JP2011036375A - Training device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a training device capable of applying to the agent a force of a variation behavior. <P>SOLUTION: According to the training device 1, according to the mode selection, it is possible to apply a force to the agent along the direction of the relative motion between an upper body and a leg (assist force) and a force to the agent reverse to the direction of the relative motion between the upper body and the leg (resistance force). <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a training device configured to transmit a force generated by an operation of an actuator to an agent's body through a brace.

Techniques have been proposed in which a force is applied to an agent (human) to assist the movement of its legs and the like (see Patent Documents 1 and 2).
JP 2007-054086 A JP 2007-259175 A

  However, since the action mode of force is constant, there is a possibility that the agent may feel uncomfortable.

  Therefore, an object of the present invention is to provide a training apparatus that can apply force to an agent in various forms.

  A training apparatus according to the present invention for solving the above problems includes a first mounting element and a second mounting element mounted on each of an upper body and a leg body of an agent, an actuator, and a control device that controls the operation of the actuator. A training device configured to transmit the force generated by the operation of the actuator to the agent through the first mounting element and the second mounting element, the upper body of the agent and An assist mode for applying an assisting force to the relative motion of the leg and a training mode for applying a resistance force to the relative motion of the upper body and the leg to the agent are used separately (first invention) ).

  According to the training apparatus of the present invention, the state in which a forward force (auxiliary force) acts on the agent relative to the direction of the relative motion of the upper body and the leg, and the upper body A situation in which a force (resistance force) in the opposite direction acts on the direction of relative motion of the legs is realized. Thereby, when the agent is performing a walking motion accompanied by the relative motion of the upper body and the leg, a force can be applied to the agent in various forms.

  In the training device according to the first aspect of the invention, the control device uses the auxiliary force and the function as a function having a part or all of a posture variable representing a relative posture of the leg to the upper body of the agent and a time derivative thereof as a variable. It may be configured to adjust at least one of the resistance forces (second invention).

  According to the training device of the configuration, the strength and direction of the force acting on the agent are determined according to at least one of the relative posture of the upper body and legs of the agent and the time-series variation of the relative posture. Can be changed.

  In the training device according to the second aspect of the invention, the control device is configured such that the posture variable is equal to or greater than a first assist reference value, and a posture change speed that is a first-order time derivative of the posture variable is negative, and the posture variable is In the first assist definition area in which the posture variable acceleration approaches 0 as the maximum value is approached, the approach variable approaches 0 as the posture variable approaches the maximum value, and becomes the minimum value as the posture variable approaches the first assist reference value. The auxiliary force is controlled according to a function having a changing characteristic that approaches, the posture variable is less than the first assist reference value, and the posture changing speed is negative and the posture variable approaches a minimum value, In the second assist definition area in which the attitude change speed approaches 0, the attitude variable approaches 0 as the attitude variable approaches a minimum value, while the attitude variable approaches the first assist reference value. The assist force is controlled according to a function having a change characteristic that approaches a minimum value, the posture variable is less than a second assist reference value, the posture change speed is positive, and the posture variable is a minimum value. In the third assist definition area in which the posture change speed approaches 0 as the value approaches, the change approaches 0 as the posture variable approaches the minimum value, while the approach variable approaches the maximum value as the posture variable approaches the second assist reference value. The posture change is controlled as the assist force is controlled according to a function having characteristics, the posture variable is greater than or equal to the second assist reference value, the posture change speed is positive, and the posture variable approaches a maximum value. In the fourth assist definition area in which the speed approaches 0, the attitude variable approaches 0 as the attitude variable approaches the maximum value, while the attitude variable approaches the second assist reference value. Which may be configured to control the assist force according to a function having the change characteristics approaching a maximum value (third invention).

  According to the training apparatus having the above configuration, the agent is walking and the measured values of the posture variable and the posture change speed are the first assist definition region, the second assist definition region, the third assist definition region, and the fourth assist definition region. When transitioning cyclically in the order of the assist domain, the operation is controlled as follows. A period in which the measurement value is included in the i-th assist definition area (i = 1 to 4) is defined as the i-th assist period.

  In the “first assist period”, the assist force is controlled to continuously decrease from 0 to its minimum value. In the “second assist period”, the assist force is controlled so as to continuously increase from its minimum value to zero. In the “third assist period”, the assist force is controlled so as to continuously increase from 0 to its maximum value. In the “fourth assist period”, the assist force is controlled to continuously decrease from the maximum value F to zero.

  As a result, the force assisting the relative movement of the leg relative to the upper body during the walking movement of the agent is changed in an appropriate change form in view of the movement state of the agent represented by the posture variable and the measured value of the posture change angle. Given to.

  In the training device according to a third aspect of the present invention, the control device has an approximate expression of a curve representing a change mode of the assist force, a posture variable and the posture change speed on a plane defined by the assist force and the posture variable. The auxiliary force may be controlled using the measured value (fourth invention).

  According to the training apparatus of the said structure, in the state where each of posture variables and posture change speeds, such as a walk exercise | movement, is changing continuously, it can adjust so that the auxiliary | assistant force given to an agent may change continuously.

  In any one of the second to fourth inventions, the control device has an attitude change rate, the attitude variable being equal to or higher than a first training reference value, and a first-order time derivative of the attitude variable. In the first training definition area, which is negative and the posture change speed approaches 0 as the posture variable approaches the maximum value, the posture variable approaches 0 as the posture variable approaches the maximum value. The resistance is controlled in accordance with a function having a change characteristic that approaches a maximum value as it approaches a reference value, the posture variable is less than the first training reference value, and the posture change speed is negative and the posture In the second training definition area in which the posture change speed approaches 0 as the variable approaches the minimum value, the approach variable approaches 0 as the posture variable approaches the minimum value. , Controlling the resistance force according to a function having a change characteristic that approaches a maximum value as the posture variable approaches the first training reference value, the posture variable is less than a second training reference value, and the posture change In the third training definition area in which the speed is positive and the posture variable speed approaches zero as the posture variable approaches the minimum value, the posture variable approaches zero while the posture variable approaches the minimum value. 2 The resistance is controlled according to a function having a change characteristic that approaches a minimum value as it approaches the training reference value, the posture variable is equal to or greater than the second training reference value, and the posture change speed is positive. In the fourth training domain in which the posture change speed approaches 0 as the posture variable approaches the maximum value, the posture variable approaches the maximum value. More while approaching 0, which may be configured to control the resistance force in accordance with the function having the change characteristics in which the attitude variable approaches as minimum value approaches the second training reference value (fifth invention).

  According to the training apparatus having the configuration, the agent is walking and the measured values of the posture variable and the posture change speed are the first training definition region, the second training definition region, the third training definition region, and the fourth. In the case of a cyclic transition in the order of the training domain, the operation is controlled as follows. A period in which the measurement value is included in the i-th training domain (i = 1 to 4) is defined as the i-th training period.

  In the “first training period”, the resistance is controlled so as to continuously increase from 0 to its maximum value. In the “second training period”, the resistance is controlled so as to continuously decrease from its maximum value to zero. In the “third training period”, the resistance is controlled to continuously decrease from 0 to its minimum value. In the “fourth training period”, the resistance is controlled so as to continuously increase from the minimum value to zero.

  As a result, the force that resists the relative movement of the leg with respect to the upper body during the walking movement of the agent is changed in an appropriate change form in consideration of the movement state of the agent represented by the posture variable and the measured value of the posture change angle. Given to.

  In the training device according to a fifth aspect of the present invention, the control device has an approximate expression of a curve representing a change mode of the resistance force, a posture variable and the posture change speed on a plane defined by the resistance force and the posture variable. The resistance may be controlled using the measured value (the sixth invention).

  According to the training apparatus having the configuration, the resistance force applied to the agent can be adjusted so as to continuously change in a state in which each of the posture variable and the posture change speed, such as walking motion, is continuously changing.

  In the training device according to any one of the first to sixth inventions, the control device is configured to control a force acting on the agent by an operation of the actuator with respect to a posture variable representing a relative posture of the leg with respect to the upper body of the agent. By controlling the phase difference, the force may be applied to the agent as one of the auxiliary force and the resistance force (seventh invention).

  In the training device according to any one of the first to seventh inventions, the control device determines whether each of the agent's legs is in a standing state and a free leg state, and the auxiliary device is determined according to the determination result. It may be configured to control the operation of the actuator so that the change forms of the force and the resistance force are different (the eighth invention).

  According to the training device of the configuration, the upper body of the agent and the body of the agent according to the standing state of the leg (the state where the leg is landing) and the free leg state (the state where the leg is leaving the floor) Each change form of the assisting force and the resistance force acting on the relative motion of the legs can be made different.

  In the training device according to an eighth aspect of the present invention, when the control device determines that the leg of the agent is in a standing state, compared to a case where it is determined that the leg of the agent is in a free leg state, The operation of the actuator may be controlled so that each of the auxiliary force and the resistance force becomes strong (9th invention).

  According to the training device having the above configuration, the strengths of the assisting force and the resistance force acting on the relative motion of the upper body of the agent and the leg are different depending on whether the leg is standing or swinging. Can be made. The strength of force here means strength of strength in a situation where factors other than the standing state and the free leg state are the same.

  In any one of the first to ninth inventions, the control device recognizes the posture of the walking surface of the agent, and each of the auxiliary force and the resistance force according to a difference in the recognition result. You may be comprised so that operation | movement of the said actuator may be controlled so that a change form may differ (10th invention).

  In the training device according to a tenth aspect of the present invention, when the control device recognizes that the posture of the walking surface is a rising inclination posture, it compares with a case where the posture of the walking surface is recognized as a flat posture or a downward inclination posture. And controlling the operation of the actuator so that the auxiliary force is strong, and recognizing that the posture of the walking surface is a flat posture, and recognizing that the posture of the walking surface is a downward inclined posture; In comparison, the operation of the actuator may be controlled so that the auxiliary force becomes stronger (the eleventh invention).

  In the training device according to a tenth aspect of the present invention, when the control device recognizes that the posture of the walking surface is a downward inclined posture, it compares with a case where the posture of the walking surface is recognized as a flat posture or an upward inclined posture. And controlling the operation of the actuator so as to increase the resistance, and recognizing that the posture of the walking surface is a flat posture, and recognizing that the posture of the walking surface is a rising inclination posture; In comparison, the operation of the actuator may be controlled so as to increase the resistance (a twelfth aspect).

  According to the training apparatus of the said structure, in light of the difference in the strength of the relative motion of the upper body and the leg required for walking according to the posture of the ground or floor (walking surface) on which the agent walks, Can apply an appropriately adjusted resistance force to the agent.

  In any one of the first to twelfth inventions, the control device is configured so that the assist force continuously changes from 0 to 0 after the assist mode starts and ends. The operation of the actuator may be controlled, and the operation of the actuator may be controlled to return to 0 after the resistance force continuously changes from 0 from the start time to the end time of the training mode. Good (13th invention).

  The training device according to any one of the first to thirteenth aspects, further comprising: a first interface that enables designation by one of the assist mode and the training mode, wherein the control device includes the first interface. The actuator may be configured to control the operation of the actuator in accordance with one of the assist mode and the training mode based on a signal corresponding to the designation output from (Aspect 14).

  According to the training apparatus having the configuration, the operation mode is switched between the assist mode and the training mode according to the intention of the agent. As a result, when the agent is performing a walking motion involving the relative motion of the upper body and legs, it is possible to apply force to the agent in various forms that reflect the intention of the agent.

  In any one of the first to fourteenth aspects of the invention, the assist device includes a plurality of assist modes having different forms of change in the assist force under the same conditions as options of the operation modes, A second interface enabling designation by one of the agents, and the control device according to one of the plurality of assist modes based on a signal according to the designation output from the second interface. The actuator may be configured to control the operation (15th invention).

  According to the training apparatus of the said structure, it can switch to the assist mode of one assist mode direction among several assist modes according to an agent's intention. As a result, when the agent is performing a walking motion involving the relative motion of the upper body and legs, the assisting force can be applied to the agent in various forms reflecting the agent's intention.

  In any one of the first to fifteenth inventions, the training apparatus includes a plurality of training modes having different resistance force change forms under the same conditions as options of the operation modes, A third interface that enables designation by one of the agents, and the control device performs the actuator according to one of the training modes based on a signal corresponding to the designation output from the third interface. It may be configured to control the operation of (No. 16th invention).

  According to the training device having the configuration, the training device can be switched to one of the plurality of training modes according to the intention of the agent. As a result, when the agent is performing a walking motion accompanied by a relative motion of the upper body and the leg, the resistance force can be applied to the agent in various forms reflecting the intention of the agent.

  In any one of the first to sixteenth inventions, the control device is configured to measure the fatigue level of the agent and switch between the assist mode and the training mode based on the measurement result. Hands are also good (17th invention).

  According to the training device of the configuration, depending on the level of fatigue of the agent, a situation where an assisting force acts on the agent in the direction of relative motion of the upper body and the leg, and the upper body of the agent And a situation in which a resistance force acts on the direction of relative motion of the legs.

  In any one of the first to seventeenth inventions, the auxiliary force and the resistance force are transmitted from the output shaft of a motor as a component of the actuator to the agent via a speed reduction mechanism. The control device may be configured to adjust each of the auxiliary force and the resistance force by switching a reduction ratio of the reduction mechanism (18th invention).

  According to the training apparatus having the configuration, the assist force and the resistance force having different strengths can be applied to the agent by adjusting the speed reduction ratio of the speed reduction mechanism.

  In any one of the first to eighteenth aspects of the invention, the training apparatus is configured to transmit the resistance to the agent by regenerative braking or friction of a motor as a component of the actuator, and by regenerative braking of the motor. You may provide the electrical storage system which stores the produced | generated electric energy (19th invention).

  According to the training device having the above configuration, it is possible to change the change form of the resistance force applied to the agent according to the difference in the mechanism of the motor friction and the motor regenerative braking. In addition, electrical energy generated by regenerative braking of the motor is stored in the power storage system.

  In the training device of the nineteenth invention, a connection terminal for supplying electric energy from the power storage system to an external electronic device may be provided (20th invention).

  According to the training device having such a configuration, the electric energy generated by the regenerative braking of the motor as described above and stored in the power storage system can be used by the external electronic device.

  In any one of the first to twentieth inventions, the control device measures at least one of the agent's walking cycle, step length, walking rate, and walking ratio as a walking state variable, and the walking state The operation of the actuator may be controlled so that the measured value of the variable matches the target value (21st invention).

  According to the training device of the configuration, when the agent is walking with relative motion of the upper body and the leg, the agent can be applied with force in various forms, and the walking form can be targeted. Can be close to form.

BRIEF DESCRIPTION OF THE DRAWINGS Structure explanatory drawing of the training apparatus of this invention. The structure explanatory view of the control device of the training device of the present invention. Explanatory drawing regarding the determination method of assist torque. Explanatory drawing regarding the change form of assist torque. Explanatory drawing regarding the change form of the walking state of an agent. Explanatory drawing regarding the determination method of training torque. Explanatory drawing regarding the change form of training torque. Explanatory drawing regarding the change form of the walking state of an agent. Explanatory drawing regarding the determination method of the torque in hybrid mode. Explanatory drawing regarding the change form of the torque in hybrid mode. Explanatory drawing regarding the change form of the walking state of an agent. Explanatory drawing regarding a regenerative current.

  An embodiment of a training apparatus according to the present invention will be described with reference to the drawings.

(Configuration of training device)
A configuration of a training apparatus as an embodiment of the present invention will be described.

  The training device 1 shown in FIG. 1 includes a first mounting element 11, a second mounting element 12, and an actuator 22. As shown in FIG. 2, the training device 1 includes a state sensor 202, an operation interface 204, a control device 20, and a battery 21.

  The first mounting element 11 includes a waistband 111 pressed against the back side of the waist of the agent (human), and a band 112 wound around the abdomen to fix the waistband to the waist. The waistband 111 is made of, for example, a flexible and moderately hard resin. Actuators 22 are attached to the lower ends of the left and right sides of the hip rest 111 with a degree of freedom of rotation about the roll axis.

  The second mounting element 12 includes a band that is wound around the thigh of the leg of the agent. Of the second mounting element 22, a link member 13 for transmitting the output of the actuator 22 to the second mounting element 12 is attached to the front side of the thigh with a degree of freedom of rotation about the roll axis. The link member 13 is made of a hard resin and has a shape curved from the left and right sides of the waist of the agent toward the front side of the left and right thighs.

  The control device 20 is configured by a computer (comprised of a CPU, ROM, RAM, I / O circuit, A / D conversion circuit, etc.) built in the waist support 111 of the first mounting element 11. The control device 20 executes arithmetic processing according to a program read from the memory as appropriate for the output signal from the state sensor 202 and the output signal from the interface 204. As a result, the control device 20 controls the charging and discharging of the battery 21 and the operation of the actuator 22.

  The battery 21 is also built in the waistband 111 of the first mounting element 11. The battery 21 supplies power to the control device 20 and the actuator 22 as a power source. The battery 21 constitutes a power storage system together with an inverter (not shown), and stores the electric energy generated by the motor 221 constituting the actuator 22 after receiving it via the inverter.

  The actuator 22 includes a motor 221 and a speed reduction mechanism 222. The operation of the motor 221 and the reduction ratio of the reduction mechanism 222 are controlled by the control device 20. The output of the motor 221 after passing through the speed reduction mechanism 222 corresponds to the output of the actuator 22. The output of the actuator 22 is transmitted to the lumbar part of the agent through the first mounting element 11, and is transmitted to the leg body (directly the thigh) through the link member 13 and the second mounting element 12. Is done.

  The state sensor 202 is configured to output signals corresponding to various states of the agent. For example, the rotary encoders placed on the left and right sides of the agent's waist that output signals according to the relative angle of the agent's waist and thigh (leg) (hereinafter referred to as the “leg angle”) It corresponds to the sensor 202. In addition, when the rotor angle of the motor constituting the actuator 22 is the basis for calculating the leg angle, a hall element provided in the motor that outputs a signal corresponding to the rotor angle is employed as the state sensor 22. sell.

  The interface 204 is configured to output a signal corresponding to the mode designated by the agent. A manual mode designation button, a touch panel mode designation button, a voice recognition device, or the like attached to the first mounting element 11 corresponds to the interface 204.

(Function of training device)
The function of the training apparatus having the above configuration will be described.

  When an ON / OFF switch (not shown) is switched from OFF to ON, power supply from the battery 21 to the control device 20 is started. The control device 20 receives an output signal from the rotary encoder or the hall element as the state sensor 202 at every sampling period or calculation period, and calculates the leg angle φ of the agent and the leg angular velocity φ ′ that is a first-order time derivative thereof. To do.

  The leg angle φ represents the inclination angle of the thigh with respect to the basic frontal plane (see FIGS. 5 and 8). The leg angle φ is defined as “positive” in the situation where the thigh is ahead of the reference frontal plane. On the other hand, the leg angle φ is defined as “negative” in the situation where the thigh is behind the reference frontal plane. The basic front face is defined by a posture that includes the waist of the agent and is tilted back and forth according to the tilt of the upper body of the agent.

  Leg body angular velocity φ ′ is defined as “positive” when the thigh is approaching the reference frontal plane from behind or the thigh is moving forward with respect to the reference frontal plane. The On the other hand, the leg angular velocity φ ′ is “negative” when the thigh is approaching from the front of the reference frontal face or the thigh is moving backward from the reference frontal face. Is defined.

  Based on the calculation result, the control device 20 controls the operation of the actuator 22 according to the mode specified by the agent through the interface 204, thereby adjusting the force acting on the agent. At this time, the strength of the force can be adjusted by controlling one or both of the output of the motor 221 and the reduction ratio (gear ratio) of the reduction mechanism 222.

  This force is considered to be a torque F for swinging the agent's thigh back and forth around the hip joint with reference to the upper body. The torque F is defined as “positive” when acting on the agent to increase the leg angle φ, while defined as “negative” when acting on the agent to decrease the leg angle φ. Is done.

(In assist mode)
When the assist mode is designated by the agent, the torque F applied to the agent is adjusted according to the assist function F ₁ (φ, φ ′) based on the measured values of the leg angle φ and the leg angular velocity φ ′.

The assist function F ₁ includes the first assist domain [α ₁ <φ] × [φ ′ <0] and the second assist domain [φ <α ₁ ] × [φ ′ shown in FIG. <0], the third assist domain [φ <α ₂ ] × [0 <φ ′], and the fourth assist domain [α ₂ <φ] × [0 <φ ′], as will be described below. Defined to show different characteristics.

The first assist reference angle α ₁ that defines the boundary position between the first assist definition area and the second assist definition area, and the second assist reference angle α that defines the boundary position between the third assist definition area and the fourth assist definition area. ₂ can be set to an arbitrary value (for example, “0”).

In the first assist definition area, the assist function F ₁ approaches “0” as the leg angular velocity φ ′ approaches 0 (as the leg angle φ approaches the maximum value φmax), while the leg angle φ is the first assist. As the reference angle α ₁ is approached, the change characteristic approaches the “minimum value Fmin”. In the first assist definition area, the partial differential function ∂F ₁ / ∂φ with respect to the leg angle φ of the assist function F ₁ is “positive”, and the leg angle φ approaches “0” as the leg angle φ approaches the first assist reference angle α _1. It has a change characteristic that approaches "".

In the second assist definition area, the assist function F ₁ approaches “0” as the leg angular velocity φ ′ approaches 0 (as the leg angle φ approaches the minimum value φmin), while the leg angle φ increases to the first assist. As the reference angle α ₁ is approached, the change characteristic approaches the “minimum value Fmin”. In the second assist definition area, the partial differential function ∂F ₁ / ∂φ with respect to the leg angle φ of the assist function F ₁ is “negative”, and “0” as the leg angle φ approaches the first assist reference angle α _1. It has a change characteristic that approaches "".

In the third assist definition area, the assist function F ₁ approaches “0” as the leg angular velocity φ ′ approaches 0 (as the leg angle φ approaches the minimum value φmin), while the leg angle φ increases to the second assist. As the reference angle α ₂ is approached, the change characteristic approaches the “maximum value Fmax”. In the third assist definition area, the partial differential function ∂F ₁ / ∂φ with respect to the leg angle φ of the assist function F ₁ is “positive”, and the leg angle φ becomes closer to the second assist reference angle α ₂ as “0”. It has a change characteristic that approaches "".

In the fourth assist definition area, the assist function F ₁ approaches “0” as the leg angular velocity φ ′ approaches 0 (as the leg angle φ approaches the maximum value φmax), while the leg angle φ increases to the second assist. As the reference angle α ₂ is approached, the change characteristic approaches the “maximum value Fmax”. In the fourth assist definition area, the partial differential function ∂F ₁ / ∂φ with respect to the leg angle φ of the assist function F ₁ is “negative”, and the leg angle φ approaches “0” as the leg angle φ approaches the second assist reference angle α _2. It has a change characteristic that approaches "".

FIG. 3B shows a trajectory showing how the assist torque changes in the F ₁ −φ plane defined by the assist function F ₁ and the leg angle φ. This trajectory is configured by a series of arrows as1 to as4 corresponding to the first to fourth assist definition areas. Each shape of the arrows as1 to as4 in the F ₁ −φ plane can be expressed by an approximate curve equation, for example, F _1i = A _1i φ ² + B _1i φ + C _1i (i = 1, 2, 3, 4). .

Here, consider a situation where the agent is walking while periodically changing the left and right leg angles φ as indicated by broken lines in FIG. In this situation, by following the assist function F ₁ having the characteristics as described above, the operation of the actuator 22 is controlled so that the assist torque (auxiliary force) acts on the agent as shown by a one-dot chain line in FIG. Is done.

  That is, in this situation, the four periods change periodically in the order of the first assist period → the second assist period → the third assist period → the fourth assist period. The i-th assist period (i = 1 to 4) means a period in which the measured values (φ, φ ′) belong to the i-th assist domain.

  In the “first assist period”, the assist torque is controlled to continuously decrease from 0 to its minimum value Fmin (see FIG. 4). That is, the auxiliary force for the relative motion of the upper body and the leg (see FIGS. 5A to 5B) that causes the agent to translate the upper body forward while supporting the leg in the standing state is initially zero. Then, it is given to the agent in a change form that gradually becomes stronger.

  In the “second assist period”, the assist torque is controlled to continuously increase from its minimum value Fmin to 0 (see FIG. 4). That is, the assisting force for the relative motion of the upper body and the leg (see FIG. 5 (b) → (c)) that causes the agent to translate the upper body forward while supporting the leg in the standing state is initially strong. Then, the agent is given to the agent in such a manner that it gradually weakens and then becomes zero at the end.

  In the “third assist period”, the assist torque is controlled to continuously increase from 0 to its maximum value Fmax (see FIG. 4). That is, the assisting force for the relative movement of the upper body and the leg (see FIG. 5 (c) → (d)) that the agent mainly translates the leg in the free leg state forward is zero at first, After that, it is given to the agent in a change form that gradually becomes stronger.

  In the “fourth assist period”, the assist torque is controlled to continuously decrease from its maximum value Fmax to 0 (see FIG. 4). That is, the assisting force for the relative motion of the upper body and the leg (see FIG. 5 (d) → (a)), in which the agent mainly translates the leg in the free leg state, is strong at first, and thereafter It is given to the agent in a change form that gradually weakens and eventually becomes zero.

(In training mode)
When the training mode is designated by the agent, the torque F applied to the agent is adjusted according to the training function F ₂ (φ, φ ′) based on the measured values of the leg angle φ and the leg angular velocity φ ′.

The training function F ₂ includes the first training domain [β ₁ <φ] × [φ ′ <0] and the second training domain [φ <β ₁ ] × [φ shown in FIG. As described below in each of '<0], the third training domain [φ <β ₂ ] × [0 <φ ′] and the fourth training domain [β ₂ <φ] × [0 <φ ′]. Are defined to exhibit different characteristics.

The first training reference angle β ₁ that defines the boundary position between the first training definition area and the second training definition area, and the second training reference angle β that defines the boundary position between the third training definition area and the fourth training definition area. ₂ can be set to an arbitrary value (for example, “0”).

In the first training domain, the training function F ₂ approaches “0” as the leg angular velocity φ ′ approaches 0 (as the leg angle φ approaches the maximum value φmax), while the leg angle φ increases to the first training. As the reference angle β ₁ is approached, the change characteristic approaches the “maximum value Fmax”. In the first training domain, the partial differential function ∂F ₂ / ∂φ with respect to the leg angle φ of the training function F ₂ is “negative”, and the leg angle φ approaches the first training reference angle β _1. It has a change characteristic that approaches “0”.

In the second training domain, the training function F ₂ approaches “0” as the leg angular velocity φ ′ approaches 0 (as the leg angle φ approaches the minimum value φmin), while the leg angle φ increases to the first training. As the reference angle β ₁ is approached, the change characteristic approaches the “maximum value Fmax”. In the second training domain, the partial differential function ∂F ₂ / ∂φ with respect to the leg angle φ of the training function F ₂ is “positive”, and as the leg angle φ approaches the first training reference angle β ₁ , “0” is obtained. It has a change characteristic that approaches "".

In the third training domain, the training function F ₂ approaches “0” as the leg angular velocity φ ′ approaches 0 (as the leg angle φ approaches the minimum value φmin), while the leg angle φ increases to the second training. As the reference angle β ₂ is approached, the change characteristic approaches the “minimum value Fmin”. In the third training domain, the partial differential function ∂F ₂ / ∂φ with respect to the leg angle φ of the training function F ₂ is “negative”, and as the leg angle φ approaches the second training reference angle β ₂ , it becomes “0”. It has a change characteristic that approaches "".

In the fourth training domain, the training function F ₂ approaches “0” as the leg angular velocity φ ′ approaches 0 (as the leg angle φ approaches the maximum value φmax), while the leg angle φ increases to the second training. As the reference angle β ₂ is approached, the change characteristic approaches the “minimum value Fmin”. In the fourth training definition area, the partial differential function ∂F ₂ / ∂φ with respect to the leg angle φ of the training function F ₂ is “positive”, and “0” becomes closer to the second training reference angle β _2. It has a change characteristic that approaches "".

FIG. 6B shows a trajectory showing how the training torque changes in the F ₂ -φ plane defined by the training function F ₂ and the leg angle φ. This trajectory is formed by connecting arrows tr1 to tr4 corresponding to the first to fourth training definition areas in a ring shape. Each shape of the arrow tr1~tr4 in F ₂ -.phi plane, for example, may be represented by an approximate curve equation as ^{_{F 2i = A 2i φ 2 +}} B 2i φ + C 2i (i = 1,2,3,4) .

Here, consider a situation where the agent is walking while periodically changing the left and right leg angles φ as indicated by broken lines in FIG. By following the training function F ₂ having the characteristics as described above in this situation, the operation of the actuator 22 is performed so that the training torque (resistance force) acts on the agent as shown by a two-dot chain line in FIG. Be controlled.

  That is, in this situation, the four periods change periodically in the order of the first training period → the second training period → the third training period → the fourth training period. The i-th training period (i = 1 to 4) means a period in which the measured values (φ, φ ′) belong to the i-th training domain.

  In the “first training period”, the training torque is controlled to continuously increase from 0 to its maximum value Fmax (see FIG. 7). That is, the resistance force to the relative motion of the upper body and the leg (see FIGS. 8A to 8B) in which the agent translates the upper body forward while supporting the leg in the standing state is initially zero. Then, it is given to the agent in a change form that gradually becomes stronger.

  In the “second training period”, the training torque is controlled so as to continuously decrease from its maximum value Fmax to 0 (see FIG. 7). That is, the resistance to the relative movement of the upper body and the leg (see FIGS. 8B to 8C) that causes the agent to translate the upper body forward while supporting the leg in the standing state is initially strong. Then, the agent is given to the agent in such a manner that it gradually weakens and then becomes zero at the end.

  While the leg of the walking agent is in a standing state, the resistance changes to be powerless at first, gradually increases, then gradually decreases and finally becomes powerless again (FIG. 8 ( a) → (b) → (c)).

  In the “third training period”, the training torque is controlled to continuously decrease from 0 to the minimum value Fmin (see FIG. 7). That is, the resistance to the relative movement of the upper body and the leg (see FIGS. 8 (c) → (d)) that the agent mainly translates the leg in the free leg state forward is zero at first. After that, it is given to the agent in a change form that gradually becomes stronger.

  In the “fourth training period”, the training torque is controlled so as to continuously increase from the minimum value Fmin to 0 (see FIG. 7). That is, the resistance to the relative movement of the upper body and the leg (see FIG. 8 (d) → (a)) in which the agent mainly translates the leg in the free leg state forward is strong at first, and thereafter It is given to the agent in a change form that gradually weakens and eventually becomes zero.

  While the leg of the walking agent is in the free leg state, the resistance force is initially powerless, gradually increases, then gradually decreases and finally becomes powerless again (FIG. 8). (See (c) → (d) → (a)).

  In the training mode, when a resistance force acts on the agent due to regenerative braking of the motor 221, the motor 221 functions as a generator to generate electric energy, and this electric energy is stored in the battery 21. As shown in FIG. 12, it is generated by regenerative braking of the left and right motors 221 while the agent is walking so that the right leg angle (one-dot chain line) and the left leg angle (two-dot chain line) change. The regenerative current (solid line) is supplied to the inverter and stored in the battery 21 as electric energy.

  By providing the training device 1 with a connection terminal for an external electronic device such as a mobile phone, electric energy can be supplied from the battery 21 to the external electronic device via the connection terminal.

(In hybrid mode)
When the hybrid mode is designated by the agent, the torque F applied to the agent is adjusted according to the hybrid function F ₃ (φ, φ ′) based on the measured values of the leg angle φ and the leg angular velocity φ ′. It can be said that the hybrid mode is a mode that selectively uses the assist mode and the training mode in each walking cycle of the agent.

The hybrid function F ₃ includes the first hybrid domain [γ ₁ <φ <γ ₂ ] × [0 <φ ′] and the second hybrid domain [γ ₂ <φ] × shown in FIG. [0 <φ ′] + [γ ₃ <φ] × [φ ′ <0], third hybrid domain [γ ₄ <φ <γ ₃ ] × [φ ′ <0] and fourth hybrid domain [φ Each of <γ ₄ ] × [φ ′ <0] + [φ <γ ₁ ] × [0 <φ ′] is defined so as to exhibit different characteristics as described below.

The first hybrid reference angle γ ₁ , the second hybrid reference angle γ ₂ , the third hybrid reference angle γ _3, and the fourth hybrid reference angle γ ₄ that define the boundary position of each hybrid domain can be set to arbitrary values. .

In the first hybrid domain, the hybrid function F ₃ approaches “0” as the leg angle φ approaches the first hybrid reference angle γ ₁ , while the leg function angle ₃ approaches “2” as the leg angle φ approaches the second hybrid reference angle γ _2. It has a change characteristic approaching the maximum value Fmax ”. In the first hybrid domain, the partial differential function ∂F ₃ / ∂φ depending on the leg angle φ of the hybrid function F ₃ is “positive”.

In the second hybrid domain, the hybrid function F ₃ approaches “0” as the leg angle φ approaches the third hybrid reference angle γ ₃ while the leg angular speed φ ′ is negative, while the leg angular speed φ ′ In a positive state, the leg body angle φ has a change characteristic that approaches the “maximum value Fmax” as it approaches the second hybrid reference angle γ ₂ . In the second hybrid domain, the partial differential function ∂F ₃ / ∂φ with respect to the leg angle φ of the hybrid function F ₃ is “negative” when φ ′ is positive, and “φ” is negative when φ ′ is negative. "Positive".

In the third hybrid domain, the hybrid function F ₃ approaches “0” as the leg angle φ approaches the third hybrid reference angle γ ₃ , while the closer the leg angle φ approaches the fourth hybrid reference angle γ ₄ , It has a change characteristic approaching the “minimum value Fmin”. In the third hybrid domain, the partial differential function ∂F ₃ / ∂φ depending on the leg angle φ of the hybrid function F ₃ is “positive”.

In the fourth hybrid domain, the hybrid function F ₃ approaches “0” as the leg angle φ approaches the first hybrid reference angle γ ₁ while the leg angular speed φ ′ is positive. In the negative state, the leg body angle φ has a change characteristic that approaches the “minimum value Fmin” as it approaches the fourth hybrid reference angle γ ₄ . In the fourth hybrid domain, the partial differential function ∂F ₃ / ∂φ with respect to the leg angle φ of the hybrid function F ₃ is “negative” when the leg angular velocity φ ′ is negative, and the leg angular velocity φ ′. If is positive, it is “positive”.

FIG. 9B shows a trajectory showing a torque change mode on the F ₃ -φ plane defined by the hybrid function F ₃ and the leg angle φ. The locus includes an arrow has1 corresponding to the first hybrid domain, arrows has2 and htr2 corresponding to the assist mode and the training mode in the second hybrid domain, and an arrow has3 corresponding to the third hybrid domain, Arrows has4 and htr4 corresponding to the assist mode and the training mode in the fourth hybrid domain are formed in a ring. F ₃ -.phi arrow in the plane Has1～has4, respective shapes of htr2 and htr4, for ^{_{example, F 3k = A 3k φ 2}} + B 3k φ + C 3k (k = as1~as4, tr2, tr4) approximate curve equation as It can be expressed by

Here, consider a situation where the agent is walking while periodically changing the left and right leg angles φ as indicated by broken lines in FIG. In accordance with the hybrid function F ₃ having the above characteristics in this situation, the operation of the actuator 22 is controlled so that the torque acts on the agent as shown by the solid line in FIG.

  That is, in this situation, the four periods change periodically in the order of the first hybrid period → the second hybrid period → the third hybrid period → the fourth hybrid period. The i-th hybrid period (i = 1 to 4) means a period in which the measured values (φ, φ ′) belong to the i-th hybrid domain.

  In the “first hybrid period”, the torque is controlled so as to continuously increase from 0 to its maximum value Fmax (see FIG. 10). That is, the “assisting force” for the relative motion of the upper body and the leg (see FIGS. 11A to 11B), in which the agent mainly translates the leg in the free leg state, is initially zero. It is given to agents in a form of change that gradually increases.

  In the “second hybrid period”, the torque is controlled so as to continuously decrease from its maximum value Fmax to 0 (see FIG. 10). That is, first, the “assisting force” with respect to the upper body and the relative movement of the leg (see FIG. 11 (b) → (c)), in which the agent mainly translates the leg in the free leg state, Is given to the agent in a changing form that is strong and then gradually weakens. Then, the “resistance” against the relative movement of the upper body and the leg (see FIG. 11 (c) → (d)) in which the agent translates the upper body forward while supporting the leg in the standing state, Is given to the agent in a change form (initial load form) that gradually weakens and then becomes 0 afterwards.

  In the “third hybrid period”, the torque is controlled to continuously decrease from 0 to the minimum value Fmin (see FIG. 10). That is, the “assisting force” with respect to the relative motion of the upper body and the leg (see FIG. 11 (d) → (e)) that causes the agent to translate the upper body forward while supporting the leg in the standing state is the first. Is 0, and is given to the agent in a variation that gradually increases thereafter.

  In the “fourth hybrid period”, the torque is controlled so as to continuously increase from its minimum value Fmin to 0 (see FIG. 10). That is, first, the “assisting force” for the relative motion of the upper body and the leg (see FIG. 11 (e) → (f)) in which the agent translates the upper body further forward while supporting the leg in the standing state. However, it is given to the agent in a change form that is strong at first and then gradually weakens. Then, the “resistance” against the relative movement of the upper body and the leg (see FIG. 11 (f) → (a)) in which the agent mainly translates the leg in the free leg state forward is strong at first, After that, it is given to the agent in a change form (initial movement load form) that gradually weakens and finally becomes 0.

  As described above, in the hybrid mode, the assist period and the training period are mixed between each walking cycle of the agent (see FIG. 10).

Note that the timing for switching from the assist period to the training period can be changed by changing the hybrid reference angle γ _i (i = 1 to 4). Further, by changing the hybrid reference angle γ _i , the change form of the resistance force given to the agent during the training period can be changed to various forms such as the end load form and the initial load form.

  Even in the hybrid mode, while the resistance force acts on the agent due to the regenerative braking of the motor 221, the motor 221 functions as a generator to generate electric energy, and this electric energy is stored in the battery 21 (see FIG. 12). ).

(Effect of training device)
According to the training device 1 that exhibits the above-described function, a situation in which a forward force (auxiliary force) acts on the agent relative to the direction of the relative motion of the upper body and the leg by using the mode properly, A situation in which a force (resistance force) in the opposite direction acts on the direction of relative motion of the upper body and the leg is realized. Thereby, when the agent is performing a walking motion accompanied by the relative motion of the upper body and the leg, a force can be applied to the agent in various forms.

  Further, by specifying the mode through the interface 204, the assist mode and the training mode are switched according to the intention of the agent. As a result, when the agent is performing a walking motion involving the relative motion of the upper body and legs, it is possible to apply force to the agent in various forms that reflect the intention of the agent.

  For example, by specifying the assist mode or the hybrid mode through the interface 204 while the agent is using the training device 1 in the training mode, the training device 1 can be used in the specified mode instead of the training mode. .

  The operation of the actuator 22 is controlled so that the assisting force continuously changes from 0 from the start time to the end time of the assist mode and then returns to 0, and the resistance force continuously increases from 0 to the end time of the training mode. Then, the operation of the actuator 22 is controlled so as to return to 0.

  For example, when the assist mode is designated by the agent while the training device 1 is operating according to the training mode, the torque is maintained at 0 after the training torque becomes 0, and the initial value of the assist torque becomes 0. The operation in the assist mode is started at such timing. The same applies to the case where the training mode is designated by the agent while the training device 1 is operating according to the assist mode.

  Thereby, each of the auxiliary force and the resistance force acting on the agent can be smoothly changed. In addition, the polarity or direction of the force acting on the agent can be smoothly reversed at the time of switching between the assist mode and the training mode.

  Further, according to the training device 1, according to at least one of the relative posture (leg angle φ) of the upper body and legs of the agent and the time-series change mode (leg angular velocity φ ′) of the relative posture. The strength and direction of the resistance force applied to the agent can be changed by regenerative braking of the motor 221 (see FIGS. 7 and 10). Accordingly, in view of the walking motion form of the agent represented by the relative posture of the upper body and the leg, the kinetic energy is converted into electric energy by regenerative braking of the motor 221 while applying an appropriate resistance force to the agent. (See FIG. 12).

(Other embodiments)
The control device 20 determines whether each of the leg of the agent is in the standing state and the free leg state, and the change form of each of the assisting force and the resistance force differs depending on the determination result. The operation may be controlled. In particular, when it is determined that the agent's leg is in the standing state, adjustment is made so that each of the assisting force and the resistance force is stronger than when the agent's leg is determined to be in the swinging state. May be. Here, the strength of the force means the strength of the force or the amplitude of the torque in a situation where the factors other than the standing state and the free leg state are the same (see FIGS. 4, 7, and 10).

  According to the training device of the configuration, the upper body of the agent and the body of the agent according to the standing state of the leg (the state where the leg is landing) and the free leg state (the state where the leg is leaving the floor) Each change form of the assisting force and the resistance force acting on the relative motion of the legs can be made different.

  Further, in view of the difference in the level of kinetic energy that changes the relative posture of the upper body and the leg depending on whether the leg is in the standing or swinging state, the strength of the resistance force, that is, regenerative braking The amount of electrical energy due to can be adjusted. Specifically, when the leg is in the standing state, the relative posture of the upper body and the leg is changed by the translational force or the translational inertial force of the upper body as compared with the case where the leg is in the free leg state. Kinetic energy is high. In view of this point, in the standing state, a large amount of electric energy can be obtained by increasing the resistance force acting on the agent. On the other hand, in the swing leg state, the electric energy obtained by weakening the force acting on the agent is suppressed to a small extent. Accordingly, in view of the walking motion form of the agent represented by the swinging leg and the standing leg, the kinetic energy can be converted into electric energy by the regenerative braking of the motor 221 while applying an appropriate resistance force to the agent.

  In addition to the measured values of the leg angle φ and leg angular velocity φ ′, the output signal of the acceleration sensor for measuring the vertical acceleration of the upper body of the agent, or the sole of the agent (preferably heel and toe Based on the output signal of the pressure sensor attached to both), a distinction between the standing state and the free leg state can be determined.

  The control device 20 may recognize the posture of the walking surface of the agent, and may control the operation of the actuator 22 so that the change forms of the auxiliary force and the resistance force differ according to the difference in the recognition result.

  For example, when the walking surface posture is recognized as a rising inclination posture, the assisting force is adjusted to be stronger than when the walking surface posture is recognized as a flat posture or a downward inclination posture. May be. Further, when the walking surface posture is recognized as a flat posture, the assisting force may be adjusted to be stronger than when the walking surface posture is recognized as a descending tilt posture.

Furthermore, when the walking surface posture is recognized as a descending tilt posture, the resistance is adjusted to be stronger than when the walking surface posture is recognized as a flat posture or a rising tilt posture. May be. In addition, when the walking surface posture is recognized as a flat posture, the resistance may be adjusted to be stronger than when the walking surface posture is recognized as a rising inclination posture. According to the configuration of the training device, in view of the difference in the strength of the relative motion of the upper body and legs required for walking according to the posture of the ground or floor surface (walking surface) on which the agent walks, Each of the appropriately adjusted auxiliary force and resistance force can be applied to the agent.

  Also, depending on the difference in the posture of the ground or floor (walking surface) on which the agent is walking, the change form of the resistance force to the relative motion of the upper body and legs of the agent, that is, the electric energy obtained by regenerative braking The changing form of can be adjusted.

  Specifically, the strength is appropriately adjusted in view of the difference in strength of the relative motion of the upper body and legs required for walking according to the posture of the ground or floor (walking surface) on which the agent walks. The applied resistance can be applied to the agent. Accordingly, in consideration of the posture of the surface on which the agent is walking, the kinetic energy can be converted into electric energy by regenerative braking of the motor 221 while applying an appropriate resistance force to the agent.

  The posture of the walking surface can be determined based on the movement pattern of the leg of the agent, for example, according to the technique described in Japanese Patent No. 3833921 or Japanese Patent No. 3908735. The state where the agent is climbing the hill or the stairs corresponds to the state where the walking surface is in the rising inclination posture. The state where the agent is walking on the flat ground corresponds to the state where the walking surface is in a flat posture. The state where the agent is going down the hill or the stairs corresponds to the state where the walking surface is in the downward inclined posture. In addition, the actual inclination angle may be recognized as the posture of the walking surface instead of the type of posture of the walking surface.

  The control device 20 may be configured to measure the fatigue level of the agent and switch between the assist mode and the training mode based on the measurement result.

  According to the training device of the configuration, depending on the level of fatigue of the agent, a situation where an assisting force acts on the agent in the direction of relative motion of the upper body and the leg, and the upper body of the agent And a situation in which a resistance force acts on the direction of relative motion of the legs.

  For example, the fatigue level of an agent can be evaluated higher as the duration of use of the training apparatus 1 by the agent (the time during which the ON / OFF switch is maintained) is longer. The slower the walking speed of the agent (the walking speed is obtained from the leg body angular speed), the higher the degree of fatigue may be evaluated. Further, the fatigue level may be evaluated higher as the accumulated use time of the training device 1 in the training mode is longer. As the cumulative work amount (work amount = training torque × angular displacement) due to the training torque (resistance force) increases, the degree of fatigue may be evaluated higher. The higher the heart rate or blood pressure of the agent, the higher the degree of fatigue may be evaluated.

  The training device 1 may include a second interface that allows designation by one agent among a plurality of assist modes in which the assist force changes in the same condition. Further, the control device 20 may control the operation of the actuator 22 according to one of the assist modes based on the signal according to the designation output from the second interface.

  According to the training apparatus of the said structure, it can switch to the assist mode of one assist mode direction among several assist modes according to an agent's intention. As a result, when the agent is performing a walking motion involving the relative motion of the upper body and legs, the assisting force can be applied to the agent in various forms reflecting the agent's intention.

  The training apparatus 1 may include a third interface that enables designation by an agent in one training mode among a plurality of training modes that have different resistance force changes under the same conditions. Further, the control device 20 may control the operation of the actuator 22 according to one of a plurality of training modes based on the signal according to the designation output from the third interface.

  According to the training device having the configuration, the training device can be switched to one of the plurality of training modes according to the intention of the agent. As a result, when the agent is performing a walking motion accompanied by a relative motion of the upper body and the leg, the resistance force can be applied to the agent in various forms reflecting the intention of the agent.

  The control device 20 measures at least one of the walking cycle, step length, walking rate, and walking ratio of the agent as a walking state variable, and controls the operation of the actuator 22 so that the measured value of the walking state variable matches the target value. It may be configured to.

  According to the training device of the configuration, when the agent is walking with relative motion of the upper body and the leg, the agent can be applied with force in various forms, and the walking form can be targeted. Can be close to form.

  The “step length” can be calculated based on the length of the leg of the agent (stored in the memory) and the maximum positive and negative values of the leg angle φ. The “walking rate (number of steps per unit time)” is grasped by the time change pattern of the output signal of the acceleration sensor that measures the walking cycle or the vertical acceleration of the agent that can be recognized by the time change pattern of the leg angle φ. It can be calculated based on the landing timing of the left and right legs. The “walking ratio” is calculated as the ratio of “step length” to “walking rate”.

  The phase of the torque F may be controlled so that the phase difference with respect to the leg angle φ matches the target phase difference. As a training apparatus 1 for realizing such control, for example, Japanese Patent No. 3930399, Japanese Patent No. 3950149, Japanese Patent No. 4008464, Japanese Patent No. 400008465, Japanese Patent No. 4220567, Japanese Patent No. 4234765, Japanese Patent No. 4271711 Further, walking assist devices (exercise assist devices or motion induction devices) described in Japanese Patent No. 4271713 may be employed.

  In the walking assistance device, the target phase difference is automatically adjusted according to the mode designated by the agent through the interface 204.

(Phase difference control method in assist mode)
The assist mode is realized by adjusting the target phase difference so that at least the leg angle φ becomes (π / 2) at the time when each of the leg angle φ shows the maximum value φmax and the minimum value φmin (see FIG. 4). .

  At the time when the leg angle φ shows the minimum value φmin (the transition point from the second assist period to the third assist period), the phase difference of the torque F (F ′> 0) with respect to the leg angle φ is the same as the target phase difference. (Π / 2). At the time when the leg angle φ shows the maximum value φmax (the transition point from the fourth assist period to the first assist period), the phase difference of the torque F (F ′ <0) with respect to the leg angle φ is the same as the target phase difference. (Π / 2).

Since the target phase difference is fluidly adjusted in the vicinity of (π / 2), the phase of the torque F is controlled so that the phase difference with respect to the leg angle φ varies. At the time when the torque F shows the minimum value Fmin (the transition point from the first assist period to the second assist period), the phase difference of the torque F with respect to the leg angle φ (φ <0, φ ′ <0) is about (π / 2) (1-α ₁ / φmin). At the time when the torque F shows the maximum value Fmax (the transition point from the third assist period to the fourth assist period), the phase difference of the torque F with respect to the leg angle φ (φ> 0, φ ′> 0) is about (π / 2) (1-α ₂ / φmax).

(Phase difference control method in training mode)
The training mode is realized by adjusting the target phase difference so that at least the leg angle φ becomes − (π / 2) at the time when the leg angle φ shows the maximum value φmax and the minimum value φmin, respectively (see FIG. 7). ).

  At the time when the leg angle φ shows the minimum value φmin (the transition point from the second training period to the third training period), the phase difference of the torque F (F ′ <0) with respect to the leg angle φ is the same as the target phase difference. -(Π / 2). At the time when the leg angle φ shows the maximum value φmax (at the time of transition from the fourth training period to the first training period), the phase difference of the torque F (F ′> 0) with respect to the leg angle φ is the same as the target phase difference. -(Π / 2).

Since the target phase difference is fluidly adjusted in the vicinity of − (π / 2), the phase of the torque F is controlled so that the phase difference with respect to the leg angle φ varies. At the time when the torque F reaches the maximum value Fmax (at the time of transition from the first training period to the second training period), the phase difference of the torque F with respect to the leg angle φ (φ> 0, φ ′ <0) is about − ( π / 2) (1-β ₁ / φmax). At the time when the torque F shows the minimum value Fmin (at the time of transition from the third training period to the fourth training period), the phase difference of the torque F with respect to the leg angle φ (φ <0, φ ′> 0) is about − ( π / 2) (1-β ₂ / φmin).

(Phase difference control method in hybrid mode)
By adjusting the target phase difference so that the leg angle φ is larger than − (π / 2) and smaller than (π / 2) at the time when the leg angle φ indicates each of the maximum value φmax and the minimum value φmin, A hybrid mode is realized (see FIG. 10).

  At the time when the leg angle φ reaches the maximum value φmax (intermediate time of the second hybrid period), the phase difference of the torque F (F> 0, F ′ <0) with respect to the leg angle φ is the same as the target phase difference (π / 2) Lower positive value. At the time when the leg angle φ shows the minimum value φmin (intermediate time point of the fourth hybrid period), the phase difference of the torque F with respect to the leg angle φ becomes a positive value lower than (π / 2), similar to the target phase difference. Yes.

Since the target phase difference is adjusted fluidly, the phase of the torque F is controlled so that the phase difference with respect to the leg angle φ varies. The phase difference of the torque F with respect to the leg angle φ (φ <0, φ ′> 0) at the time when the torque F shows 0 (F ′> 0) (the transition point from the fourth hybrid period to the first hybrid period) Is about (π / 2) (1-γ ₁ / φmin). At the time when the torque F reaches the maximum value Fmax (at the time of transition from the first hybrid period to the second hybrid period), the phase difference of the torque F with respect to the leg angle φ (φ> 0, φ ′> 0) is about (π / 2) (1-γ ₂ / φmax). The phase difference of the torque F with respect to the leg angle φ (φ> 0, φ ′ <0) at the time when the torque F shows 0 (F ′ <0) (the transition point from the second hybrid period to the third hybrid period) Is about (π / 2) (1-γ ₃ / φmax). At the time when the torque F shows the minimum value Fmin (the transition point from the third hybrid period to the fourth hybrid period), the phase difference of the torque F with respect to the leg angle φ (φ <0, φ ′ <0) is about (π / 2) (1-γ ₄ / φmin).

DESCRIPTION OF SYMBOLS 1 ... Training apparatus, 11 ... 1st mounting element, 12 ... 2nd mounting element, 20 ... Control apparatus, 21 ... Battery, 22 ... Actuator, 221 ... Motor, 222 ... Deceleration mechanism

Claims (21)

A first mounting element and a second mounting element mounted on each of an upper body and a leg of an agent; an actuator; and a control device that controls the operation of the actuator; A training device configured to transmit to the agent a force generated by operation of the actuator through an element,
The assist mode in which the assist force for the relative motion of the upper body and the leg acts on the agent and the training mode in which the resistance force for the relative motion of the upper body and the leg acts on the agent are used separately. Features training equipment. The training device according to claim 1,
The control device adjusts at least one of the auxiliary force and the resistance force as a function of a posture variable representing a relative posture of the leg with respect to the upper body of the agent and a part or all of the time derivative thereof as a variable. It is comprised so that it may carry out, The training apparatus characterized by the above-mentioned. The training device according to claim 2,
The control device is
The posture variable acceleration approaches 0 as the posture variable approaches the maximum value when the posture variable is greater than or equal to the first assist reference value and the posture change speed that is a first-order time derivative of the posture variable is negative. In the first assist definition region, the assist force is controlled according to a function having a change characteristic that approaches the minimum value as the posture variable approaches the first assist reference value while the posture variable approaches the maximum value. And
In the second assist definition area in which the posture variable is less than the first assist reference value and the posture change speed is negative and the posture change speed approaches 0 as the posture variable approaches a minimum value. The assisting force is controlled according to a function having a change characteristic that approaches the minimum value as the posture variable approaches the first assist reference value while the posture variable approaches the minimum value.
The posture variable is less than the second assist reference value, and, in a third assist domain of the attitude variable the posture change rate is positive approaches the attitude rate of change 0 closer to the minimum value, the posture The auxiliary force is controlled according to a function having a change characteristic that approaches a maximum value as the posture variable approaches the second assist reference value while the variable approaches a minimum value,
The attitude variable is not less the second assist reference value or more, and, in the fourth assist domain approaching the posture change speed is 0 closer to the posture variable maximum value the posture change rate is positive, the The auxiliary force is configured to control the auxiliary force according to a function having a change characteristic that approaches the maximum value as the posture variable approaches the second assist reference value while the posture variable approaches the maximum value. A training device characterized by. The training device according to claim 3,
In the plane defined by the auxiliary force and the posture variable, the control device uses the approximate expression of a curve representing a change mode of the auxiliary force and the measured value of the posture variable and the posture change speed. A training device configured to control force. In the training apparatus according to any one of claims 2 to 4,
The control device is
The posture change speed approaches zero as the posture variable is greater than or equal to the first training reference value and the posture change speed, which is a first-order time derivative of the posture variable, is negative and the posture variable approaches a maximum value. In the first training domain, the resistance is controlled according to a function having a change characteristic that approaches the maximum value as the posture variable approaches the first training reference value while the posture variable approaches the maximum value. And
In the second training domain where the posture variable is less than the first training reference value and the posture change rate is negative and the posture variable rate approaches 0 as the posture variable approaches a minimum value , While the posture variable approaches 0 as the minimum value approaches, the resistance is controlled according to a function having a change characteristic that approaches the maximum value as the posture variable approaches the first training reference value,
In the third training domain where the posture variable is less than the second training reference value and the posture change rate is positive and the posture change rate approaches 0 as the posture variable approaches a minimum value , The resistance is controlled according to a function having a change characteristic that approaches a minimum value as the posture variable approaches the second training reference value, while the variable approaches 0 as the variable approaches a minimum value.
In the fourth training domain where the posture variable is greater than or equal to the second training reference value and the posture change rate is positive and the posture change rate approaches 0 as the posture variable approaches a maximum value , The posture variable is configured to control the resistance force according to a function having a change characteristic that approaches the minimum value as the posture variable approaches the second training reference value while the posture variable approaches the maximum value. A training device characterized by. The training device according to claim 5, wherein
In the plane defined by the resistance force and the posture variable, the control device uses the approximate expression of a curve representing a change mode of the resistance force and the resistance variable and the measured value of the posture change speed. A training device configured to control force. In the training apparatus according to any one of claims 1 to 6,
The control device controls the phase difference of the force acting on the agent by the operation of the actuator with respect to a posture variable representing a relative posture of the leg with respect to the upper body of the agent. A training apparatus configured to act on the agent as one of the resistance forces. In the training apparatus according to any one of claims 1 to 7,
The control device determines whether the leg of the agent is in a standing state or a free leg state, and changes in the auxiliary force and the resistance force according to the determination result are different. A training apparatus configured to control operation of the actuator. The training device according to claim 8, wherein
When the control device determines that the agent's leg is in a standing state, the control device determines that the assist force and the resistance force are lower than when the agent's leg is determined to be in a swinging state. A training apparatus configured to control the operation of the actuator so that each of them becomes stronger. In the training apparatus according to any one of claims 1 to 9,
The control device recognizes the posture of the walking surface of the agent, and controls the operation of the actuator so that the change forms of the auxiliary force and the resistance force differ according to the difference in the recognition result. It is comprised in the training apparatus characterized by the above-mentioned. The training device according to claim 10, wherein
When the control device recognizes that the posture of the walking surface is an upward inclined posture, the assisting force is stronger than when the posture of the walking surface is recognized to be a flat posture or a downward inclined posture. When the movement of the actuator is controlled and the posture of the walking surface is recognized as a flat posture, the auxiliary force is less than that when the posture of the walking surface is recognized as a downward inclination posture. A training apparatus configured to control the operation of the actuator to be strong. The training device according to claim 10, wherein
When the control device recognizes that the posture of the walking surface is a downward inclined posture, the resistance force is stronger than when the controller recognizes that the posture of the walking surface is a flat posture or an upward inclined posture. When controlling the operation of the actuator and recognizing that the posture of the walking surface is a flat posture, the resistance is less than when recognizing that the posture of the walking surface is a rising inclination posture. A training apparatus configured to control the operation of the actuator to be strong. In the training apparatus according to any one of claims 1 to 12,
The control device controls the operation of the actuator so that the assist force continuously changes from 0 and returns to 0 from the start time to the end time of the assist mode and ends from the start time of the training mode. A training apparatus configured to control the operation of the actuator so that the resistance force continuously changes from 0 and returns to 0 over time. In the training apparatus according to any one of claims 1 to 13,
A first interface that enables designation by one of the assist mode and the training mode is provided, and the control device is configured to perform the assist mode based on a signal corresponding to the designation output from the first interface. And a training device configured to control the operation of the actuator according to one of the training modes. In the training apparatus according to any one of claims 1 to 14,
A plurality of assist modes having different forms of change in the assist force under the same conditions, as options for the operation mode,
A second interface that enables designation by one of the plurality of assist modes is provided, and the control device is configured to output the plurality of assist modes based on a signal corresponding to the designation output from the second interface. The training apparatus is configured to control the operation of the actuator according to one of them. In the training apparatus according to any one of claims 1 to 15,
A plurality of training modes having different forms of change of the resistance force under the same conditions, as options for the operation mode,
A third interface that enables designation by one of the plurality of training modes, and wherein the control device includes, based on a signal corresponding to the designation output from the third interface, A training apparatus configured to control the operation of the actuator according to one. The training device according to any one of claims 1 to 16,
The training device, wherein the control device is configured to measure a fatigue level of the agent and switch between the assist mode and the training mode based on the measurement result. The training device according to any one of claims 1 to 17,
Each of the auxiliary force and the resistance force is transmitted from the output shaft of a motor as a component of the actuator to the agent via a speed reduction mechanism,
The training device, wherein the control device is configured to adjust each of the auxiliary force and the resistance force by switching a reduction ratio of the reduction mechanism. In the training apparatus according to any one of claims 1 to 18,
It is configured to transmit the resistance force to the agent by regenerative braking or friction of a motor as a component of the actuator, and includes a power storage system that stores electrical energy generated by the regenerative braking of the motor. Training equipment. The training device according to claim 19, wherein
A training apparatus comprising a connection terminal for supplying electrical energy from the power storage system to an external electronic device. In the training apparatus according to any one of claims 1 to 20,
The control device measures at least one of the walking period, step length, walking rate, and walking ratio of the agent as a walking state variable, and the operation of the actuator so that the measured value of the walking state variable matches a target value. A training apparatus characterized by being configured to control.