holder 38B connected to and removed from it. The battery 38A is electrical with the transmission devices 32 and 33 , the adjustable seat post 34 , the suspension device 36 etc. connected.

The human-powered vehicle 1 further comprises at least one speed sensor S1 , a cadence sensor S2 , a torque sensor S3 , a tilt sensor S4 , an accelerometer S5 , a position sensor S6 and a weather sensor S7 . If it is not necessary to describe the respective sensors individually in the following description, they are also called the sensors S1-S7 designated.

The speed sensor S1 is a sensor for outputting a signal indicating the speed of the human-powered vehicle 1 corresponds. The speed sensor S1 is on the frame 10A or on the front fork 10B attached. The speed sensor S1 contains, for example, a Hall element and detects one on the front wheel 12 or rear wheel 14th provided magnets (not shown) and thus measures a speed of the front wheel 12 or the rear wheel 14th . The speed sensor S1 outputs a signal that indicates the speed of the front wheel 12 or the rear wheel 14th indicates. In another example, the speed sensor calculates S1 the speed of the human-powered vehicle 1 based on the measured speed of the front wheel 12 or the rear wheel 14th and outputs a signal indicating the calculated speed.

The cadence sensor S2 is a sensor for outputting a signal that indicates a cadence. The cadence sensor S2 is on the frame 10A attached. The cadence sensor S2 contains, for example, a Hall element and detects a magnet (not shown) on the right crank 22A or on the left crank 22B is attached and measures the number of revolutions of the crankshaft 22C per time unit.

The torque sensor S3 is a sensor for outputting a signal to the crankshaft 22nd applied torque. The torque sensor S3 is on any drive path from the right crank 22A via the left crank 22B and the crankshaft 22C to the front sprocket assembly 24 or a drive path from the crankshaft 22C to the front sprocket assembly 24A intended. The torque sensor S3 contains a deformation sensor, a magnetostrictive sensor, a photo sensor, a pressure sensor or the like and detects a on the crankshaft 22nd applied torque.

The determining device 100 can affect the human-powered vehicle 100 transmitted power based on the cadence sensor S2 recorded number of revolutions per unit of time of the crankshaft 22C and that of the torque sensor S3 captured on the crank 22nd Calculate the applied torque.

The tilt sensor S4 is a sensor for outputting a signal indicative of an inclination angle of the human-powered vehicle 1 corresponds. The one from the tilt sensor S4 The detected angle of inclination is, for example, an angle of rotation about an axis of inclination in the right-left direction of the human-driven vehicle 1 . The tilt sensor S4 includes, as an example, a sensor for detecting the angular velocity of an angle of inclination and calculates, as the angle of inclination, a value obtained by integrating the angular velocity about the axis of inclination. The tilt sensor S4 may be configured to rotate an angle about a roll axis along the front-rear direction of the human-powered vehicle 1 and an angle of rotation about an inclination axis along the up-down direction of the human-powered vehicle 1 measures.

The acceleration sensor S5 is a sensor for outputting a signal indicative of the acceleration in the front-rear direction of the human-powered vehicle 1 corresponds. The acceleration sensor S5 is for example on the frame 10A attached and records the displacement of the armature as a change in capacitance or resistance when accelerating in the fore-aft direction and thus measures the acceleration. The acceleration sensor S5 can be configured to both accelerate in the right-left direction of the human-powered vehicle 1 as well as the acceleration in the up-down direction of the human-powered vehicle 1 measures. In addition, the acceleration sensor S5 be configured to also accelerate about a yaw axis, a roll axis, and a pitch axis of the human-powered vehicle 1 measures.

The position sensor S6 is a sensor for outputting a signal indicating the current position of the human-powered vehicle 1 indicates. The position sensor S6 is on the frame 10A or on the handlebar 16 attached. The position sensor S6 includes, for example, a GPS communication machine (Global Positioning System) and determines the current position of the human-powered vehicle 1 by receiving a radio wave from the GPS satellite. The position sensor S6 may be configured to have a memory for individually recording the measured position information and, based on the position information recorded in the memory, to output information about a travel path on which the human-powered vehicle is traveling 1 moves.

The weather sensor S7 is a measuring device for measuring the weather around the human-powered vehicle 1 around and to output the measurement result. The weather sensor S7 is on the frame 10A attaches and measures at least one of the following quantities: wind direction, wind speed, outside temperature and outside humidity. The weather sensor S7 can be configured to communicate with an external weather server (not shown) and at least one piece of information from the wind direction, the wind speed, the outside temperature and the outside air humidity around the human-powered vehicle 1 recorded by the weather server.

The driver of the human-powered vehicle 1 can with at least one heart rate monitor M1 , a thermometer M2 , a welding knife M3 , a blood pressure monitor M4 , a respiratory meter M5 , an electromyograph M6 , an electroencephalograph M7 and a body composition monitor M8 be equipped. If it is not necessary to describe the respective devices individually in the following description, they are also called the measuring devices M1-M8 designated. The measuring devices M1-M8 are in 2 shown.

The heart rate monitor M1 is a measuring device for measuring the driver's heart rate. The heart rate monitor M1 is attached to the driver's chest, for example, and measures an electrocardiogram of the driver. The heart rate monitor M1 detects the heartbeat based on changes in the electrocardiogram and calculates the heart rate from the heartbeat interval. Alternatively, the heart rate monitor M1 be attached to the driver's wrist to calculate the heart rate by detecting the driver's pulse waves.

The thermometer M2 is a measuring device for measuring the body temperature of the driver. The thermometer M2 is attached to the driver's ear, for example, and records the infrared rays emanating from the surface of the driver's body. The thermometer M2 calculates the driver's body temperature based on the intensity of the emitted infrared rays. It may be acceptable that the thermometer M2 is a sensor capable of measuring the driver's body temperature using, for example, a thermistor, the resistance of which changes when heated.

The welding knife M3 is a measuring device for measuring the amount of sweat of the driver. The welding knife M3 is attached to the driver's hand, for example, and contains a moisture sensor that detects the moisture generated by the driver's body surface. The welding knife M3 calculates the amount of sweat the driver has to sweat based on the humidity detected by the humidity sensor.

The sphygmomanometer M4 is a measuring device for measuring the driver's blood pressure. The sphygmomanometer M4 is attached to the driver's arm, for example, and outputs a measurement result that shows the driver's blood pressure.

The respiratory meter M5 is a measuring device for measuring the driver's breathing rate. The respiratory meter M5 is attached to the driver's mouth and outputs a measurement result indicating the flow rate of the driver's breathing.

The electromyograph M6 is a measuring device for measuring an activity level in a muscle of the driver. The electromyograph M6 contains an electrode that is attached to the driver's leg, for example, and outputs an electromyogram (EMG) measured by the electrode.

The electroencephalograph M7 is a measuring device for measuring the driver's brain waves. The electroencephalograph M7 is attached to the driver's head and measures an electrical phenomenon caused by the electrical activity of nerve cells in the brain with a method such as an electroencephalogram (EEG), magnetoencephalogram (MEG), or the like. Instead of measuring the electrical phenomenon, cerebral blood flow or changes in metabolism that occur as a secondary phenomenon after the ignition of nerve cells can be measured with a method such as functional magnetic resonance imaging (fMRI), near infrared spectroscopy (NIRS), or the like.

The body composition monitor M8 is a measuring device for measuring the amount of muscle and fat in the driver. The body composition monitor M8 contains, for example, an electrode, one of which is weak Electricity flows into the driver's vital structures and measures the impedance of the vital structures, thereby measuring the amount of muscle and fat in the driver. The body composition monitor M8 outputs data on the measured amount of muscle and the measured fat percentage.

The ones on the human-powered vehicle 1 mounted determination device 100 collects driving information based on the outputs from the sensors S1-S7 . The driving information includes at least one of the following information: speed of the human-powered vehicle 1 , Speed of rotation of the front wheel 12 or the rear wheel 14th , Cadence, power, angle of inclination, acceleration, geographical position, route, wind direction, wind speed, outside temperature and outside humidity.

The determining device 100 also captures driver information from the measurement results of the measurement devices M1-M8 . The driver information includes information about gender, age, height, weight, muscle amount, fat amount, heart rate, body temperature, sweating, blood pressure, breathing, myoelectric potential and brain waves of the driver. The driver's sex, age, height, weight, muscle and fat amount can be classified as the first driver information that varies relatively little during the journey. Heartbeat, body temperature, sweat, blood pressure, breathing, myoelectric potential and brain waves can be classified as second driver information that varies relatively widely while driving. Such information is entered into the determination device by the driver 100 entered and in the determination device 100 saved.

The determining device 100 is configured to provide driving information with the sensors S1-S7 and driver information with the measuring devices M1-M8 although it may be configured to acquire part of the driving information and the driver information with a portable terminal such as a driver-held smartphone or a driver-attached portable terminal. In this case, the determining device 100 acquire at least one of the driving information and the driver information from the portable terminal or the portable terminal of the driver via wired or wireless communication.

The determining device 100 determines the physical condition of the driver based on the driving information about driving the human-powered vehicle 1 . In the present embodiment, the determining device determines 100 the physical condition of the driver based on the driving information and the driver information of the driver of the human-powered vehicle 1 .

Here the determination device 100 determine the physical condition of the driver based on the driving information and the first driver information or determine the physical condition of the driver based on the driving information, the first driver information, and the second driver information. The physical state is a synthetic state that indicates how far the driver of the human-powered vehicle is 1 is still able to drive. In other words, the physical condition is the driver's remaining physical strength.

The following is the configuration of the determination device 100 described.

2 Fig. 13 is a block diagram showing an example of the internal configuration of the determination device 100 according to the embodiment 1 shows. The determining device 100 is of the computer type. In one example, the determining device is 100 a dedicated terminal like a bike computer that is on the human-powered vehicle 1 is mounted. In another example, the determining device is 100 a general purpose terminal such as a smartphone, tablet, etc. used by the driver of the human-powered vehicle 1 is held. The determining device 100 comprises an input unit 102 , an arithmetic processing unit 104 , a storage unit 106 and an output unit 108 .

The input unit 102 has a communication interface on which the sensors S1-S7 and the measuring devices M1-M8 are connected. The one in the input unit 102 The communication interface provided is, for example, a wired interface and is connected to the sensors via communication cables S1 - S7 and the measuring devices M1-M8 connected. The input unit 102 receives data from the sensors via the communication cables S1-S7 and the measuring devices M1-M8 output signals. The one from the input unit 102 input signals can be stored in the memory unit 106 stored on a time series basis. Parts of the sensors S1-S7 and measuring devices M1-M8 on the input unit 102 are to be connected can be connected via a wireless communication interface. For the wireless communication interface, a communication interface in accordance with a communication standard such as Bluetooth (registered trademark), WiFi (registered trademark), ZigBee (registered trademark), Long Term Evolution (LTE), and other wireless local area networks (LAN) can be used.

The arithmetic processing unit 104 is equipped with a central processing unit (CPU), a read-only memory (ROM), a random access memory (RAM), etc. In the one in the arithmetic processing unit 104 provided ROM become computer programs, etc. for controlling the operation of each of the in the determination device 100 hardware components provided. The CPU in the arithmetic processing unit 104 executes the computer programs stored in the ROM to control the operation of each of the hardware components, thereby performing the determination of a physical condition of the driver based on the driving information of the human-powered vehicle 1 is achieved. In the one in the arithmetic processing unit 104 provided RAM, the data used during the execution of a calculation are cached.

The arithmetic processing unit 104 is configured to include the CPU, ROM, and RAM, although it can be one or more arithmetic circuits associated with a graphics processing unit (GPU), a field programmable gate array (FPGA), a digital signal processor (DSP) , a quantization processor, volatile or non-volatile memory, etc.

The storage unit 106 is equipped with a memory such as an electronically erasable, programmable read-only memory (EEPROM), a static random access memory (SRAM), etc. In the storage unit 106 various computer programs are stored, including a determination execution program 110 that from the computing unit 104 should be carried out, a learning model 120 to be described later, etc.

The Determination Execution Program 110 is a computer program that causes the computer to perform to determine the physical condition of the driver based on the driving information of the human-powered vehicle 1 execute. In the present embodiment, the determination execution program is 110 a computer program that causes the computer to execute to determine the physical condition of the driver based on the driving information of the human-powered vehicle 1 and the driver information. The arithmetic processing unit 104 runs the determination execution program 110 to thereby determine the physical condition of the driver based on the driving information of the human-powered vehicle 1 and determine the driver information.

Various computer programs including the determination execution program 110 can through a storage medium M. be provided, which is able to store the computer programs. The storage medium M. is, for example, portable memory such as CD-ROM, USB memory, secure digital memory card (SD), micro SD card, CompactFlash (registered trademark), or the like. According to the current embodiment, the storage medium is M. a non-temporary storage medium that stores various computer programs including the determination execution program 110 saves legibly. The arithmetic processing unit 104 reads various computer programs from the storage medium with the aid of a reading device M. and installs the various computer programs that have been read out on the storage unit 106 .

That in the storage unit 106 saved learning model 120 is described by its definition information. The definition information of the learning model 120 include structural information of the learning model 120 and various parameters such as weights between the nodes, distortions, etc., which are used in the learning model 120 be used. In the present embodiment, the learning model is 120 in the storage unit 106 , which is learned according to a predetermined learning algorithm, stored, the driving and driver information of the human-powered vehicle 1 and data indicating the physical condition of the driver are used as teaching data. An algorithm for monitored deep learning using a neural network can be used as the learning algorithm. Instead of the supervised algorithm for deep learning, an algorithm for unsupervised learning or a repetitive neural network can be used.

The output unit 108 has an output interface via which a determination result of the arithmetic processing unit 104 is issued. The determination result can be output in any form.

In one example, the output unit comprises 108 a display device such as a liquid crystal screen, an organic electroluminescent display (EL), or the like. In this case the output unit 108 Output data indicating the physical condition of the driver through character or image information on the display device. In another example, the output unit contains 108 a voice output device such as a loudspeaker or the like. In this case the output unit 108 output a voice via the voice output device indicating the physical condition of the driver. In another example contains the output unit 108 a communication interface. In this case the output unit 108 Output data showing the physical condition of the driver to external devices that are connected to the communication interface.

3 Fig. 3 is a configuration diagram of the learning model 120 from the first example. The learning model 120 is a learning model for machine learning that includes, for example, deep learning and is built up via a neural network. The learning model 120 in the first application example includes an input level 122 , the intermediate levels 124A and 124B and an output level 126 .

In the example in 3 are the two intermediate levels 124A and 124B shown, wherein not only two, but three or more intermediate levels can be provided.

The input level 122 , the intermediate levels 124A and 124B and the output level 126 each contain one or more nodes, and the nodes in each of the levels are connected in one direction to the respective nodes in the previous level and the respective nodes in the subsequent level by predetermined weights and biases. Vector data showing the same number of components as the nodes in the input plane 122 have become the learning model 120 provided as input data. For example, data based on the output signals of the sensors S1-S7 based, and data showing the measurement results of the measuring devices M1-M8 show the node in the input level 122 made available. Alternatively, the output signals of the sensors S1-S7 and the measurement results of the measuring devices M1-M8 not provided as they are, but provided by the arithmetic processing unit 104 Data adapted to the input form can be assigned to the node in the input level 122 to provide.

In the input level 122 becomes driving information for driving the human-powered vehicle 1 entered. In the present embodiment, the input level 122 Driving information about driving the human-powered vehicle 1 and driver information about the driver of the human powered vehicle 1 entered. The data that the node in the input plane 122 are made available to the corresponding nodes in the first intermediate level 124A issued. On the intermediate level 124A the outputs are calculated with an activation function including weightings and biases and the calculated values are sent to the corresponding nodes in the following intermediate level 124B issued. This transfer is repeated one level after the other until outputs of the output level 126 can be obtained. Various parameters such as weightings, which connect the nodes, and preloads are taught in according to a predetermined learning algorithm. For the learning algorithm that learns various parameters, a learning algorithm for deep learning is used, for example. In the present embodiment, various parameters can be learned by using a predetermined learning algorithm that uses the driving and driver information of the human-powered vehicle 1 and data indicating the physical condition of the driver is used as learning data. The intermediate levels 124A and 124B learn about the relationship between driving information and the physical condition of the driver. In the present embodiment, the intermediate levels learn 124A and 124B the relationship between the driver and driver information and the physical condition of the driver.

The output level 126 outputs data indicating the driver's physical condition, that is, a determination result. The output level 126 can output the determination result in any form. The output level 126 consists for example of eleven nodes from the 0th to the 10th node. Data can be output from each of the nodes indicating the remaining physical strength of the driver. For example, the probability that the driver's remaining physical strength is 100% is output from the 0th node, the probability that the driver’s remaining physical strength is 90% is output from the first node, ... the probability that the remaining physical strength of the driver is 0% is output from the 10th node. The state in which the remaining physical strength is 100% represents a state in which the physical state of the driver is not significantly different from the state before the start of driving the human-powered vehicle 1 has changed. The number of nodes that make up the output level 126 and the output data to be assigned to each of these nodes are not limited to the example described above and can be designed accordingly.

4th FIG. 12 is a flow chart illustrating a determination method in Embodiment 1. FIG. The arithmetic processing unit 104 the determining device 100 captured in step S10 Driving and driver information of the human-powered vehicle 11 . More precisely, the arithmetic processing unit 104 records the output signals from the sensors S1-S7 and the measurement results of the measuring devices M1-M8 via the input unit 102 . The arithmetic processing unit 104 can read the output signals from the sensors S1-S7 and those of the Measuring devices M1-M8 Process recorded measurement results and generate input data that corresponds to the input level 122 of the learning model 120 to provide.

The arithmetic processing unit 104 enters the recorded driving and driver information in the input level 122 of the learning model 120 and thereby leads in step S102 Calculations using the learning model 120 out. At the node of the input level 122 data is provided with the driving information and the driver information. The data that represents the input level node 122 are made available to the appropriate nodes in the adjacent intermediate level 124A issued. On the intermediate level 124A calculations with an activation function including weightings between the nodes and biases are carried out and the calculation results are sent to the subsequent intermediate level 124B issued. On the intermediate level 124B the calculations with an activation function including the weightings between the nodes and bias voltages are carried out further and the calculation results are sent to the corresponding nodes in the output level 126 issued. With this, the determination result of the physical condition of the driver can be obtained from the learning model 120 be won. In other words, the learning model 120 is used to carry out the execution of calculations in the intermediate levels 124A and 124B based on the input level 122 input driving information and the output of data showing the physical condition of the driver from the output level 126 used. In the present embodiment, the learning model 120 for performing calculations in the intermediate levels 124A and 124B based on the driving information and the driver information entered in the input level 122 and for the output of data relating to the physical condition of the driver from the output level 126 ads, used.

The arithmetic processing unit 104 captured in step S103 the results of the calculation based on the learning model 120 from the output level 126 . From the nodes that are the output level 126 form, data can be obtained that indicate the physical condition of the driver. The data indicating the physical condition represent, for example, a probability that the remaining physical strength of the driver is X% (where, for example, X is 100, 90, ..., 0).

The arithmetic processing unit 104 determined in step S104 the physical condition of the driver based on the results of the calculations obtained from the output plane 126 of the learning model 120 were won. The arithmetic processing unit 104 determines, for example, that the state is most likely from those of the nodes in the output level 126 issued probabilities corresponds to the physical condition of the driver.

The determining device thus determines 100 the physical condition of the driver using the learning model 120 configured to output data showing the physical condition of the driver in accordance with an input of driving information of the human-powered vehicle 1 Show. The determining device 100 determines the physical condition of the driver with the help of the learning model 120 configured to output data indicating the physical condition of the driver in accordance with the inputs of driving information of the human-powered vehicle 1 and the driver information about the driver of the human powered vehicle 1 Show. The determining device 100 in Embodiment 1 determines the driver's physical condition using the learning model 120 based on the driving information of the human-powered vehicle 1 by the sensors S1-S7 and the driver information of the measuring devices M1-M8 can be obtained.

The present embodiment is configured to determine the physical condition of the driver using the learning model 120 although it can be configured to determine the physical condition of the driver using a function or table from which data indicative of the physical condition of the driver is outputted as inputs driving information of the human-powered vehicle 1 by the various sensors S1-S7 and driver information obtained from the various measuring devices M1-M8 can be obtained.

The present embodiment is configured so that a probability that the remaining physical strength of the driver is X% (where X is 100, 90, ..., 0) from the nodes in the output plane 126 of the learning model 120 although it can be configured to give a probability that the number of calories consumed by the driver corresponds to a certain value Y (where Y is a discrete value such as Y1, Y2, ..., Yn, or the like). The arithmetic processing unit 104 can calculate the number of calories consumed by the driver based on the output of the output level 126 and derive the physical condition by comparing the derived number of calories consumed with the value of the Determine the number of calories measured before the driver started.

As a further output form, it can be configured in such a way that it outputs a probability that a load on the driver has a certain value Z (Z is a discrete value such as Z1, Z2, ..., Zn or the like) from the nodes in the output level 126 of the learning model 120 corresponds. The arithmetic processing unit 104 can load the driver based on the output from the output level 126 and determine the physical condition of the driver based on the derived load.

Embodiment 2 provides a control system for controlling a component of the human-powered vehicle 1 based on a determination result of the determination device 100 described.

5 FIG. 13 is a block diagram illustrating a control system according to Embodiment 2. FIG. The control system according to Embodiment 2 includes a determination device 100 and a control device 200 which is a component of the human powered vehicle 1 based on a determination result of the determination device 100 controls.

The internal configuration of the determination device 100 is similar to that of the determining device 100 in Embodiment 1, and therefore a detailed description thereof is not given here.

The control device 200 is with a control unit 202 , a storage unit 204 , an input unit 206 and an output unit 208 fitted. The control unit 202 includes, for example, a CPU, a ROM, a RAM and so on. In the one in the control device 202 provided ROM are control programs, etc. for controlling the operation of those in the control device 200 intended hardware components stored. The CPU in the control unit 202 runs the control programs stored in ROM and various in the storage unit 204 stored programs to control the operation of the hardware components, whereby the control device according to the present application is achieved. In particular, the control unit generates 202 a control signal for controlling each of the components of the human powered vehicle 1 based on the determination result from the determination device 100 that is via the input unit 206 is input, and outputs the generated control signal via the output unit 208 out.

The configuration of the control unit 202 is not limited to the configuration described above. At the control unit 202 it can be one or more control circuitry including a single core CPU, a multicore CPU, an FPGA, volatile or non-volatile memory, and so on. In addition, the control unit 202 have functions of a clock for outputting date and time information, a timer for measuring an elapsed time from the issue of a timing start command to the issue of a timing end command, a counter for counting the number, and so on.

The storage unit 204 is equipped with a memory such as an EEPROM or an SRAM. In the storage unit 204 are those of the control unit 202 computer programs to be executed, the data to be used by the computer programs or the like.

The input unit 206 has an interface to which the determination device 100 connected with a cable. In the input unit 206 becomes one of the determining device 100 output determination result entered. The determination result that is in the input unit 206 should be entered is data that indicate a physical condition of the driver. The input unit 206 gives the inputted determination result to the control unit 202 out. In the present embodiment, the control device is 200 configured so that it is via the cable with the determination device 100 is connected, the control device 200 can be configured to send data to the destination device via a wireless communication interface 100 sends and from the destination device 100 receives. A wireless transmission system in accordance with a communication standard such as Bluetooth, WiFi, ZigBee, LTE, other wireless LANs or the like can be used for wireless communication.

The output unit 208 has an interface to which each of the components to be controlled is connected via a cable. The output unit 208 gives a from the control unit 202 generated control signal to a component to be controlled. In the present embodiment, the control device is 200 configured so that it is wired to each of the Components of the human powered vehicle 1 can be connected to the control device 200 can be configured to send data to any of the components of the human-powered vehicle via a wireless communication interface 1 sends and from each of the components of the human powered vehicle 1 receives. For the wireless communication, a wireless transmission system according to the various communication standards described above can be used.

To the to the control device 200 Connected components include, for example, a reporting device 220 . The reporting device 220 comprises a display device 220A or a speech output device 220B .

The display device 220A includes a liquid crystal display panel or the like for displaying information based on one from the output unit 208 output control signal. The display device 220A is at a suitable location on the human-powered vehicle 1 installed visible to the driver. The control device 200 generates character or image information to be notified to the driver based on the determination result of the determination device 100 and outputs the generated character or image information via the output unit 208 to the display device 220A out.

The speech output device 220B is with a loudspeaker or similar for outputting speech based on one of the output unit 208 output control signal. The speech output device 220B is at a suitable location on the human-powered vehicle 1 installed so that the driver can hear the output voice. The control device 200 generates voice information to be notified to the driver based on the determination result by the determination device 100 and outputs the generated speech information via the output unit 208 to the speech output device 220B out.

6th Fig. 13 is a flowchart showing a control process according to the embodiment 2 illustrated. If in step S201 a determination result by the determination device 100 is detected, the control unit generates 202 the control device 200 in step S202 a control signal for controlling the component to be controlled based on the determination result by the determining device 100 . For example, if the component to be controlled is the display device 220A acts, the control unit generates 202 a control signal to display character or image information on the display device 220A to display. If the component to be controlled is the speech output device 220B acts, the control unit generates 202 a control signal to the speech information via the speech output device 220B to spend.

The control unit 202 gives that in step S202 generated control signal via the output unit 208 to the component to be controlled and controls in step S203 the component to be controlled.

The display device 220A shows character or image information based on the control signal from the control device 200 on. 7th Fig. 13 is a schematic view showing an example of that on the display device 220A shows information to be displayed. The example in 7th Fig. 13 shows a state in which the character information indicating that the remaining physical strength of the driver decreases to 40% and the character information indicating that eating and / or drinking or a rest is necessary on the display device 220A is shown.

8th Fig. 13 is a schematic view showing another example of information displayed on the display device 220A should be displayed. The example in 8th Fig. 13 shows a state in which image information shows a route recommended to the driver on the display device 220A are displayed. If there are multiple routes from the current location of the human-powered vehicle 1 to the destination, the control unit calculates 202 the control device 200 the route for the respective routes and causes the display device 220A to display the route of the shortest route determined by the calculation as the recommended route. The destination can be determined by the driver himself. The control unit 202 can the display device 220A cause the route with the smallest height difference through the driveway to be displayed as the recommended route instead of recommending the shortest route. The control device 200 can use a known method route information from the current position of the human-powered vehicle 1 record to the destination. When the control device 200 can access an external map server, it uses an application programming interface (API) offered by the map server and provides the API with information about the current position and destination, thereby collecting route information.

Thus, in Embodiment 2, each of the components of the human-powered vehicle becomes 1 controlled in accordance with the determination result of the physical condition of the driver, so that the driver can be asked to eat and / or drink or to take a break or, if necessary, can be presented with a recommended route.

In Embodiment 2, that is by the control device 200 component to be controlled, although not limited to the Display device 220A or the speech output device 220B . The component to be controlled can be the transmission device 32 or 33 be those in human-powered vehicles 1 is provided. If the component to be controlled is the transmission device 32 acts, the control unit generates 202 the control device 200 a control signal for controlling the operation of the transmission device 32 based on the determination result of the determination device 100 and outputs the generated control signal via the output unit 208 to the transmission device 32 out. So can the control device 200 the transmission device 32 control in such a way that in cases where the physical condition of the driver deteriorates, the gear ratio or gear stage is changed and the physical burden on the driver is reduced. If the component to be controlled is the transmission device 33 acts, the control unit generates 202 the control device 200 a control signal for controlling the operation of the transmission device 33 based on the determination result of the determination device 100 and outputs the generated control signal via the output unit 208 to the transmission device 33 out. So can the control device 200 the transmission device 33 control in such a way that in cases where the physical condition of the driver deteriorates, the gear ratio and the gear stage are changed and the physical burden on the driver is reduced.

In Embodiment 2, the determining device 100 and the control device 200 described as separate devices, the determining device 100 and the control device 200 can be formed as an integrated device.

In the embodiment 3 describes a communication system in which communication between a destination device 100 and an external device.

9 FIG. 13 is a block diagram illustrating a communication system according to Embodiment 3. FIG. The communication system according to Embodiment 3 includes a determination device 100 and a communication device 300 which the determination device 100 transfers the output information to an external device. The internal configuration of the determination device 100 is similar to that of embodiment 1 , and therefore no detailed description thereof is given here.

The communication device 300 is with a control unit 302 , a storage unit 304 , an input unit 306 and a communication unit 308 fitted. The control unit 302 includes, for example, a CPU, a ROM, a RAM and so on. In the ROM of the control unit 302 become control programs, etc. for controlling the operation of the in the communication device 300 intended hardware components stored. The CPU in the control unit 302 runs the control programs stored in ROM and various in the storage unit 304 stored programs in order to control the operation of the hardware components and thus to reach the communication device according to the present application. In particular, the control unit transmits 302 a determination result received via the input unit 306 from the determining device 100 was entered to an external device. In one example, the external device is a communication device 300A attached to another human-powered vehicle 1A is mounted. 9 shows a single communication device 300A , being multiple communication devices 300A can be available for communication. In another example, the external device is a server device 400 that is connected to a communication network such as the Internet. The server device 400 will be described in embodiment 4 with reference to FIG 11 described.

The configuration of the control unit 302 is not limited to the configuration described above. At the control unit 302 it can be one or more control circuits, including a single-core CPU, a multi-core CPU, an FPGA, a volatile or non-volatile memory, etc. In addition, the control unit 302 Have functions of a clock for outputting date and time information, a timepiece for measuring an elapsed time from the issuing of a timing start command to the issuance of a timing end command, a counter for counting the number, and so on.

The storage unit 304 is equipped with a memory such as an EEPROM, an SRAM, etc. In the storage unit 304 are those of the control unit 302 computer programs to be executed, the data to be used by the computer programs, etc. are stored.

The input unit 306 has an interface to which the determination device 100 connected with a cable. In the input unit 306 becomes one of the determining device 100 output determination result entered. The determination result that is in the input unit 306 is data that indicate a physical condition of the driver. The input unit 306 gives the entered determination results to the control unit 302 out. In the present embodiment, the communication device is 300 configured so that it is via the cable with the determination device 100 can be connected to the communication device 300 can be configured to send data to the destination device via a wireless communication interface 100 sends and from the destination device 100 receives. A wireless transmission system according to Bluetooth, WiFi, ZigBee, LTE and other wireless LANs can be used for wireless communication.

The communication unit 308 has a communication interface.

A communication interface based on a communication standard such as WiFi, LTE and other wireless LANs can be used for the wireless communication interface. The communication device 300 transmits the information to be reported via the communication unit 308 to a communication device 300A attached to another human-powered vehicle 1A is mounted. The communication device 300 receives the information transmitted to it from the communication device 300A attached to another human-powered vehicle 1A is mounted, via the communication unit 308 .

10 FIG. 13 is a flowchart illustrating a control process in Embodiment 3. FIG. The control unit 302 the communication device 300 captured in step S301 a determination result obtained by the determination device 100 and generated in step S302 Information sent to the communication device 300A running on another human-powered vehicle 1A mounted, based on the determination result obtained from the determination device 100 was determined. The control unit 302 generates, for example, character or image information indicating the determination result. The from the control device 302 The character information generated may include character information indicating the remaining physical strength of the driver, or character information prompting the driver to eat or drink or to rest, as in FIG 7th is shown. The one from the control unit 302 The generated image information can contain image information about a route recommended to the driver, as in 8th is shown.

The control unit 302 transfers in step S303 the in step S302 generated information about the communication unit 308 to the communication device 300A another human-powered vehicle 1A .

The communication device 300 According to Embodiment 3, it is configured to receive the information on the determination result by the determination device 100 to the communication device 300 another human-powered vehicle 1A transmits, the communication device 300 may be configured to receive information about the physical condition of another driver provided by the determination device of another human-powered vehicle 1A determined and by the communication device 300A this human powered vehicle 1A be transmitted.

The communication device 300 does not have to be every time the determination device 100 receives a determination result, information on the determination result to the communication device 300A in another human-powered vehicle 1A and can only issue alarm information about the physical condition of the driver when the physical condition of the driver deteriorates.

The communication device 300 can receive driving environment information including information about at least one traffic light or traffic volume from the external device. More specifically, when the communication device 300 can communicate with an external server that transmits driving environment information about at least one of the traffic lights and traffic volume, it can communicate with the external server via the communication unit 308 received transmitted driving environment information.

When the communication device 300 to the display device described in Embodiment 2 220A connected, the control unit 302 can be configured to provide information about a recommended route to the driver based on the information provided via the communication unit 308 generated driving environment information received and the generated information on the display device 220A indicates.

Therefore, in Embodiment 3, the driver of another human-powered vehicle can 1A Be informed of the physical condition of a particular driver by the human-powered vehicle 1 attached determination device 100 is determined. If several drivers are driving in a group and one of the drivers is in poor physical condition, the driver can be asked to eat and / or drink something or to rest, or he can reduce the driving speed in order to reduce the physical strain on the driver.

In Embodiment 3, the determining device 100 and the communication device 300 described as separate devices, the determination device 100 and the communication device 300 can be formed as an integrated device.

The communication system in Embodiment 3 is configured to be the determination device 100 and the communication device 300 although it can be configured to additionally include the control device described in Embodiment 2.

In Embodiment 4, a method of generating the learning model is described 120 that is in the determination device 100 is used.

That in the determination device 100 learning model used 120 is e.g. from an external server device 400 generated. 11 Fig. 13 is a block diagram showing an example of the internal configuration of the server device 400 shows. The server device 400 is with a control unit 402 , a storage unit 404 , an input unit 406 and a communication unit 408 fitted.

The control unit 402 includes, for example, a CPU, a ROM, a RAM and so on. In the one in the control unit 402 provided ROM become control programs, etc. for controlling the operation of the in the server device 400 hardware components provided. The CPU in the control unit 402 runs the control programs stored in ROM and various in the storage unit 404 saved programs to control the operation of the hardware components.

The configuration of the control unit 402 is not limited to the configuration described above. The control unit 402 is not limited to having the CPU, the ROM and the RAM. At the control unit 402 it can be one or more control or arithmetic circuits including a GPU, an FPGA, a DSP, a quantum processor, a volatile or non-volatile memory, etc. In addition, the control unit 402 have functions of a clock for outputting date and time information, a timer for measuring an elapsed time from the issue of a timing start command to the issue of a timing end command, a counter for counting the number, and so on.

The storage unit 404 is equipped with a storage device such as a hard disk or the like. In the storage unit 404 are different computer programs that are run by the control unit 402 to be executed, one from the server device 400 generated learning model 410 etc. saved.

The input unit 406 has an input interface through which data and a program can be obtained from a recording medium that records various data or programs. The various via the input unit 406 captured data and programs are stored in the storage unit 404 saved.

The communication unit 408 has a communication interface.

The communication unit 408 receives data from the communication device 300 to the server device 400 are transmitted while simultaneously sending to the communication device 300 transmits data to be transmitted.

The following is a method for generating a learning model 410 in the server device 400 described.

12 Figure 13 is a flow diagram illustrating a method for generating the learning model 410 illustrated. The control unit 402 the server device 400 recorded via the input unit 406 and the communication unit 408 in step S401 the teach-in data for teaching in the learning model 410 . The learning data includes, for example, driving information about driving the human-powered vehicle 1 and driver information about the driver of the human powered vehicle 1 as well as label data that indicate the physical condition of the driver. At an early stage in the creation of the learning model 410 can be the learning data from the administrator or the like of the server device 400 can be set. In the course of the learning, the learning data can be based on the result of conclusions by the learning model 410 be determined. In the latter case, the input for the learning model 410 and the label data obtained as a result of conclusions are set as learning data.

The control device 402 there in step S402 the driving information and the driver information that are entered as learning data in the learning model 410 be included. In a phase before the start of teaching, input settings for the definition information that the learning model 410 describe given. The learning model 410 includes an input level, intermediate levels, and an output level similar to that in FIG Embodiment 1 illustrated learning model 120 and that in illustrated learning model 120 . The nodes resulting from the input level of the learning model 410 exist, driving and driver information is provided. The information that is made available to the nodes in the input level is output to the corresponding nodes in the adjacent intermediate level. In the intermediate level, calculations with an activation function including weightings between the nodes and biases are carried out and the calculation results in the intermediate level are output to the corresponding nodes in the output level. The control unit 402 relates in step S403 the calculation results by the learning model 410 of the nodes in the output level.

The control unit 402 evaluates in step S404 those in the crotch S403 obtained calculation results and sets in step S405 determines whether the teaching of the learning model 410 completed or not. More precisely, the control unit 402 evaluate the calculation results with the help of an error function based on the in step S403 obtained calculation results and the learning data. The error function can also be referred to as the objective function, loss function or cost function. The control unit 402 determines that the teaching of the learning model 410 is completed when the error function is equal to or less than a threshold value or equal to or greater than a threshold value in the process of optimizing the error function by stochastic gradient descent, such as a method of the steepest descent. The control unit 402 can end the teach-in at an appropriate time by using a process such as cross-validation, early stop, or the like to avoid over-learning.

If in step S405 If it is determined that the learning is not completed, the control unit updates 402 the weights between the nodes and the pre-stresses in the intermediate levels in step S406 and returns to step with processing S401 back. The control unit 402 can update the weightings between the nodes and the biases using a backpropagation method in order to successively update the weightings between the nodes and the biases from the output level to the input level.

If in step S405 If it is determined that the teaching-in has been completed, the control unit saves 402 in step S407 the learning model 410 as a taught-in learning model in the storage unit 404 and ends the processing according to this flowchart.

Thus, in the method for generating the in the server device 400 executed learning model, the learning model used for performing the determination of the driver's physical condition from the driving information based on the driving information about driving the human-powered vehicle 1 to be generated. The server device 400 in embodiment 4, the learning model 410 generate that for performing the determination of the driver's physical condition from the driving information and the driver information based on the driving information and the driver information of the human-powered vehicle 1 and the data indicating the physical condition of the driver is used. The determining device 100 loads the taught-in learning model 410 from the server device 400 and saves the learning model in the storage unit 106 whereby the physical condition of the driver with the help of the learning model 410 is determined.

In embodiment 5 becomes the configuration for determining the physical condition of the driver in the server device 400 described.

If the learning model 410 that is in the storage unit 404 the server device 400 has already been learned in, the physical condition of the driver can be determined with the help of this learning model 410 be determined. In embodiment 5, there is a communication device 300 on the human-powered vehicle 1 assembled. The determining device 100 does not have to be on a human-powered vehicle 1 be mounted. The internal configuration of the communication device 300 is similar to that of the communication device in Embodiment 3.

The communication device 300 in embodiment 5 detected via the input unit 306 Driving or driver information from the sensors S1-S7 and the measuring devices M1-M8 . The communication device 300 transmits the from the sensors S1-S7 and the measuring devices M1-M8 captured driving information and driver information via the communication unit 308 to the server device 400 . The server device 400 receives the driving information and the driver information received from the communication device 300 and determines the physical condition of the driver based on the received driving information and driver information. That is, in Embodiment 5 is the server device 400 configured to function as the determination device of the present application.

13 FIG. 13 is a flowchart showing a determination method in Embodiment 5. FIG illustrated. The communication device 300 captured in step S501 Driving and driver information of the human-powered vehicle 1 . More specifically, the communication device detects 300 via the input unit 306 the output signals of the sensors S1-S7 and the measurement results of the measuring devices M1-M8 .

The communication device 300 transfers in step S502 the driving information and the driver information provided in step S501 are detected to the server device 400 .

The server device 400 receives in step S503 the driving information and the driver information received from the communication device 300 be transmitted. In step S504 gives the control unit 402 the server device 400 those in the crotch S503 received driving information and the driver information in the input level of the learning model 410 and thus performs calculations with the learning model 410 out. The nodes in the input level are provided with data which specify the driving information and the driver information. The data made available to the nodes in the input level are output to the corresponding nodes in the adjacent intermediate level.

In the intermediate level, calculations with an activation function including the weights between the nodes and the biases are performed, and the calculation results in the intermediate level are output to the following output level. The output level of the learning model 410 Outputs data showing the physical condition of the driver.

The control unit 402 captured in step S505 the calculation results of the learning model 410 from the output level of the learning model 410 . From the nodes, which form the output level, data can be obtained that indicate the physical condition of the driver. The data showing the physical condition are, for example, the probability that the remaining physical strength of the driver is X% (X is e.g. 100 , 90 , ..., 0). The learning model 410 may output the probability that the number of consumed calories corresponds to a certain value, a probability that the driver's load becomes a certain value, or the like, instead of the probability of the remaining physical strength.

The control device 402 determined in step S506 the physical condition of the driver based on the calculation results obtained from the output level of the learning model 410 were won. The control unit 402 determines, for example, from the probabilities output by the respective nodes in the output level that the state with the highest probability corresponds to the physical state of the driver.

The control unit 402 transfers in step S507 the determination result of the physical condition via the communication unit 408 to the communication device 300 .

The control unit 302 the communication device 300 receives in step S508 the determination result of the physical condition obtained by the server device 400 via the communication unit 308 is transmitted.

Therefore, in Embodiment 5, even in the case that the human-powered vehicle 1 not with the determination device 100 is equipped, the determination result of the physical condition of the driver via the server device 400 can be obtained.

In Embodiment 6, the second application example of the learning model will be described.

14th Fig. 13 is a configuration diagram of a learning model 140 in the second example. The learning model 140 The second example is a sequence-to-sequence model (seq2seq model), which is a type of recurrent neural network. The learning model 140 is e.g. from the server device 400 generated and in the storage unit 106 the determining device 100 saved.

The learning model 140 includes m pieces of encoders E1 to Em in each of which time series data is input, and n pieces of decoders D1 to Dn, from which time series data are output. The indices m and n can assume an integer greater than 1. As in 14th each is the encoder E1 to Em and the decoder D1 through Dn are shown as a single block, although it contains a plurality of about two to eight levels including the input level and the hidden levels. The internal configuration of the encoder E1 to Em and the decoder D1 to Dn as well as the teach-in procedure for the internal parameters in the encoders E1 to Em and the decoders D1 to Dn are already known, so that a detailed description is dispensed with here. In the present embodiment, the internal parameters of the learning model 140 learned so that time series data is one of the driving information and the driver information of the human-powered vehicle 1 as input for the learning model 140 and time series data showing the physical condition of the driver are output.

In 14th shows the horizontal direction one time step and shows the progress of the process from left to right in the drawing. In each of the encoders E1 to Em become time series data of one of the driving information and the driver information of the human-powered vehicle 1 entered. For example, if there is an output from the cadence sensor S2 as input for the learning model 140 is used, gives the arithmetic processing unit 104 the determining device 100 successive values of the cadence K1 , K2 , ... Km received as time series data into the encoder E1 to em a.

In the hidden level of the encoder E1 to Em, the input time series data of the cadence are recorded as an internal vector ct for the internal state. Here t represents a time step and takes place in the respective encoder E1 to Em values from 1 to m. The internal vector ct is transferred to a level with the corresponding value for each input, and the internal vector cm is transferred to the decoder D1 is to be passed is received after all entries are completed.

The internal vector cm in the last encoder Em is sent to the decoder D1 to hand over. A reserved word to instruct the beginning of an output is written into the decoder D1 entered. In the example in 14th If <go> is set as reserved word, any word is acceptable if it is a fixed value set in advance. In the event that the internal vector cm is passed by the encoder Em and the reserved word is entered to instruct the start of an output, the decoder gives D1 outputs an output h1 to change the internal vector to cm + 1. The output h1 of the decoder D1 is used as input for the decoder in the next step D2 used. The internal vector cm + 1 for the decoder D1 is used in the next step as the internal state for the decoder D2 used. Thus the output ht and the internal vector ct for the decoders D1 , D2 ... Dn-1 one after the other in the next decoder D2 , D3 , ..., Dn is input to perform the calculation sequentially until the last decoder Dn outputs a reserved word <eos> which represents the end of the output.

Therefore, n-1 pieces of the outputs h1 to hn-1 correspond to the respective decoders D1 to Dn-1 the final editions of the learning model 140 . The outputs h1 to hn-1 represent time series data indicating the physical condition of the driver.

The arithmetic processing unit 104 captures the time series data indicating the physical condition of the driver from the decoders D1 to Dn-1 of the learning model 140 . The arithmetic processing unit 104 sequentially repeats the entry of time series data into the encoder E1 to Em and the acquisition of time series data from the decoders D1 to Dn-1 are output, thereby determining the time series variation of the physical condition of the driver.

In embodiment 7th the third application example of the learning model is described.

15th Fig. 13 is a configuration diagram of a learning model 160 in the third example. The learning model 160 in the third application example is a learning model based on a convolutionary neural network (CNN), a region-based CNN (R-CNN) or the like and includes an input level 162 , an intermediate level 164 and an output level 166 . The learning model 160 is learned in such a way that it outputs data that shows the physical condition of the driver according to an input image of the graph, the time series variation of the driving information or the driver information of the human-powered vehicle 1 represents, show. The learning model 160 is e.g. from the server device 400 generated and in the storage unit 106 the determining device 100 saved.

When collecting the driving information and the driver information of the human-powered vehicle 1 via the input unit 102 generates the arithmetic processing unit 104 the determining device 100 Graphs showing their time series and outputs the generated graph images.

In the input level 162 the image of the graphs showing the time series variations of the driving information and driver information of the human-powered vehicle is inputted 1 Show.

For example, a time series diagram is entered in which the time as the horizontal axis and the values of the cadence of the human-powered vehicle 1 and the driver's heartbeat are shown as a vertical axis. The diagram may contain time series variations of several driving information and driver information.

The intermediate level 164 consists for example of a folding plane 164a , a pooling level 164b and a fully connected plane 164c . On the intermediate level 164 can alternate the folding levels 164a and the pooling levels 164b be provided. The folding level 164a and the pooling level 164b extract features from the via the input level 162 entered graph image. Among the features brought about by the folding plane 164a and the pooling level 164b are extracted include, for example, a slope of the graph, a maximum point, a minimum point or the like. The one from the folding plane 164a and the pooling level 164b extracted features are beyond the fully connected layer 164c to the output level 166 issued.

The output level 166 outputs data showing the physical condition of the driver based on that of the fully connected plane 164c Display the characteristics entered.

The arithmetic processing unit 104 captures the data showing the physical condition of the driver from the output level 166 of the learning model 160 . The arithmetic processing unit 104 may determine the physical condition of the driver based on the collected data indicative of the physical condition of the driver.

It is to be understood that the embodiments disclosed herein are in all respects illustrative and not restrictive. The scope of the present invention is defined by the appended claims, and all changes that come within the meanings and limits of the claims, or the equivalency of such meanings and limits, are intended to be embraced by the claims.

List of reference symbols

1

human powered vehicle

1A

human powered vehicle

10

Vehicle main body

10A

frame

10B

Front fork

12

Front wheel

14th

Rear wheel

16

Handlebar

16A

Handle

18th

saddle

20th

Drive mechanism

22nd

crank

22A

right crank

22B

left crank

22C

crankshaft

24A

front sprocket assembly

24B

rear sprocket assembly

26th

Chain

28

pedal

30th

Control device

30A

Control unit

32

Transmission device

32A

electric actuator

33

Transmission device

33A

electric actuator

33B

Chain guide

34

adjustable seat post

34A

electric actuator

36

Suspension device

36A

electric actuator

38

Battery unit

38A

battery

38B

Battery holder

100

Determining device

102

Input unit

104

arithmetic processing unit

106

Storage unit

108

Output unit

110

Determination implementation program

120, 140, 160

Learning model

122

Input level

124A

Intermediate level

124B

Intermediate level

126

Output level

160

Learning model

162

Input level

164

Intermediate level

164a

Folding plane

164b

Pooling level

164c

fully connected level

166

Output level

200

Control device

202

Control unit

204

Storage unit

206

Input unit

208

Output unit

220

Reporting device

220A

Display device

220B

Speech output device

300

Communication device

300A

Communication device

302

Control unit

304

Storage unit

306

Input unit

308

Communication unit

400

Server device

402

Control unit

404

Control unit

406

Input unit

408

Communication unit

410

Learning model

E1-Em

Encoder

D1-Dn

decoder

S1

Speed sensor

S2

Cadence sensor

S3

Torque sensor

S4

Inclination sensor

S5

Accelerometer

S6

Position sensor

S7

Weather sensor

M.

Storage medium

M1

Heart rate monitor

M2

thermometer

M3

Welding knife

M4

Sphygmomanometer

M5

Respiratory meter

M6

Electromyograph

M7

electroencephalograph

M8

Body composition monitor

S101-S508

Procedural steps

QUOTES INCLUDED IN THE DESCRIPTION

This list of the documents listed by the applicant was generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Patent literature cited

• JP 2015131533 A [0002]

Claims (20)

A determining device (100) that determines the physical condition of a driver based on driving information about driving a human-powered vehicle (1). The determination device (100) according to Claim 1 wherein the physical condition is determined by using a learning model (160) configured to output data indicating the physical condition corresponding to an input of the driving information. The determination device (100) according to Claim 1 or 2 , wherein the driving information contains information about at least one of the following variables: speed, acceleration, a geographical position, an angle of inclination, a driving distance, a driving time, a cadence, a power, a rotational speed of a wheel (12, 14), a route, a Wind direction, a wind speed, an outside temperature and an outside humidity of the human-powered vehicle (1). The determination device (100) according to one of the Claims 1 to 3 wherein the physical condition of the driver is determined based on the driving information and driver information about the driver of the human-powered vehicle (1). The determination device (100) according to Claim 4 wherein the physical condition is determined by using a learning model (160) configured to output data indicating the physical condition according to inputs of the driving information and the driver information. The determination device (100) according to Claim 4 or 5 wherein the driver information includes information about at least one of the following characteristics: gender, age, height, weight, amount of muscle, amount of fat, heart rate, body temperature, sweat, blood pressure, breathing, myoelectric potential and brain waves of the driver. A control system comprising: the determining device (100) according to any one of Claims 1 to 6th ; and a control device (200) that controls a component of the human-powered vehicle (1) based on the information output from the determination device (100). The control system according to Claim 7 , wherein the component includes a reporting device. The control system according to Claim 8 wherein the control device (200) controls the reporting device to report information about a recommended route. The control system according to Claim 8 or 9 wherein the control device (200) controls the reporting device to report information on at least one of eating, drinking and a rest. The control system according to one of the Claims 7 to 10 wherein the component includes a transfer device (32,33). A communication system comprising: the determining device (100) according to any one of Claims 1 to 6th ; and a communication device (300) that communicates with an external device, the communication device (300) transmitting the information output from the determination device (100) to the external device. The communication system according to Claim 12 wherein the information output by the determining device (100) includes warning information about the physical condition. The communication system according to Claim 12 or 13 wherein the communication device (300) receives information about a physical condition of another driver determined by a determination device (100) of another human-powered vehicle (1) and transmitted from a communication device (300) of the other human-powered vehicle (1) . The communication system according to one of the Claims 12 to 14th wherein the communication device (300) receives driving environment information including information on at least one traffic light and a traffic volume from the external device. The communication system according to Claim 15 , further comprise a reporting device that transmits information about a recommended route based on the information about the driving environment. A learning model (160) comprising: an input plane (122) into which driving information about driving a human-powered vehicle (1) is input; an intermediate level (124A, 124B) that learns a relationship between the driving information and a driver's physical condition; and an output level (126) via which data indicating the physical condition are output, the learning model (160) being used to process the calculation in the intermediate level on the basis of the information entered in the input level, and outputting data from the output level showing the physical condition. A method of generating a learning model (160) include: Generating the learning model (160) which is used in the processes of determining the physical condition of a driver from driving information about driving a human-powered vehicle (1) based on driving information and from data indicating the physical condition of the driver by means of a computer. A computer program that causes a computer to execute the processes of determining the physical condition of a driver based on driving information about driving a human-powered vehicle (1). A storage medium (M) on which the computer program is after Claim 19 is stored.

Images