JP5392671B2 - Walking measurement device - Google Patents

Description

  The present invention relates to a gait measuring device, and more particularly to a gait measuring device for grasping the movement of a lower limb accompanying pedestrian walking.

  Appropriate walking evaluation is required to grasp the decline in walking function caused by aging and to effectively perform training for preventing such function deterioration. In addition, when a walking disorder occurs due to aging or illness, it is necessary to perform an appropriate rehabilitation after performing an appropriate walking evaluation. The above gait evaluation is performed by a doctor or a physical therapist visually observing the subject (patient), but it depends largely on the sense of each doctor or the physical therapist, and the evaluation result may vary depending on the viewer. .

By the way, Patent Document 1 discloses a walking analysis system capable of measuring the movement of a foot of a pedestrian serving as a subject. This gait analysis system includes a sensor that is worn on a pedestrian's foot and measures the acceleration data and angular velocity data of the foot, and outputs wirelessly, an access point that receives data from the sensor, and the access point The above-mentioned data is input through a network, and a walking analysis device for obtaining an analysis item for rehabilitation such as a walking speed of a pedestrian from the data.
JP 2008-161228 A

  However, in the gait analysis system, as the analysis items for rehabilitation, only the stride length, step length, walking speed, etc. of the foot are obtained, so the movement of the entire limb of the pedestrian is quantitatively obtained. However, there is a problem that the evaluation of the movement of the entire lower limbs still depends on the visual observation of a doctor or a physical therapist. In the gait analysis system, the data from the sensor is transmitted to the gait analysis device through the network, so the data from the sensor can be used only in the area where data can be transmitted and received, and can be used anywhere. There is also a problem that it cannot be done. Furthermore, in the gait analysis system, if the sensor is attached not only to the foot part but also to a plurality of parts of the leg part, the amount of data communication increases, so another function for performing data communication efficiently and instantaneously is provided. There is a problem that the system configuration becomes large.

  The present invention has been devised by paying attention to such a problem, and its purpose is to quantitatively determine the movement of the lower limbs of a pedestrian regardless of the installation location with a simple device configuration. The object is to provide a walking measurement device.

(1) In order to achieve the above object, the present invention is a gait measuring device that measures the movement of a limb of a pedestrian while moving with the pedestrian,
A position and orientation sensor that measures the three-dimensional position and posture of the pedestrian's thigh and toe, and a plurality of feature points set in the lower limb to specify the state of the lower limb at predetermined time intervals A feature point calculation means to be obtained,
The feature point calculation means employs a configuration in which the position of each feature point is calculated at each predetermined time by substituting the measurement value of the position and orientation sensor into an arithmetic expression based on a predetermined lower limb model. Yes.

(2) Moreover, this invention is a walk measurement apparatus which measures the motion of the said leg of the said pedestrian, following a pedestrian automatically,
From a moving body that autonomously moves the entire apparatus by a driving device, a position and orientation sensor that measures the three-dimensional position and orientation of the thigh and toe of the pedestrian, and the measured value of the position and orientation sensor, the pedestrian Movement control means for controlling the operation of the driving device so as to maintain the relative positional relationship in a preset constant state, and a plurality of feature point positions set in the lower limb to specify the state of the lower limb. And feature point calculation means for obtaining every predetermined time,
The feature point calculation means adopts a configuration in which the position of each feature point is calculated at each predetermined time by substituting the measurement value of the position and orientation sensor into an arithmetic expression based on a predetermined lower limb model. Is preferred.

  According to the present invention, since the positions of a plurality of feature points necessary for specifying the state of the lower limb are measured at regular time intervals, the position of each obtained feature point is determined at each measurement time. By connecting in order, the state of the lower limb for each hour can be obtained, and the movement of the lower limb can be quantitatively obtained by comparing the lower limb state in time series. Further, in the walking measurement apparatus of the present invention, since the movement of the lower limbs of the pedestrian is measured while moving with the pedestrian, a communication environment such as network communication such as the Internet is not required, and the simple apparatus configuration can Even in a place where the communication environment is not established, the movement of the lower limbs of the pedestrian can be obtained. Furthermore, it is not necessary to attach sensors to all of the plurality of feature points required to specify the state of the lower limbs, and by sensing the position and orientation of at least two locations of the thigh and toe, the position of each feature point Can be requested. For this reason, not only complicated exchange of data becomes unnecessary, but it is not necessary to attach a sensor to the joint portion mentioned as the feature point, so that the walking burden on the pedestrian by the sensor can be reduced. In addition, since the sensor only needs to be attached to the thigh and the toe portion where the sensor is easily fixed, it is possible to prevent the sensor from being accidentally dropped during walking. As a result, lower limb data can be reliably acquired with a simple device configuration.

  In particular, by configuring as described in (2) above, it is possible to obtain the movement of the lower limbs of the pedestrian while automatically following the pedestrian, so that the apparatus constituting the body of the position and orientation sensor, the feature point calculation means, etc. Even if the device is large and the device cannot be attached to a part of the pedestrian's body, there is no need for the pedestrian or other person to artificially move the device along with the pedestrian's walk. The movement of the lower limb can be obtained without taking time and effort. In addition, since the automatic tracking control of the moving body is also performed using the measurement value of the position and orientation sensor for measuring the walking motion of the pedestrian, a new sensor or the like for the automatic tracking control is provided to the pedestrian. There is no need to attach it, and the increase in size and complexity of the device configuration associated with the automatic tracking control can be suppressed.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  FIG. 1 shows a schematic configuration diagram of a mobile walking measurement apparatus according to the present embodiment. In this figure, the walking measurement apparatus 10 includes a moving body 11 that moves the entire apparatus, and a position and orientation sensor 13 and a measurement control device 14 that are mounted on the moving body 11.

  The mobile body 11 moves autonomously so that it can automatically follow behind the pedestrian H with the walking of the pedestrian H as a subject by the operation control described later by the measurement control device 14. The moving body 11 includes a main body 16 on which the position / orientation sensor 13 and the measurement control device 14 are disposed, casters 17 and 17 rotatably supported at two front left and right positions of the bottom of the main body 16, and a bottom of the main body 16. Drive wheels 18 and 18 rotatably supported at two positions on the left and right sides of the rear side, motors 20 as drive devices for independently driving the drive wheels 18 and 18 on both the left and right sides, and the drive wheels 18 on both the left and right sides, And an encoder 21 for measuring 18 rotation angles.

  The position / orientation sensor 13 is a sensor capable of measuring the three-dimensional position / orientation of a predetermined point existing on the lower limb K of the pedestrian H. In the present embodiment, a known magnetic sensor is used. The position / orientation sensor 13 includes a transmitter 22 that is fixed to the main body 16 and generates an alternating magnetic field, a receiver 23 (marker) that includes a coil that generates current according to the alternating magnetic field generated around the main body 16 by the transmitter 22, The sensor body 25 is fixed to the main body 16 and specifies the three-dimensional position / orientation of the receiver 23 according to the current from the receiver 23.

  The receiver 23 is mounted on the pedestrian H and includes a mounting receiver 27 for measuring the position and posture of the mounting site, and an initial setting receiver 28 used for a prior setting. The mounting receiver 27 is detachably mounted on the left and right thighs K1 and left and right toe portions K4 of the pedestrian H with a mounting tool 30 such as a belt. As will be described later, the initial setting receiver 28 measures the positions and postures of a plurality of feature points A to E set on the lower limbs K of the pedestrian H in a stationary state before the pedestrian H starts walking. Used when. Each of the receivers 27 and 28 is configured to transmit a current generated by the receiver 27 and 28 to the sensor body 25 through the cable 32. In the following description, the attachment point of the attachment receiver 27 attached to the thigh K1 is referred to as “thigh measurement point P1”, and the attachment point of the attachment receiver 27 attached to the toe portion K4 is “ It is referred to as “toe part measurement point P2”. The position and orientation detected by the mounting receiver 27 and the initial setting receiver 28 described above are the position and orientation in the coordinate system of the position and orientation sensor 13 with the predetermined position of the transmitter 22 attached to the moving body 11 as the origin G1. Hereinafter, the coordinate system is referred to as a “moving body coordinate system”. Here, the moving body coordinate system has three orthogonal axes (x axis, y axis, z axis), and hereinafter, unless otherwise specified, the coordinates in the moving body coordinate system are represented as (x, y, z), and the z axis. A rotation angle A around, a rotation angle E around the y-axis, and a rotation angle R around the x-axis, and the posture in the moving object coordinate system is represented as (A, E, R).

  The position / orientation sensor 13 used here is a known sensor, and since its configuration and structure are not the essence of the present invention, detailed description thereof will be omitted. The position / orientation sensor 13 may be any sensor that can measure the three-dimensional position / orientation of the pedestrian's thigh K1 and toe part K4. It is also possible to use it. However, even if the magnetic sensor is used outdoors where the measurement environment changes, such as the presence of a shielding object, the three-dimensional positions and postures of the thigh K1 and toe K4 can be more accurately determined. Can be measured.

  The measurement control device 14 includes a computer including a CPU board, a serial board, a counter board, a DA board, a motor driver, a memory, and the like, and includes a plurality of program modules and / or processing circuits.

  This measurement control device 14 is preliminarily assumed from the movement control means 34 for controlling the operation of the motor 20 so as to maintain the relative positional relationship with the pedestrian H in a preset constant state based on the measurement values of the position and orientation sensor 13. And a feature point calculating means 35 for calculating the positions of the feature points A to E by substituting the measured values of the position / orientation sensor 13 detected every predetermined time into an arithmetic expression based on the lower limb model.

The movement control means 34 controls the operation of the motor 20 as follows. First, at the initial time when the pedestrian H starts walking (FIG. 2 (A)), the sensor body 25 attached to the moving body 11 located behind the pedestrian H causes the left and right sides in the moving body coordinate system. The positions of the thigh measurement points P1, P1 are detected, and the positions of the midpoints of these measurement points P1, P1 are calculated. The position of the midpoint is set as an origin G2 of a coordinate system based on the pedestrian H (hereinafter referred to as “pedestrian coordinate system”). Then, a position vector ^Hp extending from the origin G2 of the pedestrian coordinate system to the origin G1 of the moving object coordinate system is obtained. The position vector ^H p is stored so as to be constant during movement. The position vector ^H p may be stored in advance as a constant value.

Then, after a predetermined time, when the pedestrian H is moved to (FIG. 2 (B)), it points to a certain of the position vector ^{H p} extending from the origin G2 pedestrian coordinate system reaches, momentarily, the mobile coordinate The point deviates from the origin G1 of the system, and this point becomes the movement target point M of the mobile body 11. At this time, the sensor body 25 detects the positions and postures of the thigh measurement points P1 and P1 on both the left and right sides in the moving body coordinate system. A plane position in the coordinate system is calculated, and a rotation angle around the z-axis that is the vertical direction in the moving object coordinate system (the orthogonal direction in the drawing) is calculated.

ここで、位置ベクトル^Ｈｐは、前述したように、移動体１１の静止状態で予め記憶された一定値であり、変換マトリクス^ＥＴ_Ｈは、前述のように算出された原点Ｇ２の平面位置（ｘ，ｙ）及びｚ軸回りの回転角Ａから、次式で求められる。

Then, the position vector ^{E p} from the origin G1 of the moving body coordinate system to the movement target point M is calculated using the following expression.

Here, as described above, the position vector ^H p is a constant value stored in advance when the moving body 11 is stationary, and the transformation matrix ^E T _H is the plane position of the origin G2 calculated as described above ( x, y) and the rotation angle A around the z axis are obtained by the following equation.

なお、以上において、ＫとＫ_ｓは、実験により算出された比例定数である。
更に、平均角速度ωから差分速度ω_ｓを減算することで、左側の駆動輪１８の角速度ω_１が決定される一方、平均角速度ωに差分速度ω_ｓを加算することで、右側の駆動輪１８の角速度ω_２が決定される。 Then, an angle θ formed by the magnitude of the position vector ^Ep obtained by the above equation and the y axis (vertical direction in FIG. 2B) of the moving object coordinate system is obtained by an inner product, and the left and right drive wheels are obtained by the following equation: The average angular speed ω of 18 and 18 and the differential speed ω _{s of the} left and right drive wheels 18 and 18 necessary for the left and right rotation (steering) of the moving body 11 are calculated.

In the above, K and K _s is a proportionality constant which is calculated by experiments.
Further, by subtracting the differential speed ω _s from the average angular speed ω, the angular speed ω ₁ of the left drive wheel 18 is determined, while by adding the differential speed ω _s to the average angular speed ω, the right drive wheel 18. of the angular velocity ω ₂ is determined.

The drive of the motor 20 is controlled so that the drive wheels 18 and 18 rotate at the angular velocities ω ₁ and ω ₂ obtained as described above. As a result, the moving body 11 moves autonomously to reach the movement target point M, automatically follows the walking of the pedestrian H, and moves to a fixed position behind the pedestrian H. In addition, this movement control is performed for every fixed time so that the followable | trackability of the mobile body 11 is not impaired.

  In the feature point calculation means 35, the positions (coordinates) of the feature points A to E during walking of the pedestrian H are obtained as follows. These feature points A to E are set at five locations on the left and right lower limbs K, which are essential for specifying the state of the lower limbs K. A simulated leg line that matches the condition is obtained. Here, as shown in FIG. 1, the feature point A is the greater trochanter which is one part of the femur, and the feature point B is the outer femoral condyle (outer knee point) near the knee joint center. Yes, the feature point C is an ankle portion (saddle point) near the ankle joint, and the feature point D is a portion (fifth metatarsal head point) that projects outwardly in the foot portion near the metatarsal joint. ), And the feature point E is a portion (first distal phalanx tip) that protrudes most forward in the foot.

  As the lower limb model, as shown in FIG. 3, one lower limb K is considered to be composed of four rigid bodies including a thigh K1, a crus K2, a hind leg K3, and a toe K4. Therefore, it is assumed that the feature point A and the feature point B exist on the rigid body of the thigh K1, and the feature point D and the feature point E exist on the rigid body of the toe portion K4. Further, the thigh K1 and the crus K2 are connected by a three-degree-of-freedom knee joint S1 that enables relative rotation about three orthogonal axes, and the crus K2 and the rear foot K3 are connected. The middle foot with one degree of freedom that is connected by the foot joint S2 having the same degree of freedom and that allows the toe portion K4 to rotate only in the vertical direction with respect to the rear foot portion K3 between the rear foot portion K3 and the toe portion K4. It is considered that it is composed of a bone joint S3.

Therefore, in the feature point calculation means 35, first, a rigid body position vector ^th p extending from the thigh measurement point P1 to the feature point A and the feature point B in the local coordinate system with the thigh measurement point P1 as the origin, and Then, in the local coordinate system with the toe portion measurement point P2 as the origin, a rigid body position vector ^th p extending from the toe portion measurement point P2 to the feature point D and the feature point E is calculated. These rigid body position vectors ^th p have a relative position relationship between the thigh measurement point P1 and the feature point A and the feature point B, and a relative position relationship between the toe portion measurement point P2 and the feature point D and the feature point E. Since the inside does not change, even if the pedestrian H is walking, it is always constant. Further, when the pedestrian H is walking, the mounting receiver 27 is mounted on the thigh K1 and the toe K2, and the three-dimensional position / posture of the thigh measurement point P1 and the toe measurement point P2 is measured. Therefore, the coordinates of the feature points A, B, D, E during walking are calculated from these measured values and the constant position vector ^th p in the rigid body, and the remaining features during walking are calculated from the coordinates. The coordinates of the point C are calculated. The calculation procedure will be described in detail below using the flowchart of FIG.

First, in a stationary state before the pedestrian H starts walking (hereinafter referred to as “pre-measurement start state”), each rigid body position vector ^th p is obtained as follows (step S101).

ここで、^ｒｅｐは、初期測定用レシーバ２８の所定部位を原点としたローカル座標系において、当該原点から棒状体３７の先端に延びる位置ベクトルであり、予め設定された一定値となっている。また、^ｔｒＴ_ｒｅは、初期測定用レシーバ２８からの測定に基づく座標変換マトリクスであり、特徴点Ａ〜Ｅそれぞれに棒状体３７の先端が接触したときの各初期測定用レシーバ２８の三次元位置（ｘ，ｙ，ｚ）及び姿勢（Ａ，Ｅ，Ｒ）から、各特徴点Ａ〜Ｅそれぞれの場合について次式で求められる。

As shown in FIG. 1, a rod-shaped body 37 is attached to the initial measurement receiver 28 used at this time, and a doctor, a physical therapist, or the like can connect the left and right sides of the pedestrian H to be measured. The site corresponding to the feature points A to E in the lower limb K is identified by visual observation or the like, and the tip of the rod-shaped body 37 is brought into contact with each site, so that the sensor body 25 of the moving body 11 performs initial measurement at each contact. The three-dimensional position / orientation of the receiver 28 in the moving body coordinate system is detected. Then, the initial position vector ^tr p of the moving body coordinate system extending from the origin G1 of the moving body coordinate system to each of the feature points A to E is obtained by the following equation.

Here, ^rep is a position vector extending from the origin to the tip of the rod-shaped body 37 in the local coordinate system with the predetermined part of the initial measurement receiver 28 as the origin, and is a predetermined constant value. ^Tr _Tre is a coordinate transformation matrix based on the measurement from the initial measurement receiver 28, and the three-dimensional position of each initial measurement receiver 28 when the tip of the rod-shaped body 37 contacts each of the feature points A to E. From each of (x, y, z) and posture (A, E, R), the respective characteristic points A to E are obtained by the following equations.

ここで、^ｔｒＴ_ｔｈは、装着用レシーバ２３からの測定に基づく座標変換マトリクスであり、前述した座標変換マトリクス^ｔｒＴ_ｒｅの式と同一の上式（５）で求められる。すなわち、特徴点Ａへの剛体内位置ベクトル^ｔｈｐは、大腿部Ｋ１に装着された装着用レシーバ２７で検出された三次元の位置（ｘ，ｙ，ｚ）及び姿勢（Ａ、Ｅ、Ｒ）を上式（５）に代入して座標変換マトリクス^ｔｒＴ_ｔｈを特定し、特徴点Ａの初期位置ベクトル^ｔｒｐを上式（６）に代入することで求められる。また、特徴点Ｂへの剛体内位置ベクトル^ｔｈｐも、特徴点Ａと同様に求められる。一方、特徴点Ｄへの剛体内位置ベクトル^ｔｈｐは、爪先部Ｋ４に装着された装着用レシーバ２７で検出された三次元の位置（ｘ，ｙ，ｚ）及び姿勢（Ａ、Ｅ、Ｒ）と、特徴点Ｄの初期位置ベクトル^ｔｒｐとにより、特徴点Ａ、Ｂの場合と同様に求められる。また、特徴点Ｅへの剛体内位置ベクトル^ｔｈｐも、特徴点Ｄと同様に求められる。 Next, the position (x, y, z) and posture (A, E, R) of the thigh measurement point P1 and the toe part measurement point P2 are set by the wearing receiver 23 while the pedestrian H remains stationary. Using the position (x, y, z) and posture (A, E, R) respectively detected from the initial position vectors ^tr p of the feature points A, B, D, E obtained previously, Each rigid body position vector ^th p is obtained.

Here, ^tr T _th is a coordinate conversion matrix based on the measurement from the mounting receiver 23, and is obtained by the same equation (5) as the equation of the coordinate conversion matrix ^tr T _re described above. That is, the rigid body position vector ^th p to the feature point A is the three-dimensional position (x, y, z) and posture (A, E, R) detected by the mounting receiver 27 mounted on the thigh K1. ) Is substituted into the above equation (5) to specify the coordinate transformation matrix ^tr T _th and the initial position vector ^tr p of the feature point A is substituted into the above equation (6). The rigid body position vector ^th p to the feature point B is also obtained in the same manner as the feature point A. On the other hand, the rigid body position vector ^th p to the feature point D is the three-dimensional position (x, y, z) and posture (A, E, R) detected by the mounting receiver 27 mounted on the toe portion K4. And the initial position vector ^tr p of the feature point D, as in the case of the feature points A and B. Also, the rigid body position vector ^th p to the feature point E is obtained in the same manner as the feature point D.

すなわち、ここでは、特徴点Ａへの移動時位置ベクトル^ｔｒｐ（以下、「移動時位置ベクトル^ｔｒａ」と称する。）を求める際には、対応する大腿部Ｋ１に装着された装着用レシーバ２７で検出された三次元位置（ｘ，ｙ，ｚ）及び姿勢（Ａ、Ｅ、Ｒ）を上式（５）に代入し、座標変換マトリクス^ｔｒＴ_ｔｈを特定し、特徴点Ａへの剛体内位置ベクトル^ｔｈｐを上式（７）に代入することで求められる。また、特徴点Ｂへの移動時位置ベクトル^ｔｒｐ（以下、「移動時位置ベクトル^ｔｒｂ」と称する。）も、特徴点Ａと同様に求められる。一方、特徴点Ｄへの移動時位置ベクトル^ｔｒｐ（以下、「移動時位置ベクトル^ｔｒｄ」と称する。）を求める際には、対応する爪先部Ｋ４に装着された装着用レシーバ２７で検出された三次元位置（ｘ，ｙ，ｚ）及び姿勢（Ａ、Ｅ、Ｒ）と、特徴点Ｄへの剛体内位置ベクトル^ｔｈｐとにより、特徴点Ａ、Ｂの場合と同様に求められる。また、特徴点Ｅへの移動時位置ベクトル^ｔｒｐ（以下、「移動時位置ベクトル^ｔｒｅ」と称する。）も、特徴点Ｄと同様に求められる。 Thereafter, the pedestrian H starts moving, and the position (x, y, z) and posture (A, E, and) of the thigh measurement point P1 and the toe measurement point P2 are determined by the mounting receiver 27 at regular intervals. R) is detected respectively. At this time, using the detected position (x, y, z) and posture (A, E, R) and each rigid body position vector ^th p obtained previously, A moving position vector ^tr p extending from the origin G1 of the moving object coordinate system to the feature points A, B, D, and E is obtained at regular intervals (step S102).

That is, here, when obtaining a moving position vector ^tr p to the feature point A (hereinafter referred to as “moving position vector ^tr a”), a mounting receiver mounted on the corresponding thigh K1. The three-dimensional position (x, y, z) and posture (A, E, R) detected in step 27 are substituted into the above equation (5), the coordinate transformation matrix ^tr T _th is specified, and the _stiffness to the feature point A is determined. It is obtained by substituting the in-body position vector ^th p into the above equation (7). Further, a position vector ^tr p during movement to the feature point B (hereinafter referred to as “position vector ^tr b during movement”) is also obtained in the same manner as the feature point A. On the other hand, when obtaining a moving position vector ^tr p (hereinafter referred to as “moving position vector ^tr d”) to the feature point D, it is detected by the mounting receiver 27 mounted on the corresponding toe portion K4. Further, the three-dimensional position (x, y, z) and posture (A, E, R) and the rigid body position vector ^th p to the feature point D are obtained in the same manner as in the case of the feature points A and B. In addition, the moving position vector ^tr p (hereinafter referred to as “moving position vector ^tr e”) to the feature point E is also obtained in the same manner as the feature point D.

Next, a moving position vector ^tr c extending from the origin G1 of the moving object coordinate system to the feature point C is calculated as follows using the moving position vectors ^tr b, ^tr d, and ^tr e (step S103). .

At this time, as shown in FIG. 5, a metatarsal joint coordinate system serving as a local coordinate system is defined for the metatarsal joint S3 of the lower limb model. This metatarsal joint coordinate system includes an X axis, a Y axis, and a Z axis that are orthogonal to each other with the feature point D as the origin G3. Here, a straight line passing through the characteristic point E from the origin G3 X 'as an axis, a straight line passing through the fixed point D ₁ set in advance from the origin G3 and Y-axis, X' defines the Z-axis by the cross product between the axis and Y-axis Then, the X axis is defined by the outer product of the Z axis and the Y axis. Fixing point D ₁ is a point which corresponds in the first metatarsal head point, even pedestrians walking in H, at a preset fixed point so that the relative positional relationship between the feature point D becomes constant, characteristic point D It is assumed to move with. Therefore, the Y axis is an axis along the axial direction of the metatarsal joint, that is, the lateral direction of the foot.

First, the distance l _S between the feature points BC and the distance l _h between the feature points CD are calculated from the initial position vectors ^tr p of the feature points B, C, and D in the pre-measurement state obtained as described above. Stored in memory. These distances l _S and l _h are always constant during walking due to the structure of the lower limb model described above. Further, based on the movement during position vector ^{tr d} to the feature point D determined previously, when moving the position vector ^tr d ₁ is obtained which extends from the origin G1 of the moving body coordinate system fixed point D _1.

ここで、θは、中足骨関節座標系のＺ軸とローカル位置ベクトル^ｍｅｃのなす角度であり、θは、次式で求められる。なお、以下では、θの解が二つ得られる場合があるが、足首の構造から小さい方の値が選択される。

上式（９）では、中足骨関節座標系における特徴点Ｂの座標（Ｂ_ｘ，Ｂ_ｙ，Ｂ_ｚ）が代入されるが、当該座標（Ｂ_ｘ，Ｂ_ｙ，Ｂ_ｚ）は、先に求めた移動体座標系における特徴点Ｂへの移動時位置ベクトル^ｔｒｂが、次式により中足骨関節座標系における特徴点Ｂへのローカル位置ベクトル^ｍｅｂに変換されることで求められる。

ここで、^ｔｒＴ_ｍｅは、中足骨関節座標系における位置ベクトルを移動体座標系における位置ベクトルに変換する座標変換マトリクスであり、移動体座標系における中足骨関節座標系のＸ軸基底ベクトルをｉ、同Ｙ軸基底ベクトルをｊ、同Ｚ軸基底ベクトルをｋ、同Ｘ’軸基底ベクトルをｉ’、移動体座標系から中足骨関節座標系への平行移動マトリクスをｔとしたときに、次式の通りとなる。

Next, a local position vector ^me c extending from the origin G3 of the metatarsal joint coordinate system to the feature point C is as follows.

Here, θ is an angle formed by the Z-axis of the metatarsal joint coordinate system and the local position vector ^me c, and θ is obtained by the following equation. In the following, two solutions of θ may be obtained, but the smaller value is selected from the ankle structure.

In the above equation (9), the coordinates (B _x , B _y , B _z ) of the feature point B in the metatarsal joint coordinate system are substituted, but the coordinates (B _x , B _y , B _z ) The position vector ^tr b at the time of moving to the feature point B in the moving body coordinate system is converted to the local position vector ^me b to the feature point B in the metatarsal joint coordinate system by the following equation.

Here, ^tr T _me is a coordinate conversion matrix for converting a position vector in the metatarsal joint coordinate system into a position vector in the moving body coordinate system, and an X-axis base vector of the metatarsal joint coordinate system in the moving body coordinate system Is i, Y-axis basis vector is j, Z-axis basis vector is k, X'-axis basis vector is i ', and the translation matrix from the moving body coordinate system to the metatarsal joint coordinate system is t. In addition, the following equation is obtained.

Then, the position vector ^tr c at the time of movement of the feature point B is calculated from the local position vector ^me c obtained above by the following equation using the transformation matrix ^tr T _me according to the above equation (11).

ここで、^ＷＴ_ｔｒは、移動体座標系の位置ベクトルをワールド座標系の位置ベクトルに変換するための変換マトリクスであり、次のように求められる。

上式（１４）のｘ_ｔ、ｙ_ｔは、歩行者Ｈが歩行を開始してからの時間ｔにおける原点Ｇ１のワールド座標系における平面位置を表す。また、θ_ｔは、前記時間ｔにおける原点Ｇ１のワールド座標系のｚ軸回りの移動体１１の回転角度を表す。これら平面位置（ｘ_ｔ、ｙ_ｔ）と回転角度θ_ｔは、前記エンコーダ２１の測定値に基づき、次式で求められる。ここでは、エンコーダ２１の測定値から求めた左右の各駆動輪１８，１８の角速度をω_ｌ、ω_ｒとし、各駆動輪１８，１８の間隔を予め記憶された一定値ｌとし、各駆動輪１８，１８の半径を予め記憶された一定値ｒとしている。

Finally, each moving position vector ^tr p ( ^tr a, ^tr b, ^tr c, ^tr d, ^tr e) of the moving object coordinate system is obtained by the following expression, and the moving object 11 before the moving object 11 starts moving Is converted into a position vector ^W p at the time of movement in the world coordinate system with the predetermined portion as the origin (step S104).

Here, ^W T _tr is a conversion matrix for converting the position vector of the moving object coordinate system into the position vector of the world coordinate system, and is obtained as follows.

In the above equation (14), x _t and y _t represent plane positions in the world coordinate system of the origin G1 at time t after the pedestrian H starts walking. Θ _t represents the rotation angle of the moving body 11 around the z axis of the world coordinate system of the origin G1 at the time t. These planar positions (x _t , y _t ) and the rotation angle θ _t are obtained from the following formula based on the measured values of the encoder 21. Here, the angular velocities of the left and right drive wheels 18, 18 obtained from the measured values of the encoder 21 are ω _l , ω _r, and the interval between the drive wheels 18, 18 is a predetermined stored value l, and each drive wheel The radii of 18 and 18 are set to a constant value r stored in advance.

Here, the plane positions x _t and y _t and the rotation angle θ _t are obtained by calculation based on the measurement values of the encoder 21, but the present invention is not limited to this, and the measurement values are obtained using GPS. It can also be applied as it is.

As described above, when moving the position vector ^{W p} of the transformed feature point A~E the world coordinate system are obtained, respectively at predetermined intervals, the from the movement time of the position vector ^{W p,} the feature point in the world coordinate system A Each of the three-dimensional coordinates of ~ E is obtained. The three-dimensional coordinates of these feature points A to E are input to analysis software connecting the coordinates with straight lines, so that a simulated leg line representing the state of the lower limb K of the pedestrian H is formed for each measurement time, The movement of the lower limbs of the pedestrian H can be grasped quantitatively by displaying the simulated leg line in a time series on a monitor or the like (not shown).

  The moving body 11 is moved by the rotation of the wheels including the casters 17 and the drive wheels 18, but may be replaced with a moving structure such as a caterpillar or a leg. Moreover, as a drive device, as long as a drive force can be given to the main body 16, you may substitute for another actuator.

  Moreover, in the said embodiment, although the said mobile body 11 has automatically followed the back of the pedestrian H, this invention is not limited to this, The mobile body 11 is made to automatically follow the side and the front of the pedestrian H. It may be.

  Further, the movement control means 34 may be omitted, and the moving body 11 may be moved manually or automatically by the operation of the pedestrian H or another person. However, the operation of the mobile body 11 is not required by performing as in the above embodiment, and the movement of the lower limb K of the pedestrian H can be easily measured.

  In addition, when the pedestrian H has a size and weight that can be carried or worn by the pedestrian H, the device including the position and orientation sensor 13 and the feature point calculation unit 34 is not mounted on the moving body 11 by being held by the pedestrian H. In this case, the moving body 11 and the movement control means 34 can be omitted.

  In addition, the configuration of each part of the apparatus in the present invention is not limited to the illustrated configuration example, and various modifications are possible as long as substantially the same operation is achieved.

1 is a schematic system configuration diagram of a mobile walking measurement device according to the present embodiment. The conceptual top view of the moving body for demonstrating control by a movement control means. The conceptual diagram for demonstrating a leg model. The flowchart showing the process sequence in a feature point calculation means. The figure for demonstrating the process using a metatarsal joint coordinate system.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Mobile walking control apparatus 11 Mobile body 12 Position and orientation sensor 20 Motor (drive device)
34 Movement control means 35 Feature point calculation means

Claims (2)

A gait measuring device that measures the movement of the lower limbs of a pedestrian while moving with the pedestrian,
A position and orientation sensor that measures the three-dimensional position and posture of the pedestrian's thigh and toe, and a plurality of feature points set in the lower limb to specify the state of the lower limb at predetermined time intervals A feature point calculation means to be obtained,
The feature point calculating means calculates the position of each feature point at each predetermined time by substituting the measured value of the position and orientation sensor into an arithmetic expression based on a predetermined lower limb model. Measuring device. A gait measuring device that measures the movement of the lower limb of the pedestrian while automatically following the pedestrian,
From a moving body that autonomously moves the entire apparatus by a driving device, a position and orientation sensor that measures the three-dimensional position and orientation of the thigh and toe of the pedestrian, and the measured value of the position and orientation sensor, the pedestrian Movement control means for controlling the operation of the driving device so as to maintain the relative positional relationship in a preset constant state, and a plurality of feature point positions set in the lower limb to specify the state of the lower limb. And feature point calculation means for obtaining every predetermined time,
The feature point calculating means calculates the position of each feature point at each predetermined time by substituting the measured value of the position and orientation sensor into an arithmetic expression based on a predetermined lower limb model. Measuring device.

Images