JP4449133B2 - Control method for limb body drive device - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a limb drive device that drives a limb along a predetermined trajectory, corrects a decrease in whole body function, and performs training for maintaining and improving a healthy state.
[0002]
[Prior art]
In the following invention, among so-called exercise therapies, an object is exercise training accompanied by generation of muscular strength by the voluntary will of the patient, that is, automatic assistance exercise. This training is performed after the joint range of motion training immediately after the occurrence of a disorder such as a central nervous system disease such as stroke or cough injury, and proceeds to standing balance training or walking training in standing posture through this training. . Patients undergoing this training have weak muscle strength and are often difficult to support their legs. In the automatic assistance exercise, training to relearn the motor function is performed instead of simply restoring the muscle strength. Therefore, it is important to sense the weak muscular strength generated by the patient, relieve the weight of the limbs, repeat the exercise along the predetermined trajectory, and relearn the exercise.
[0003]
Conventionally, articulation devices used in orthopedics or exercise therapy devices used in the physical therapy department include the following.
1) For example, in JP-A-60-179062, JP-A-60-232158, JP-A-61-170464, JP 4-14028, the operating angle of a limb is set numerically, A so-called continuous passive movement is performed in which the limb of the patient is moved at a constant speed according to the angle. This is mainly intended for joint range of motion training. For example, Manson's L4K is a product based on these inventive techniques.
2) In addition, in Japanese Patent Publication No. 57-44337 and Japanese Patent Publication No. 3-54587, in addition to the passive movement, automatic movement such as isometric movement, isotonic movement, and constant speed movement can be performed for the purpose of strengthening muscle strength. . In the passive motion, if the working angle is set as time series data using so-called direct teaching means for directly moving the limb in addition to the previous angle input, the limb can be moved according to the time series data. In addition, the automatic movement is a movement in which the limb actively exerts force unlike the passive movement in which the limb is moved by the apparatus.
[0004]
[Problems to be solved by the invention]
However, the conventional techniques are insufficient for automatic assistance exercise training in both cases 1) and 2).
The reason for this is that in the case of 1), the range of motion of the joint can be trained, but only passive motion can be performed, so that automatic assisting motion according to the muscular strength generated by the patient cannot be performed. In the case of the above 2), the limbs can certainly perform automatic exercises that actively exert their power, but isometric, isotonic, and constant velocity exercises recover muscle strength to some extent. It can be applied to patients with the above condition and cannot be applied after joint range of motion training after the onset of the above-mentioned diseases. This is because
In any case, among so-called exercise therapy, there is a problem that it cannot be applied to exercise training accompanied by generation of weak muscle force of a patient, that is, automatic assistance exercise.
[0005]
[Means for Solving the Problems]
In order to solve the above problems, the control method of the limb body drive device according to the present invention is the control method according to claim 1, wherein the limb body of the patient is supported by the grip portion and moved, and the grip portion is provided. Load detection means for detecting a load applied to the limb, trajectory setting means for storing a periodic motion pattern of the grasping portion taught in advance and outputting it as trajectory information, the trajectory information output from the trajectory setting means, and the trajectory In the control method of the limb body drive device, comprising: a drive unit that corrects the trajectory using load information obtained by load detection means and outputs and controls the trajectory, and performs joint movement of the patient by the corrected trajectory. The drive unit outputs a time series number assigned along the time series of the teaching trajectory to the trajectory setting means, and a point on the teaching trajectory corresponding to the time series number is transmitted from the trajectory setting means. Read and compare the distance r to be moved from the point corresponding to the current time series number with the distance from the current teaching point to the next teaching point, and if the distance r is smaller than the distance between the teaching points The target position of the gripping part is updated, a drive command for the target position calculated by an internal impedance model is output to the mechanism part, and the mechanism part drives the gripping part by servo control for each control cycle. It is characterized in that it is to be performed.
When calculating the distance r, the inclination of the vector t from the point corresponding to the current time series number to the point on the teaching path corresponding to the next time series number is calculated as the inclination of the teaching path. When the inclination of the vector t with respect to the work coordinate system is other than vertical, the position deviation is calculated from the load information using a virtual impedance model in the trajectory setting means, and corresponds to the current time series number. The distance r is a projection of a vector from the point on the teaching trajectory to the deviation position onto the vector t.
Furthermore, when the inclination of the vector t with respect to the work coordinate system is vertical, the distance r is set to a predetermined value.
Then, when calculating the position deviation in the virtual impedance dance model in the trajectory setting means, the load information is converted from the coordinate system based on the load detection means to the work coordinate system, The limb body generated force vector obtained by subtracting the weight of the limb from the load information by a predetermined ratio is used as a limb body generated force vector, the coordinates of the limb body generated force vector are converted into a coordinate system based on the taught trajectory, and the virtual impedance model is followed on the taught trajectory A positional deviation is calculated, and the positional deviation is converted from a coordinate system on the teaching trajectory into a positional deviation of the work coordinate system.
According to a second aspect of the present invention, when the target position is updated, there is a restriction on either the number of updated time series numbers or the size of the distance r.
According to a third aspect of the present invention, the virtual impedance dance model in the trajectory setting means is a second-order dynamic model based on virtual inertia, virtual viscosity, or virtual stiffness, or a first order based on virtual viscosity / virtual stiffness. It is a dynamic model of the system.
According to a fourth aspect of the present invention, when the target position is updated, at least one of the time series numbers is updated.
According to the invention of claim 5 , the load information includes a load generated arbitrarily from the limb of the patient, a load due to the weight of the limb, and a load generated in the interaction between the limb and the device. It is a value measured by a measuring means as vector information having a direction and a large size.
Furthermore, according to the sixth aspect of the present invention, the ratio of subtracting the weight of the limb can be adjusted when obtaining the limb body force vector.
According to the driving method of the limb body driving device configured as described above, it is possible to perform an automatic assisting exercise accompanied by the generation of muscle force by the voluntary will of the patient.
[0006]
Hereinafter, a driving method according to the present invention for a limb driving apparatus will be described with reference to the drawings. FIG. 1 is a block diagram showing the configuration of a limb body driving device to which the driving method of the present invention is applied. In the figure, reference numeral 7 denotes a lower limb of a patient who performs treatment or training. Reference numeral 1 denotes a mechanism for moving the lower limb 7, which includes a base 2, drive shafts 3 A, 3 B, and 3 C, links 4 A and 4 B that connect between the base, a grip 6 that grips the lower limb 7, and load measuring means 5. Has been. The drive shafts 3A, 3B, and 3C are equipped with a motor, a speed reducer, an angle / angular velocity sensor, and the like, and the drive shafts 3A, 3B, and 3C are driven by these and a servo amplifier for driving the motor built in the base 2. With this configuration, when the lower limb 7 is fixed to the grasping portion 6, the mechanism portion 1 can cause the lower limb 7 to move. On the other hand, the load measuring means 5 measures the load applied to the lower limb 7 to generate load information f, which is amplified by the load sensor amplifier 8 and sent to the drive unit 9. In the drive unit 9, a virtual impedance model set inside, trajectory information Xt sent from the trajectory setting unit 10 storing the taught trajectory in advance, load information F, and built-in teaching point updating means Based on the result, the trajectory is corrected and a drive command Xo is sent to the mechanism unit 1. The mechanism unit 1 is driven by the drive shafts 3A, 3B, and 3C, and drives the limb body 7 along the corrected trajectory. In the trajectory setting unit 10, time-series data of the position / posture of the gripping unit 6, time-series data of the angles of the drive shafts 3 </ b> A, 3 </ b> B, and 3 </ b> C, or teaching trajectory information such as the bending angle of each joint of the lower limb 7 is obtained. It is received from the mechanism unit 1 and stored in advance as time series data in a built-in storage unit. The time-series data is a pattern in which the exercise cycle from the beginning to the end of exercise is a pattern, and the treatment operation is performed by repeating the pattern a certain number of times.
[0007]
Next, the control method of the present invention will be described.
FIG. 2 is a flowchart showing an outline of the flow of the control method of the present invention. In the figure, when the trajectory setting unit 10 (FIG. 1) receives a time series number index (hereinafter referred to as “index”) of a teaching trajectory from the drive unit 9 (FIG. 1), the time is read from the storage unit stored in advance. The data of the teaching point tdata corresponding to the sequence number index is read, and the teaching point tdata is updated (S1).
When the teaching point tdata is updated, the impedance is controlled in the drive unit 9 using the teaching point as a target value (S2). At that time, a drive command Xo is generated as a new target value by processing to be described later, and is sent to the mechanism unit 1.
When the mechanism unit 1 receives the drive command Xo that has become the new target value, the mechanism unit 1 performs servo control, and the grip unit 6 is driven along the corrected trajectory (S3).
Next, it is determined whether or not the control of the above procedure is to be ended (S4). If not, the procedure returns to step S1 and the above procedure is repeated. The above procedure, which is repeated, is performed every control cycle.
[0008]
Next, a specific procedure for updating the teaching point tdata (S1) will be described with reference to the flowchart of FIG.
First, an index indicating a time-series number of data of the teaching point tdata is initially set (S10).
Next, it is determined whether this index is equal to or less than the maximum value max_index (S11). If it is below, the time series number index of the teaching point tdata is updated and newly set as new_index (S12).
Next, the teaching point tdata is based on the teaching points tdata (old_index) and tdata (new_index) corresponding to the old_index and the updated new_index in the storage unit (not shown) of the trajectory setting unit 10 (FIG. 1). A distance r traveling from (old_index) is calculated (S13).
Then, it is determined whether the distance r is smaller than the distance between the teaching points (S14). If the distance r is smaller than the distance between the teaching points, the index and the target position tdata (index) are determined, and again set as index = new_index (S15), and the series of processes is terminated.
Here, it is assumed that at least one index is updated. In this way, at least one of the reasons for updating the index is that when providing a limb driving apparatus that can perform exercise training accompanied by generation of muscle force by the patient's voluntary intention, that is, automatic assistance exercise, the load caused by the voluntary movement of the patient is increased. If it does not occur, perform a passive movement, and if a patient's voluntary movement causes a load, an automatic assistance movement can be performed, so that automatic load generation is not sufficient in patients with extremely weak muscle strength However, a series of training operations can be performed without stopping.
If the distance r is not smaller than the distance between the teaching points in step S14, the process returns to step S11 and the above procedure is repeated.
If the index is not less than or equal to the maximum value max_index in step S11, the process in step S15 is performed and the series of processes is terminated.
[0009]
Next, the procedure (S13) for calculating the distance r traveling from the teaching point tdata (old_index) will be described with reference to the flowchart of FIG.
First, when the time point number index of the teaching point tdata and the data of the teaching point tdata are received, a relative vector t is set between the teaching points as in the following equation (S20).
t = tdata (index + 1) −tdata (index)
Next, it is determined whether or not the inclination of the relative vector t is vertical (S21).
In the case of the vertical, for example, an exception process calculation such as assigning a constant value to r is performed (S25), and the series of processes ends.
If the result of determination in step S21 is not vertical, a position deviation is calculated from the teaching point tdata (index) based on the vector t and load information F measured by a force sensor or the like (S22). Then, a relative vector x from the teaching point tdata (index) to the displaced point is calculated, and the position deviation vector on the work coordinate system is changed to x (S23).
Further, the projection r from the vector x to the vector t is calculated by the following equation, and the series of processes is terminated (S24).
r = (x · t) / (t · t)
Here, “·” represents an inner product of vectors.
[0010]
Next, the procedure (S22) for calculating the position deviation from the teaching point tdata (index) will be described with reference to the flowchart of FIG. A relative vector defined between teaching points is expressed as t = (tx, ty) in a coordinate system on a two-dimensional plane, for example. Based on the data, the slope η of the relative vector t in the work coordinate system is calculated as η = atan2 (ty, tx) (S30).
On the other hand, when the load information F is received, a coordinate conversion calculation in the so-called robot engineering from the sensor coordinate system of the load to the work coordinate system is performed (S34). When the conversion result is received, in order to calculate the limb body generating force, the limb body generating force calculated in advance is subtracted from the external force to calculate the limb body generating force (S35). From the calculation result and the inclination η, a coordinate system on the teaching trajectory of the limb generated force is defined along the inclination η, and coordinate conversion to the coordinate system on the teaching trajectory is performed on the generated force (S31). ). Then, a position deviation is calculated by impedance control, which will be described later (S32), and coordinate conversion from the coordinate system on the teaching trajectory to the working coordinate system is performed for the position deviation (S33), and a series of processing is completed. The position deviation of the working coordinate system is obtained.
[0011]
Here, the impedance control performed in step S32 will be described. In so-called impedance control, a so-called secondary dynamic model of inertia M, viscosity B, and rigidity K or a so-called primary dynamic model of viscosity B and rigidity K is used as a virtual impedance model. When external force, that is, load information F is input and the position correction amount ΔX is output, these relationships are expressed by the following equations. However, the following is expressed in the frequency domain.
ΔX (s) = F (s) / { Ms ² + Bs + K}
At this time, the target trajectory Xo of the arm is corrected as follows.
Xo (s) = Xt (s) + ΔX (s)
[0012]
Next, some examples of the load measuring means 5 will be described below.
For this, for example, as shown in FIG. 1, a force sensor provided at the base of the grasping portion 6 that supports the lower limb 7 is used as a means for measuring the load on the apparatus from the limb of the patient. The load on the device can be measured.
Although not shown, as another means, each drive shaft 3A, 3B, 3C of the mechanism unit 1 in FIG. 1 is provided with a torque sensor, and from the position information of the gripping part 6, the lower limb 7 is applied to the gripping part 6. The load can be measured.
As another means, the load on the device from the patient's limb can be measured by using the current value of the motor provided on each axis of the arm shown in FIG.
Although not shown, as another means, a so-called disturbance observer using the current value of each motor and the speed information of each motor provided in the arm shown in FIG. Can be measured.
Although not shown, as another means, the device is applied from the patient's limb to the apparatus from the myoelectric information of the main muscles of the driven limb shown in FIG. 1 and the angle information of each joint of the limb at that time. The load can be measured.
Although not shown in the figure, as another means, the size of the bulge of the main muscles of the driven limb shown in FIG. 1 and the angle information of each joint of the limb at that time are measured, whereby the patient It is possible to measure the load on the device from the limbs.
Further, as another means, the load from the patient's limb to the apparatus can be measured more precisely by combining the above measuring means. This can be realized, for example, by adding a weight to each output.
Here, the load information detected by the load measuring means includes a load generated arbitrarily by the patient's limb, a load due to the weight of the limb, and a load generated in the interaction between the limb and the device, A value measured by a measuring means as so-called vector information having a magnitude is used as load information.
Further, when the position deviation according to the impedance control coefficient different from the driving means is converted from the coordinate system on the teaching trajectory to the position deviation of the work coordinate system, the ratio of subtracting the own weight can be adjusted.
For example, the weight compensation amount of the limb can be set in advance, and can be adjusted by a dial (not shown), and the load on the patient can be adjusted according to the purpose of treatment. For example, if this is compensated 100%, the patient does not need to support the leg by his / her own weight, and most of the generated muscular strength can be used as the force to move the leg, and the load on the patient is reduced. This applies to patients immediately after onset. On the other hand, if the compensation is set to 0%, the patient needs to support his / her own weight, and the burden on the patient increases. This applies to patients whose training is progressing and restoring function. It is important that this intermediate value can be adjusted according to the patient's recovery level.
[0013]
Next, the trajectory setting unit 10 will be described. Here, the impedance control is executed, the spring coefficient of the impedance is further set low, the lower limb 1 is fixed to the grasping portion 6 and can be moved manually, and the mechanism portion 1 is directly moved to teach the trajectory. Teach is performed, and the trajectory at that time is stored in the trajectory setting unit 10 from the mechanism unit 1 through the taught trajectory information . The apparatus of the present invention is generally intended for the treatment of joint tissues or muscles or ligaments or other joint tissues, and is used for the treatment of tissues such as hip and knee joints and muscles as shown in the embodiments. It is not limited to those intended for treatment, and can be easily applied to joints and muscles of each part of the limb. In the above embodiment, an electric motor is assumed as the limb body driving means. However, the present invention is not limited to this, and the same effect can be obtained by using a hydraulic servo driving means or a pneumatic servo driving means.
[0014]
【The invention's effect】
As described above, according to the control method of the present invention, there is an effect that, in so-called exercise therapy, exercise training accompanied by generation of weak muscle force of a patient, that is, automatic assistance exercise can be performed. In addition, in the conventional technology, it is possible to exercise the range of motion of the joint, but training that can be applied only to patients whose muscle strength has recovered to some extent, such as only passive movement, isometric exercise, isotonic exercise, constant velocity exercise, etc. There is an effect that it is possible to solve the problem that was not possible.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a configuration of a limb body driving device to which a method of the present invention is applied.
FIG. 2 is a flowchart showing the main procedure of the method of the present invention.
FIG. 3 is a flowchart showing a part of the procedure of the method of the present invention.
FIG. 4 is a flowchart showing a part of the procedure of the method of the present invention.
FIG. 5 is a flowchart showing a part of the procedure of the method of the present invention.
[Sharpness of sign]
DESCRIPTION OF SYMBOLS 1 Mechanism part 2 Base part 3A, 3B, 3C Drive shaft 4A, 4B Link 5 Load measuring means 6 Gripping part

Claims (6)

A mechanism that moves the patient's limb while supporting the limb with a gripping part;
A load detecting means provided on the grip portion for detecting a load applied to the limb;
A trajectory setting means for storing a periodic motion pattern of the grip portion taught in advance and outputting it as trajectory information;
A driving unit that corrects the trajectory using the trajectory information output from the trajectory setting unit and load information from the load detection unit and outputs the corrected trajectory to the mechanism unit to control the joint of the patient In the control method of the limb body drive device for performing exercise,
The drive unit outputs a time series number assigned along the time series of the teaching trajectory to the trajectory setting means, reads a point on the teaching trajectory according to the time series number from the trajectory setting means,
The distance r to be moved from the point corresponding to the current time series number is compared with the distance from the current teaching point to the next teaching point, and if the distance r is smaller than the distance between the teaching points, the gripping Update the target position of the
A drive command for the target position calculated by an internal impedance model is output to the mechanism unit,
The mechanism unit performs a procedure of driving the gripping unit by servo control every control cycle,
When calculating the distance r, the inclination of the vector t from the point corresponding to the current time series number to the point on the teaching trajectory corresponding to the next time series number is defined as the inclination of the teaching trajectory. (1) When the inclination of the vector t with respect to the work coordinate system is other than vertical, a position deviation is calculated from the load information using a virtual impedance model in the trajectory setting means, and the current time series number is calculated. The projection of the vector from the corresponding point on the teaching trajectory to the deviation position onto the vector t is used as the distance r, and (2) a predetermined value when the inclination of the vector t with respect to the work coordinate system is vertical. And the distance r,
When calculating the position deviation in the virtual impedance dance model in the trajectory setting means, the load information is converted from the coordinate system based on the load detection means to the work coordinate system, and the converted load information The limb body generated force vector is obtained by subtracting the weight of the limb from the predetermined ratio from the limb body generated force vector, the limb body generated force vector is transformed into a coordinate system based on the teaching trajectory, and the position deviation is performed according to the virtual impedance model on the teaching trajectory. And calculating the position deviation from the coordinate system on the teaching trajectory to the position deviation of the work coordinate system . The limb body drive device control method according to claim 1 , wherein, when the target position is updated, a restriction is provided on either the number of updates of the time-series number or the size of the distance r.   The virtual impedance dance model in the trajectory setting means is a second-order dynamic model based on virtual inertia, virtual viscosity, or virtual rigidity, or a first-order dynamic model based on virtual viscosity / virtual rigidity. The control method of the limb body drive device of Claim 1 or 2.   The limb body drive device control method according to any one of claims 1 to 3, wherein at least one of the time series numbers is updated when the target position is updated.   The load information includes a load arbitrarily generated from the limb of the patient, a load due to the weight of the limb, and a load generated in the interaction between the limb and the device by measuring means as vector information having a direction and a magnitude. 2. The control method for a limb body driving device according to claim 1, wherein the control method is a measured value.   The method for controlling a limb body driving device according to claim 1, wherein a ratio of subtracting the weight of the limb body can be adjusted when obtaining the limb body generation force vector.

Images