Analysis apparatus, racket selection diagnosis and racket design support system

Abstract

To provide a technique of simulating a swing motion of a racket with high accuracy.SOLUTION: A three-dimensional equation of motion model 101 consists of a collection of multiple segments and joints that represent a racket and the upper limb as one. The output of the equation of motion model 101 includes at least the acceleration and angular acceleration of each segment. An external force calculation unit 105 calculates the translational force Fext exerted on the shoulder joint by a moving person from motion trajectory data when swinging the racket. A joint torque calculation unit 110 calculates a joint torque Thuman that is exerted on the joint corresponding to the shoulder, elbow, and wrist joints by the moving person from the motion trajectory data. A restoration torque calculation unit 120 calculates a restoration torque Tshaft acting on the joint that virtually divides the racket according to the racket deformation, using the output of the equation of motion model and a constant parameter of the racket.SELECTED DRAWING: Figure 3

Description

[Technical field]

[0001] The present disclosure relates to a technique for simulating the swing motion of a racket for badminton or the like. [Background technology]

[0002] A technique is used to simulate the swing motion of a sports tool based on a motion equation model. For example, Japanese Patent No. 6845433 (Patent Document 1) discloses a technique for analyzing with high precision the deformation during the swing motion of a golf club (hitting tool).

[0003] Patent document 1 describes a technology that calculates the center of rotation of a hitting tool during a swing based on measurement data of the behavior of a part of the hitting tool (the grip end) during a swing, and analyzes the deformation of the hitting tool during a swing based on the measurement data and the calculated center of rotation.

[0004] In addition, the following non-patent document 1 proposes a method for analyzing the swing motion of a badminton racket by constructing an upper limb-racket coupled model and clarifying the effect on body motion when the inertia parameters of the racket are changed based on a dynamic analysis method. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Patent No. 6845433 [Non-patent literature]

[0006] [Non-Patent Document 1] Sekiya Koike and Tomohiro Hashiguchi, "Dynamic Analysis of Badminton Smash Motion Considering Deformation of Racket Shaft (Effect of Racket Inertia Parameters on Swing Motion)," Proceedings of the Symposium: Sports and Human Dynamics 2012, Session ID225, November 14, 2012. Summary of the Invention [Problem to be solved by the invention]

[0007] The main design elements of badminton rackets are inertia elements such as mass, static balance, and moment of inertia, rigidity elements such as flex and torque, or vibration characteristics resulting from both. Racket design combines these elements to improve resilience and swing speed, or to reduce the load on the body.

[0008] On the other hand, the operating characteristics during actual use differ depending on the player swinging the racket, and there is a concern that the cooperation of many testers and the production of prototypes will be necessary to efficiently determine the design parameters suitable for each target user. Therefore, it is desirable to improve the efficiency of design by analyzing the swing motion of the racket using simulation.

[0009] However, the simulation in Patent Document 1 inputs measurement data of the behavior of a part (grip end) included in the hitting tool during a swing as a result of physical movement, and there is a concern that there is a limit to how highly accurate it can be because it is not possible to simulate physical movement and the behavior of the hitting tool together. In addition, although Non-Patent Document 1 proposes the construction of an upper limb-racket coupling model, it does not fully describe a specific simulation method using this model.

[0010] The present disclosure has been made to solve such problems, and an object of one aspect of the present disclosure is to provide a technique for simulating the swing motion of a racket with high accuracy. [Means for solving the problem]

[0011] In an embodiment of the present disclosure, an analysis device is provided. The analysis device includes an equation of motion model, an external force calculation unit, a first torque calculation unit, and a second torque calculation unit. The equation of motion model is a three-dimensional equation of motion model that integrates the racket and the upper limbs of an actor swinging the racket and is expressed as a collection of multiple segments and multiple joints. The multiple segments include multiple first segments that are components of the upper limbs and multiple second segments that are virtually divided into the racket. The multiple joints include a shoulder joint that corresponds to the end of the upper limbs opposite the racket and a first joint that corresponds to each of the joints that connect adjacent first segments, and a second joint that is a virtual joint having a rotational spring that connects adjacent second segments. Inputs of the equation of motion model include a joint torque that the actor applies to the first joint, a restoring torque that applies to the second joint in response to the deformation of the racket, a motion-dependent term that indicates the centrifugal force and the Coriolis force, a translational force that the actor applies to the end of the upper limbs opposite the racket, and gravitational acceleration. The output of the equation of motion model includes the acceleration and angular acceleration of each segment, the constraint force generated in each joint, and the constraint moment acting on the joint. The external force calculation unit calculates a translational force from the motion trajectory data when the player swings the racket. The first torque calculation unit calculates a joint torque from the motion trajectory data. The second torque calculation unit calculates a restoring torque using the rotation behavior of each second joint calculated using the output of the equation of motion model and constant parameters of the racket.

[0012] In another embodiment of the present disclosure, a racket selection diagnostic system is provided. The racket selection diagnostic system includes the above-mentioned analysis device, a first data input unit, a second data input unit, and a diagnostic information generation unit. The first data input unit selectively inputs one racket data among a plurality of racket data of a plurality of different models to the analysis device. At least a part of the constant parameters of the racket differs between the plurality of racket data. The second data input unit inputs motion trajectory data when a user of the racket selection diagnostic system swings a racket to the analysis device. The analysis device calculates a simulation output value, which is an output of the motion equation model, using a motion equation model constructed by reflecting the one racket data input by the first data input unit and the motion trajectory data input by the second data input unit. The diagnostic information generation unit generates information for comparing the simulation output value when each of the plurality of racket data is input to the analysis device.

[0013] In yet another embodiment of the present disclosure, a racket design support system is provided. The racket design support system includes the above-mentioned analysis device, a first data input unit, a second data input unit, and a design support information generation unit. The first data input unit inputs one of M1 racket data (M1: natural number) to the analysis device. The second data input unit inputs one of M2 motion trajectory data (M2: natural number) to the analysis device. At least one of M1 and M2 is 2 or more, and when M1 is 2 or more, at least a part of the constant parameters of the racket is different between the M1 racket data, and when M2 is 2 or more, swing performers of the M2 motion trajectory data are different. The analysis device calculates a simulation output value, which is an output of the motion equation model, using the motion equation model constructed by reflecting one racket data input by the first data input unit and the motion trajectory data input by the second data input unit. The design support information generating unit generates information for comparing simulation output values ​​when each combination of one of the M1 racket data and one of the M2 movement trajectory data is input to the analysis device. Effect of the Invention

[0014] According to the present disclosure, it is possible to realize a highly accurate simulation of a racket swing motion using a three-dimensional simulation model in which the upper limbs and the racket are integrated, and further, it is possible to construct a racket selection diagnostic system and a racket design support system using the results of the simulation. [Brief description of the drawings]

[0015] [Figure 1] 2 is a conceptual diagram illustrating a structure of an analysis object model used in the analysis device according to the present embodiment. FIG. [Diagram 2] 2 is a conceptual diagram illustrating a process of deriving an equation of motion model from the three-dimensional model shown in FIG. 1. [Diagram 3] 1 is a block diagram illustrating a configuration of an analysis device according to an embodiment of the present invention. [Figure 4] FIG. 13 is a conceptual diagram illustrating an example of attaching markers to an actor during motion capture. [Diagram 5] FIG. 13 is a conceptual diagram illustrating an example of attaching a marker to a racket during motion capture. [Figure 6] FIG. 1 is a conceptual diagram illustrating an example of virtual segment division of a racket and measurement of stiffness parameters of each segment. [Figure 7] FIG. 13 is a block diagram illustrating an example of the configuration of a joint torque calculation unit having a torque correction function using an observed value. [Figure 8] 1 is a conceptual diagram illustrating an example of a hardware configuration of an analysis device according to a first embodiment. [Figure 9] FIG. 2 is a block diagram illustrating an example of a hardware configuration of an information processing device. [Figure 10] 4 is a flowchart illustrating a control process of a simulation performed by the analysis device according to the first embodiment. [Figure 11] 4 is a first waveform diagram showing an example of a simulation result obtained by the analysis device according to the first embodiment; FIG. [Figure 12]FIG. 5 is a second waveform diagram showing an example of a simulation result obtained by the analysis device according to the first embodiment. [Figure 13] FIG. 13 is a waveform diagram showing a part of the data in FIG. 12. [Figure 14] FIG. 11 is a block diagram illustrating an example of the configuration of a racket selection diagnostic system according to a second embodiment. [Figure 15] 15 is a flowchart illustrating a control process by the racket selection diagnostic system shown in FIG. 14. [Figure 16] 15 is a conceptual diagram showing an example of display of diagnostic information by the racket selection diagnostic system shown in FIG. 14. [Figure 17] FIG. 11 is a block diagram illustrating an example of the configuration of a racket design assistance system according to a second embodiment. [Figure 18] 18 is a flowchart illustrating a first example of a control process by the racket design support system shown in FIG. 17. [Figure 19] 18 is a flowchart illustrating a second example of a control process by the racket design support system shown in FIG. [Figure 20] 18 is a conceptual diagram showing an example of design support information provided by the racket design support system shown in FIG. 17. [Figure 21] 18 is a conceptual diagram showing another example of design support information provided by the racket design support system shown in FIG. 17. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0016] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings. In the following, the same or corresponding parts in the drawings are denoted by the same reference characters, and their description will not be repeated in principle. [Embodiment 1] <Derivation of the equation of motion model>

[0017] The analysis device according to the first embodiment is characterized in that it uses a three-dimensional simulation model that integrates a racket (more specifically, a badminton racket) and the upper limbs of a person swinging the racket. Therefore, the derivation of a motion equation model for constructing the three-dimensional simulation model will be described first.

[0018] FIG. 1 is a conceptual diagram illustrating the structure of an analysis object model to be analyzed by an analysis device according to the first embodiment.

[0019] 1, the three-dimensional model to be analyzed is composed of a collection of multiple segments SGh1-SGh3, SGr1-SGr6 and multiple joints JNh0-JNh3, JNr1-JNr5 that are an integrated model of a human upper limb 10 and a racket 20. Segments SGh1-SGh3 correspond to the "multiple first segments," and segments SGr1-SGr6 correspond to the "multiple second segments."

[0020] The model of the upper limb 10 includes a segment SGh1 corresponding to the upper arm, a segment SGh2 corresponding to the forearm, and a segment SGh3 corresponding to the hand. The segment SGh1 (upper arm) and the segment SGh2 (forearm) are connected by a joint JNh1 corresponding to the elbow. The segment SGh2 (forearm) and the segment SGh3 (hand) are connected by a joint JNh2 corresponding to the wrist. The joint JNh0 between the upper limb 10 (segment SGh1) and the trunk (not shown) corresponds to the shoulder joint. The shoulder joint (joint JNh0) also corresponds to the end of the upper limb 10 on the opposite side to the racket 20. Each of the joints JNh0 to JNh3 corresponds to a "first joint", and each of the joints JNr1 to JNr5 corresponds to a "second joint".

[0021] The model of the racket 20 includes a segment SGr1 that corresponds to the grip, segments SGr2 to SGr5 that correspond to the shaft, and a segment SGr6 that corresponds to the racket head. The segment SGr1 (grip) is connected to a segment SGh3 (hand) by a joint JNh3 for holding the racket 20. Furthermore, the segment SGr1 (grip) and the segment SGr2 (shaft base) are connected by a joint JNr1 that has a rotational spring.

[0022] The shaft of the racket 20 is virtually divided into segments SGr2 to SGr5 by joints JNr2 to JNr4, which are virtual joints having rotational springs. Segment SGr5 (shaft tip) and segment SGr6 (racket head) are connected by joint JNr5 having a rotational spring.

[0023] FIG. 2 is a conceptual diagram illustrating a process of deriving an equation of motion model from the three-dimensional model in FIG.

[0024] For each segment SG (collectively referring to segments SGh1 to SGh3, SGr1 to SGr6) and each joint JN (collectively referring to JNh0 to JNh3, JNr1 to JNr5) that make up the three-dimensional simulation model shown in Figure 1, a motion equation model is constructed by obtaining the three processes shown in Figure 2(a) to (c).

[0025] First, as shown in FIG. 2(a), for each segment SG, equations of motion are established for the translational and rotational motion of the center of gravity CG within the three-dimensional coordinates (x, y, z).

[0026] 2(b), two adjacent segments SG are connected by a joint JN equivalent to a joint, etc. Therefore, the position (displacement) of the end point of the segment SG is restricted by the connection constraint equation.

[0027] Furthermore, as illustrated in FIG. 2(c), it is understood that the rotation directions of the joints JNh0 to JNh3, which correspond to the joints, are restricted by constraint equations for the axial degrees of freedom corresponding to the inversion and eversion forces of the elbow joints and the like.

[0028] Therefore, the equation of motion of each segment SG in Fig. 2(a) is simultaneously solved with the connection constraint equation in Fig. 2(b) at each joint JN and the constraint equation of the axial degree of freedom in Fig. 2(c) to derive the equation of motion model. The final form of the equation of motion model is shown by the following equation (1). The method of deriving the equation of motion by simultaneously solving Figs. 2(a) to 2(c) is the same as that in Non-Patent Document 1.

number

[0029] The variables and constants in equation (1) are as follows:

number

[0030] The acceleration of each segment and each acceleration are variables of a three-dimensional vector, "V" is the velocity of the translational motion of each segment SG (center of gravity CG), and "ω" is a variable indicating the angular velocity of the rotational motion of each segment SG (center of gravity CG). In the equation, "·" indicates time differentiation, with the acceleration being represented by the first-order derivative of the velocity V, and the angular acceleration being represented by the first-order derivative of the angular velocity ω. The output vector X that constitutes the left-hand side of equation (1) is a column vector composed of a group of variables that are three-dimensional vectors.

[0031] In equation (1), the output vector X on the left side is the coefficient matrix A -1 The coefficient matrix A is multiplied by the input matrix B. -1 corresponds to a set of constant terms that appear in the derivation process described below. As mentioned above, the coefficient matrix A -1 It is understood that the components include the weight and moment of inertia of each segment.

[0032] Regarding the equations of motion for the translational and rotational motion of each segment shown in Figure 2(a), the equation of motion for the translational motion of the center of gravity CG of each segment SG in the three-dimensional coordinate space is given by the following equation (2). Similarly, the equation of motion for the rotational motion is given by the following equation (3).

number

[0033] The variables and constants in equations (2) and (3) are as follows:

number

[0034] FIG. 2(b) shows an example of a joint constraint equation in which segment SGh1 (upper arm) and segment SGh2 (forearm), which are components of the upper limb 10 of a human body, are connected via joint JNh1 (elbow).

[0035] In the segment SGh1, the displacement vector xi of the center of gravity CG and the vector r i,cgD The sum of these gives the position of joint JNh1.

[0036] Similarly, in segment SGh2, the displacement vector x of the center of gravity CG i+1 and vector r from the center of gravity to joint JNh1 i+1,cgP The sum of these gives the position of joint JNh1.

[0037] The connection constraint equation, which expresses that the two adjacent segments SG do not separate during movement (i.e., are constrained) by the endpoints (displacement) of the vectors from the center of gravity of each of the segments SGh1 and SGh2 to the joint JNh1 coinciding, is expressed by the following equation (4).

number

[0038] Since equation (4) is a conditional equation related to displacement, it cannot be directly used simultaneously with equation (2) in terms of acceleration. Therefore, by differentiating equation (4) twice, the following equation (5) is derived.

number

[0039] In equation (5), in the process of second differentiation, angular velocity (ω) and angular acceleration (first derivative of ω) as well as the second derivative term of the displacement vector x, i.e., the acceleration term (first derivative of velocity V), appear. Since acceleration and angular acceleration are also used in equations (2) and (3), equation (5) makes it possible to simultaneously establish the coupling constraint equation with the equations of motion for translational and rotational motion.

[0040] In FIG. 2(c), the constraint equation at the joint JNh1 (elbow) is shown as an example to explain the constraint equation of the axial degree of freedom at the joint. In this example, the coordinate system (e 1,x ,e 1,y ,e 1,z ) and the coordinate system of segment SGh2 (forearm) (e 2,x ,e 2,y ,e 2,z ) the constraint condition shown in the following equation (6) is set.

number

[0041] Equation (6) introduces a condition that the vectors constituting the segment coordinate system are always orthogonal (i.e., the dot product of the vectors is 0). This expresses that, for the elbow joint, the forearm can move inward and outward and inward and outward, but cannot move inward and outward.

[0042] Although details are omitted, equation (6) expressed by a displacement (three-dimensional coordinate) vector can also be differentiated twice to make it possible to simultaneously solve the equations of motion of equations (2) and (3).In addition, in the entire three-dimensional simulation model, the constraints of equation (6) are set so as to express three degrees of freedom of the shoulder joint (abduction / adduction, horizontal abduction / adduction, internal / external rotation), two degrees of freedom of the inflexible joint (flexion / extension, pronation / supination), and two degrees of freedom of the wrist joint (palmar / dorsiflexion, radial / ulnar flexion).

[0043] The equation of motion model of the above equation (1) can be obtained by solving the second-order derivatives of equations (2), (3), (5), and (6) simultaneously. In this case, the description of the equations can be simplified by introducing the method of multibody dynamics in robotics.

[0044] As a result, as shown in equation (7), the acceleration (second derivative of displacement x or first derivative of velocity V) and angular acceleration (first derivative of angular velocity) of the center of gravity of n segments (n: natural number, in the example of Figure 1, n = 9), the translational force f (constraint force) acting on the end, and the constraint moment n acting on the elbow (joint JNh1) and wrist (joint JNh2), respectively, are elbow and n wrist The output vector X on the left side of equation (1) is obtained as a collection of all the variables (three-dimensional vectors) in

number

[0045] The input matrix B on the right side of equation (1) corresponds to the sum of sub-input matrices b1 to b5, as shown in equation (8).

number

[0046] The sub-input matrix b1 corresponds to the input term of the restoring torque Tshaft acting on the joints JNr1 to JNr5 of the racket shaft. The sub-input matrix b1 includes a conversion coefficient matrix Qs and a zero matrix for multiplying the restoring torque Tshaft (three-dimensional vector) by a variable related to the angular acceleration of the segments SGr1 to SGr5 of the racket 20.

[0047] The sub-input matrix b2 is an input term of the joint torque Thuman that is applied to the joints (shoulder, elbow, wrist) of the upper limb 10 and the virtual grip (joint JNh3) by the actor swinging the racket 20. The sub-input matrix b1 is configured with a conversion coefficient matrix Qh and a zero matrix for multiplying the joint torque Thuman (three-dimensional vector) by variables related to the angular accelerations of the segments SGh1 to SGh3 via the moment vector n in equation (3).

[0048] The sub-input matrix b3 corresponds to the input term of the external force (translational force) acting on the shoulder joint of the human body. The sub-input matrix b3 is described together with the conversion coefficient matrix Pe and the zero matrix for multiplying the shoulder joint force Fext (three-dimensional vector) by the variable related to the acceleration (translational motion) of each segment.

[0049] Furthermore, the sub-input matrix b4 corresponds to the motion-dependent terms expressing the centrifugal force and Coriolis force acting on the rotational motion, and is composed of the last term on the right side of the formula (3) and each term on the right side of the formula (5). Furthermore, the sub-input matrix b5 corresponds to the input term of the gravitational acceleration acting on the translational motion of each segment.

[0050] In addition, the coefficient matrix A shown on the right side of equation (1) -1 corresponds to a set of constants by which each variable is multiplied in the second-order derivatives of equations (2), (3), (5), and (6).

[0051] <Overall simulation using analytical equipment> FIG. 3 is a block diagram illustrating a configuration of analysis device 100 according to this embodiment. 3, analysis device 100 includes equation of motion model 101 corresponding to equation (1), an external force calculation unit 105, a joint torque calculation unit 110, a restoring torque calculation unit 120, and an evaluation quantity calculation unit 130. Joint torque calculation unit 110 corresponds to an embodiment of a "first torque calculation unit", and restoring torque calculation unit 120 corresponds to an embodiment of a "second torque calculation unit".

[0052] The external force calculation section 105 and the joint torque calculation section 110 calculate a shoulder joint force Fext (vector) and a joint torque Thuman (vector), respectively, based on the motion trajectory data during a swing acquired by the three-dimensional motion analysis device 102.

[0053] The three-dimensional motion analysis device 102 typically uses optical motion capture to measure the trajectories of the actor and racket when the actor swings the racket 20 (e.g., a smash motion), and outputs motion trajectory data.

[0054] During motion capture, markers MRh1 to MRh8 are attached to the performer in correspondence with the shoulders, elbows, wrists, fingers (parts of the racket that are gripped), etc., as shown in Figures 4(a) and 4(b). Similarly, markers MRr1 to MRr10 are attached to the racket 20 in correspondence with the tip, end, and base of the racket head, various parts of the shaft (e.g., assumed positions of virtual joints), the grip end, etc., as shown in Figure 5.

[0055] For example, the three-dimensional motion analysis device 102 captures the swing of the racket 20 in FIG. 5 by the performer in FIG. 4 from different angles using multiple cameras. The two-dimensional position of each marker MR (collectively referring to MRh1 to MRh8, MRr1 to MRr10) is acquired by the multiple cameras, and the three-dimensional position of each marker MR can be measured by further performing image processing. The three-dimensional motion analysis device 102 can generate motion trajectory data based on changes in the three-dimensional position of each marker MR. The motion trajectory data can include values ​​of the displacement, velocity, acceleration, angular velocity, and angular acceleration of each marker MR over time.

[0056] The three-dimensional motion analysis device 102 is not limited to optical motion capture, and can also be configured with motion capture using an inertial sensor. In this case, by attaching inertial sensors to at least the forearm, upper arm, and hand of the performer and the grip end of the racket 20, it is possible to obtain motion trajectory data when the performer swings (smashes) the racket 20. In this way, the three-dimensional motion analysis device 102 can be configured in any way as long as it is possible to obtain appropriate motion trajectory data as an input to the equation of motion model 101.

[0057] 3 again, the external force calculation unit 105 calculates the shoulder joint force Fext (vector) based on the behavior data of the markers MRh1, MRh2, MRh4, etc. attached to the shoulders of the performer. Each directional component of the shoulder joint force Fext is calculated so as to indicate the change over time from the start of the swing (swing phase 0(%)) to the end of the swing (swing phase 100(%)).

[0058] The joint torque calculation unit 110 calculates the joint torque Thuman (vector) as the torque to be applied to each joint of the shoulder, elbow, and wrist to generate the motion grasped from the motion trajectory data, by known inverse dynamics calculation. Each directional component of the joint torque Thuman is also calculated so as to show the change over time from the start of the swing to the end of the swing.

[0059] The shoulder joint force Fext (vector) from the external force calculation unit 105 and the joint torque Thuman (vector) from the joint torque calculation unit 110 are input as sub-input matrices b3 and b2 in equation (8) to the equation of motion model 101. As a result, the equation of motion model 101 provides a simulation output of the motion (translation and rotation) of each segment SG into which the upper limb 10 and the racket 20 are divided.

[0060] At this time, due to the arrangement of virtual joints having rotational springs, a difference in displacement corresponding to the bending deformation of the shaft occurs between the segments SGr1 to SGr6 constituting the racket 20. In response to the occurrence of bending deformation, which is an elastic deformation, a restoring torque Tshaft (vector) is generated. That is, the initial value of each directional component of the restoring torque Tshaft is 0 before the start of the swing of the racket 20 (deformation amount 0), and becomes a value different from 0 as bending occurs in the shaft of the racket 20 after the start of the swing.

[0061] In this embodiment, in order to model this bending deformation (elastic deformation), the racket 20 is divided into a plurality of segments by virtual joints.

[0062] FIG. 6 shows a conceptual diagram illustrating an example of virtual segment division of the racket 20 and measurement of the stiffness parameters of each segment.

[0063] As shown in Fig. 6(a), racket 20 is provided with joints JNr2-JNr4 as virtual joints having rotational springs so as to divide the shaft into four parts between joint JNr1, which connects the grip part (segment SGr1) and the base of the shaft, and joint JNr5, which connects the tip of the shaft and the racket head. The positions of joints JNr2-JNr4 can be defined by distances l1-l4 between adjacent joints shown in Fig. 6(a).

[0064] As shown in Fig. 6(a), a measurement force ftp for stiffness distribution can be applied by adding an additional mass to the tip of racket 20 while the grip portion (segment SGr1) of racket 20 is restrained on measurement stand 30. The stiffness parameter [N / red] of each joint JNr1-JNr5 can be obtained from the deformation amount of each segment SGr2-SGr6 under the action of measurement force ftp.

[0065] The rotational displacement (angle) at each joint JNrj (j: 1 to 5) can be defined by the angle θs,j generated between adjacent segments SGrj and SGr(j+1), as shown in FIG. 6(a).

[0066] The stiffness parameters can be measured by applying measuring forces in the x direction (front bending direction of the face surface, i.e., the hitting direction), y direction (side bending direction), and z direction (torsion direction of the face surface) shown in Figure 6(b).

[0067] The angle θs,j of each joint is also represented by a three-dimensional vector having components in the x, y, and z directions. The angle θs,j of each joint is determined according to the definition shown in FIG. 6(a) based on the displacement (translational and rotational motion) of the segments SGr1 to SGr6 of the racket 20 calculated from the acceleration and angular acceleration (first-order derivative of ω) of the segments SGr1 to SGr6 included in the output vector of the equation of motion model 101. It is also possible to determine the angular velocity by first-order differentiation of the angle θs,j. In this way, the "rotation behavior" of each joint can be determined from the displacement of the segments SGr1 to SGr6 calculated from the output of the equation of motion model 101.

[0068] The restoring torque Tshaft (vector) calculated by the restoring torque calculation unit 120 is expressed as the rotational torque τ s,1 ~τ s,5 It is represented by a collection of (vectors).

number

[0069] In addition, the rotational torque τ in equation (9) s,1 ~τ s,5 Collectively, τ s,j is expressed by the following equation (10).

number

[0070] As shown in equation (10), the rotational torque τ s,j is the angle θ in the x, y, and z directions between the segments at each joint JNrj s,j,x ~θ s,j,z and stiffness parameter k of racket 20 s,j,x ~k s,j,z The multiplication of each of these, the angular velocity in the x, y, and z directions (first derivative of the angle θ) and the viscosity coefficient c of the racket 20 s,j,x ~c s,j,z The above can be expressed as the sum of the respective multiplication values ​​of

[0071] For example, the stiffness parameter k, which corresponds to the spring constant of a rotational spring, s,j,x ~k s,j,z can be set as a constant using the experimental values ​​in Figure 6(a). Also, the viscosity coefficient c, which indicates the damping characteristics of the rotational spring, s,j,x ~c s,j,z can be a constant according to the material (type of carbon, composition, etc.). The restoring torque Tshaft (vector) obtained by equations (9) and (10) is input to the motion equation model 101 as the sub-input matrix b1 in equation (8).

[0072] Stiffness parameter k s,j,x ~k s,j,z and the viscosity coefficient c s,j,x ~c s,j,z corresponds to an example of the "constant parameters" of the joints JNr1 to JNr5. It is understood that the constant parameters include at least a stiffness parameter indicating the stiffness (spring coefficient) of the rotational spring, which is a virtual joint.

[0073] 3 again, the equation of motion model 101 executes a forward dynamics simulation using as input the shoulder joint force Fext and joint torque Thuman obtained from the motion capture, and the restoring torque Tshaf calculated based on the simulation output value of the equation of motion model 101. Specifically, a numerical integration operation of a differential equation (e.g., the fourth-order Runge-Kutta method) is used to calculate the change in value over time (swing phase) for each variable (three-dimensional vector) constituting the output vector X on the left side of equation (1).

[0074] <Improved precision of joint torque> In analysis device 100, when simulating equation of motion model 101 using the joint torque Thuman obtained by inverse dynamics calculation as an input value, there is a concern that differences may occur between the simulation results and the data at the time of measurement due to inertia parameter errors during modeling, errors during numerical integration, etc.

[0075] Therefore, in the analysis device 100 according to this embodiment, the joint torque calculation unit 110 in Fig. 3 can be configured by a joint torque calculation unit 110X (Fig. 7) having a torque correction function based on the observed values ​​of the joint angle and the joint angular velocity. The joint torque calculation unit 110X corresponds to a "first torque calculation unit" having a function of calculating a torque correction amount.

[0076] FIG. 7 is a block diagram illustrating an example of the configuration of a joint torque calculation unit 110X having a torque correction function.

[0077] 7, the joint torque calculation unit 110X includes an inverse dynamics calculation unit 115 and a correction calculation unit 150. The inverse dynamics calculation unit 115 has the same function as the joint torque calculation unit 110 in FIG. 3, and calculates, by inverse dynamics calculation, basic torques (vectors) τ acting on the joints JNh0 to JNh2 corresponding to the shoulder, elbow, and wrist joints of the upper limb 10, respectively. j,inv (j: an integer between 0 and 2) is calculated. That is, the basic torque τ j,invis equal to the joint torque Thuman (vector) output from the joint torque calculation unit 110. The inverse dynamics calculation unit 115 corresponds to one embodiment of the "basic torque calculation unit".

[0078] The correction calculation unit 150 includes subtraction units 152 and 157, a proportional control calculation unit 154, a differential control calculation unit 155, and an addition unit 156. The correction calculation unit 150 calculates the observed values ​​θ of the angles θ (joint angles) of three joints JNhj (j=0 to 2) corresponding to each joint by motion capture. j,obs and the simulation value θ by the equation of motion model 101 j The correction torque amount τ to reduce the deviation from j,crt Calculate the observed value θ j,obs can be obtained as part of the motion trajectory data.

[0079] Simulation value θ in Fig. 7 j Similar to the angle of each joint of the racket 20 described in equation (10), the observed value θ can be calculated as the angle between adjacent segments based on the displacements (translational and rotational movements) of the segments SGh1 to SGh3 of the upper limb 10 calculated from the acceleration and angular acceleration (first-order differential of ω) of the segments SGh1 to SGh3 included in the output vector of the equation of motion model 101. j,obs corresponds to one embodiment of the "first rotation angle", and the simulation value θ j corresponds to one example of the "second rotation angle."

[0080] Furthermore, for each joint, the basic torque τ j,inv Corrected torque amount τ j,crt Joint torque τ minus j is set as the joint torque Thuman (sub-input matrix b1) input to the motion equation model 101.

[0081] Specifically, the subtraction unit 152 subtracts the simulation value θ j and the observed value θ j,obs Deviation Δθ jThe proportional control calculation unit 154 calculates the deviation Δθ of the joint angle. j The differential control calculation unit 155 executes a control calculation to multiply the deviation Δθ by a proportional gain kp. j The control calculation is performed to multiply the time derivative of by the derivative gain kd. j It is understood that the time derivative of corresponds to the deviation between the simulated and observed values ​​for the angular velocity of the joint.

[0082] The adder 156 adds the output values ​​of the proportional control calculator 154 and the differential control calculator 155 to obtain a correction torque amount τ j,crt The subtraction unit 157 calculates the basic torque τ j,inv Corrected torque amount τ j,crt The joint torque τ input to the motion equation model 101 is obtained by subtracting j Calculate.

[0083] Therefore, the joint torque τ j is calculated according to the following formula (11): This calculation is performed for each of the joints JNh0 to JNh2 (j: 0 to 2) and for each of the components in the x direction, y direction, and z direction.

number

[0084] The correction calculation unit 150 calculates the deviation Δθ of the joint angle. j The PD (proportional-differential) control calculation for the rotation behavior (angle and angular velocity) of each joint is performed to bring the difference between the simulation value and the observed value closer to zero. j,crt Thus, the joint torque calculation unit 110X can set the joint torque waveform input to the equation of motion model 101, that is, the change over time of the joint torque Thuman (vector) from the start of the swing to the end of the swing, so as to reduce the error between the movement shown in the simulation result and the actual movement. As a result, it can be understood that the application of the joint torque calculation unit 110X can improve the accuracy of the simulation result.

[0085] <Introduction of evaluation quantities>

[0086] Referring again to FIG. 3, the evaluation amount calculation unit 130 can extract or calculate the evaluation amount q from variables included in the simulation output value by the equation of motion model 101, thereby making it possible to efficiently evaluate the racket.

[0087] As an example, assuming that the head speed of the racket 20 is the evaluation amount q, the head speed corresponds to the speed (vector magnitude) of the segment SGr6 of the racket 20. Therefore, the evaluation amount q can be calculated based on the integral value of the acceleration (first derivative of V) of the segment SGr6, which is one of the variables included in the output vector X.

[0088] Therefore, when the selection matrix S determined so as to extract the acceleration of the segment SGr6 from the output vector X in the equations (1) and (7) is used, the evaluation amount q is expressed by the following equation (12).

number

[0089] C in formula (12) Shaft , C Human , C Force , C M.D.T. and C. Grav is expressed by the following formulas (13a) to (13e). Formulas (13a) to (13e) are obtained by multiplying the right-hand side of formula (1) by the selection matrix S in the same manner as the left-hand side, and then integrating the result over time, and dividing the result into sub-input matrices b1 to b5 (formula (1)).

number

[0090] C in formula (12) Shaft indicates a quantitative breakdown of the contribution (degree of action) of the restoring torque Tshaft input by the sub-input matrix b1 to the evaluation quantity q (head speed). Similarly, CHuman , C Force , C M.D.T. and C. Grav indicates a quantitative breakdown of the contribution (degree of action) of each of the joint torque Thuman, shoulder joint force Fext, motion-dependent terms (centrifugal force and Coriolis force), and gravitational acceleration to the evaluation quantity q (head speed). This makes it possible to quantitatively evaluate the magnitude of the influence of each factor (input) on the evaluation quantity q (for example, head speed).

[0091] The evaluation amount q can be arbitrarily extracted or calculated by setting the selection matrix S. For example, the constraint moment n acting on the joint JNh1 (elbow) or the joint JNh2 (wrist) from the output vector X in the formulas (1) and (7) is elbow or n wrist In this case, it is possible to quantitatively evaluate the evaluation amount q that quantifies the strain on the athlete's elbow and wrist, and the magnitude of the influence of each factor (input) on the evaluation amount. <System configuration example>

[0092] Next, a configuration example for implementing analysis device 100 according to the first embodiment will be described. FIG. 8 is a conceptual diagram illustrating an example of a hardware configuration of analysis device 100 according to the first embodiment.

[0093] With reference to Fig. 8, the function of the analysis device according to the first embodiment is realized, for example, by the execution of a predetermined analysis program by the information processing device 201. In addition, the other information processing device 203 may include, for example, a computer (arithmetic device) in which a program for generating swing motion trajectory data (Fig. 3) from a motion capture result is implemented. By transmitting data from the computer to the information processing device 201, input of the motion trajectory data to the analysis device 100 in Fig. 3 can be realized.

[0094] When information processing device 201 executes the above analysis program, the functions of the blocks constituting analysis device 100 shown in FIG.

[0095] That is, the function of each block shown in each block diagram including Fig. 3 is basically realized by software processing by executing a program. However, at least a part of the function of any block can be realized by a digital circuit such as an FPGA (Field Programmable Gate Array) or an ASIC (Application Specific Integrated Circuit), or an analog circuit.

[0096] FIG. 9 shows an example of the hardware configuration of the information processing device 201. 9, an information processing device 201 is configured on a computer base so as to include a CPU (Central Processing Unit) 220, a memory 230, and an input / output (I / O) circuit 240. The CPU 220, the memory 230, and the I / O circuit 240 can exchange data with each other via a bus 245.

[0097] A program including an analysis program is stored in advance in a partial area of ​​the memory 230, and the CPU 220 executes the program to perform a simulation by the analysis device 100. The I / O circuit 240 can input and output signals and data to and from other devices, for example, the information display device 202, the information processing device 203, and the cloud 204 shown in FIG. 8, via a communication device (not shown).

[0098] 8 again, the simulation results obtained by the information processing device 201 can be displayed on the information display device 202 configured with a liquid crystal display or the like. Note that, although the information processing device 201 and the information display device 202 are separate in the example of FIG. 8, they may be configured as one body.

[0099] Furthermore, the simulation results may be displayed on the information display device 202 or may be stored in the information processing device 201. The destination where the simulation results are stored is not limited to the information processing device 201 that executed the simulation, but may be another information processing device 203 connected to the information processing device 201 via a wired or wireless network, or may be the cloud 204.

[0100] Moreover, data for constructing the equation of motion model 101 may also be input to the information processing device 201 from another information processing device 203 or the cloud 204. Similarly, the motion trajectory data of the swing may be directly input from the other information processing device 203 to the information processing device 201 that executes the simulation, or may be input via the cloud 204. This allows past measurement data to be shared, etc.

[0101] 10 is a flowchart illustrating a control process of a simulation by analysis device 100. The control process shown in FIG.

[0102] 10, the CPU 220 acquires swing motion trajectory data generated by motion capture in step (hereinafter simply referred to as "S") 110. As described above, the motion trajectory data can include data values ​​on the time axis (swing phase) of the displacement, velocity, angular velocity, acceleration, and angular acceleration of predetermined parts of the upper limb 10 and the racket 20 based on the trajectories of the performer and the racket when the performer swings the racket 20 (for example, a smash motion). As described above, for example, the motion trajectory data can be acquired by data transmission from another information processing device 203 or the cloud 204.

[0103] The CPU 220 acquires human body data (including, for example, height, weight, sex, and competitive level) and racket data (mass, moment of inertia, shaft stiffness distribution, viscosity coefficient, and the like) of the motion trajectory data in S120. These data may be acquired by direct input to the information processing device 201, or may be acquired by data transmission from another information processing device 203 or the cloud 204.

[0104] In step S130, the CPU 220 constructs the equation of motion model 101 shown in equation (1) by reflecting the racket data acquired in step S120.

[0105] In S140, the CPU 220 executes a simulation calculation by the equation of motion model 101 constructed in S130 using the movement trajectory data acquired in S120. Specifically, the external force calculation unit 105, the joint torque calculation unit 110 (110X), and the restoring torque calculation unit 120 in Fig. 3 operate simultaneously in parallel with the equation of motion model 101, thereby calculating a shoulder joint force Fext (vector) and a joint torque Thuman (vector) to be input to the equation of motion model 101 based on the movement trajectory data. Furthermore, the restoring torque calculation unit 120 sequentially reflects the displacement amount (flexibility) of the racket shaft calculated by the equation of motion model 101 to calculate a restoring torque Tshaft (vector) to be input to the equation of motion model 101.

[0106] As a result, in S150, the CPU 220 acquires the values ​​of each variable (three-dimensional vector) constituting the output vector X on the left side of equation (1) at each time point from the start of the swing (swing phase 0(%)) to the end of the swing (swing phase 100(%)) as simulation output values.

[0107] Furthermore, in S160, the CPU 220 can extract or calculate a predetermined evaluation amount q by operating the evaluation amount calculation unit 130 in Fig. 3. At this time, as shown in equations (12), (13a) to (13d), the contribution (influence) of each factor (restoring torque, joint torque, shoulder joint force, motion-dependent term, and gravitational acceleration) to the evaluation amount can be quantitatively calculated.

[0108] Next, examples of simulation results obtained by the analysis device according to the first embodiment will be described with reference to the waveform diagrams of Figures 11 to 13. In the following, simulation results are shown when the evaluation amount q is the head speed of the racket 20.

[0109] In FIG. 11, the simulation values ​​(solid line) and the observed values ​​(dotted line) by motion capture for the head speed are plotted on the same time axis (swing phase is 0(%) to 100(%)).

[0110] In addition, FIG. 11 shows simulation values ​​obtained using the joint torque calculation unit 110X (FIG. 7) having a torque correction function using observed values, and it can be seen that the effect of the torque correction makes it possible to accurately simulate the head speed during the racket swing.

[0111] 12 shows a waveform diagram in which an example of an evaluation amount obtained by a simulation result using the analysis device according to the first embodiment is plotted. The horizontal axis of FIG. 12 is the time axis similar to that of FIG. 11. The vertical axis shows the evaluation amount q on the left side of the formula (12) and each term C on the right side. Shaft , C Human , C Force , C M.D.T. and C Grav The value of is shown.

[0112] Reference numeral 181 indicates a plot of the head speed [m / s], which is the evaluation quantity q, and reference numerals 182 to 184 indicate the influence term C of the restoring torque acting on the shaft. Shaft The component values ​​[m / s] of the x-direction, y-direction, and z-direction (Figure 6(b)) are plotted respectively.

[0113] Also, reference numeral 185 denotes the influence term of the joint torque C Human The magnitude of the vector obtained by integrating the x-, y-, and z-components (vectors) of is plotted. Reference numeral 186 denotes the influence term C of the motion-dependent term. M.D.T. The value (magnitude) of 187 is the influence term of gravitational acceleration C Grav The value (magnitude) of the shoulder joint force influence term C Force As for , the value was small compared to the other influence terms, so it is not plotted.

[0114] FIG. 13 shows a waveform diagram in which the influence terms (reference numerals 182 to 184) of the head speed (reference numeral 181) and the x-, y- and z-directions of the restoring torque acting on the shaft are extracted from the waveform diagram of FIG.

[0115] From Figure 13, we can see the influence of the restoring torque in the x-, y-, and z-directions of the racket shaft on the head speed. For example, we can see that the deformation in the x-direction (front bending direction) after the start of the racket shaft swing has the effect of reducing the head speed, while the deformation in the y-direction (side bending direction) and z-direction (twisting direction) contribute to improving the head speed.

[0116] As described above, the analysis device according to the first embodiment can build a three-dimensional simulation model (motion equation model) that integrates the upper limbs and the racket, and can simulate the swinging motion of a badminton racket with high accuracy by a dynamic analysis method using the motion equation. At this time, it can be understood that the simulation result can be made more accurate by introducing a joint torque correction function for bringing the difference between the simulated value and the observed value of the joint angle and angular velocity closer to zero, as described in FIG. 7.

[0117] Furthermore, since it is possible to quantify the evaluation quantity and the influence of each input term, including the joint torque and the restoring torque of the shaft reflecting the racket's design parameters (rigidity, etc.), on the evaluation quantity, it is understood that the simulation results can provide design support for optimizing the racket shaft rigidity. [Embodiment 2]

[0118] In the second embodiment, a configuration example of a racket selection diagnostic system and a racket design support system that utilizes the analysis device 100 according to the first embodiment will be described.

[0119] FIG. 14 is a block diagram illustrating an example of the configuration of a racket selection diagnostic system 250a according to the second embodiment.

[0120] 14, racket selection diagnostic system 250a includes analysis device 100 according to embodiment 1, racket data input unit 103, movement trajectory data input unit 104, and diagnostic information generation unit 106. In Fig. 14, racket data input unit 103 corresponds to the "first data input unit," and movement trajectory data input unit 104 corresponds to the "second data input unit."

[0121] The racket selection diagnostic system 250a is installed, for example, at a racket sales shop, and outputs diagnostic information for a racket suitable for a user using motion trajectory data obtained by three-dimensional motion analysis of the user's swing motion. The racket selection diagnostic system 250a can be realized by a hardware configuration similar to that of the analysis device 100 shown in FIG.

[0122] The motion trajectory data input unit 104 is configured, for example, by the three-dimensional motion analysis device 102 described in the first embodiment, and inputs motion trajectory data when a user swings (smashes) any racket to the analysis device 100. Alternatively, the motion trajectory data input unit 104 can be realized by inputting motion trajectory data previously acquired by the three-dimensional motion analysis device 102 to the analysis device 100 via the cloud 204.

[0123] The racket data input unit 103 inputs racket data (mass, moment of inertia, shaft stiffness distribution, viscosity coefficient, etc.) of N (N: an integer of 2 or more) models (types) into the analysis device 100 one by one.

[0124] Analysis device 100 executes a simulation of a swing motion of a badminton racket indicated by the racket data, using the motion trajectory data from motion trajectory data input unit 104 and the racket data input from racket data input unit 103, by analysis device 100 according to the first embodiment. As described in the first embodiment, analysis device 100 may output a predetermined evaluation amount q from variables included in the output (output vector X) of analysis device 100 in the simulation. In the second embodiment, it is assumed that the head speed of the racket and the constraint moment (joint load moment) acting on the elbow or wrist are calculated as the evaluation amount q and input from analysis device 100 to diagnosis information generating unit 106.

[0125] In racket selection diagnosis system 250a, the racket data input unit 103 inputs to analysis device 100 is switched under the same movement trajectory data, so that evaluation value q can be calculated for each of N model rackets.

[0126] Fig. 15 is a flowchart explaining the control process by the racket selection diagnosis system 250a. The control process shown in Fig. 15 can also be executed by the CPU 220 of the information processing device 201 activating a racket diagnosis program stored in the memory 230.

[0127] 15, in S210, the CPU 220 receives input of motion trajectory data of the user of the racket selection diagnostic system 250a from the motion trajectory data input unit 104. In S210, user data (height, weight, sex, competitive level, etc.) may also be input.

[0128] Furthermore, the CPU 220 receives, via S220, the input of racket data of N models. The N models for which racket data is input at S220 can be determined by direct specified input by the user. Alternatively, it is also possible to determine the N models at S220 by selecting one category based on user data under the condition that a set of a plurality of (N) racket models corresponding to each of a plurality of categories such as for women, for athletes, for beginners, etc. is predetermined. Also, in S220, a count value i for simulation calculation is set to an initial value (i = 1).

[0129] In S225, the CPU 220 selects, as the input to the analysis device 100, the racket data of the i-th (i is a natural number from 1 to N) model among the N models (models) input at S220.

[0130] The CPU 220 constructs, via S230, a motion equation model 101 of the analysis device 100 in the same manner as S130 (FIG. 10), reflecting the racket data selected in S225. Furthermore, in S240, in the same manner as S140 (FIG. 10), a simulation calculation is executed using the motion trajectory data received in S210 and the motion equation model constructed in S230. In S250, a simulation output value (from swing start to end) when the i-th racket model is swung is obtained by the simulation calculation in S240. The CPU 220 extracts or calculates a predetermined evaluation amount from the simulation output value obtained in S140 in the same manner as S150 (FIG. 10) via S260.

[0131] In S270, the CPU 220 determines whether the count value i is N or more. When i < N (when the NO determination is made in S270), the count value i is incremented by 1 (S275), and the process returns to S225. As a result, by repeating the processes of S225 to S260 until the count value i becomes from 1 to N, the simulation output value and the evaluation amount of each of the N racket models can be obtained.

[0132] When i≧N (YES in S270), CPU220 stores the evaluation amount calculated for each of the N racket models in S280. Furthermore, in S290, diagnostic information indicating the correspondence between the N racket models and the evaluation amount can be generated. The diagnostic information corresponds to information for comparing multiple simulation output values ​​(or evaluation amounts calculated from multiple simulation output values) for different racket data.

[0133] FIG. 16 shows an example of how the diagnostic information is displayed. 16 shows an example in which a head speed q1 and a joint load moment q2 are calculated as the evaluation quantity q. Generally, for a user, a large value of the head speed q1 is desirable, and a small value of the joint load moment q2 is desirable.

[0134] As shown in Fig. 16, by outputting and displaying a comparison of the evaluation quantities (q1, q2) for each racket model A, B, C, D, ..., it is possible to provide the user with diagnostic information for selecting an appropriate racket model (type of racket). The display example of Fig. 16 can be presented to the user using, for example, the information display device 202 (Fig. 8).

[0135] Alternatively, the evaluation amount q can be calculated by integrating multiple evaluation amounts q1 and q2 through weighting. That is, it is possible to set weighting coefficients wg1 and wg2 for the evaluation amounts q1 (head speed) and q2 (joint load moment) by the user, respectively, and generate diagnostic information using the integrated evaluation amount q* according to the following formula (14). In this case, the magnitude comparison of the evaluation amount q* for each racket model becomes clearer.

[0136] q*=wg1·q1+wg2·q2 …(14)

[0137] The weighting coefficient wg1 corresponds to the head speed q1, which is preferably large, and is therefore set to a positive value. Conversely, the weighting coefficient wg2 corresponds to the joint load moment q2, which is preferably small, and is therefore set to a negative value. The weighting coefficients wg1 and wg2 can be input as part of the user data based on the setting input by the user in S120.

[0138] In this way, the racket selection diagnostic system according to the second embodiment can simulate the swing motion of a plurality of racket models using motion trajectory data acquired from one swing motion of the user. Furthermore, by calculating an evaluation amount based on the simulation output value of each racket model, it becomes possible to provide the user with diagnostic information that quantitatively indicates the user's suitability among a plurality of racket models by comparing the evaluation values.

[0139] FIG. 17 is a block diagram illustrating an example of the configuration of a racket design assistance system 250b according to the second embodiment.

[0140] As shown in Fig. 17, racket design support system 250b includes analysis device 100, movement trajectory data input unit 104, and racket data input unit 103, similar to racket selection diagnosis system 250a in Fig. 14, and design support information generation unit 107. Racket design support system 250b can also be realized by a hardware configuration similar to that of analysis device 100 shown in Fig. 8. In Fig. 17 as well, racket data input unit 103 corresponds to the "first data input unit," and movement trajectory data input unit 104 corresponds to the "second data input unit."

[0141] In racket design support system 250b, M1 racket data are input to racket data input unit 103, which inputs each piece of racket data to analysis device 100.

[0142] Similarly, the movement trajectory data for M2 players is input to the movement trajectory data input unit 104, and the racket data input unit 103 inputs the movement trajectory data for each player to the analysis device 100.

[0143] M1 and M2 are integers equal to or greater than 1, and at least one of M1 and M2 is set to equal to or greater than 2. When M1≧2, at least one of the design parameters indicating the mass, moment of inertia, shaft stiffness distribution, viscosity coefficient, etc. is set to a different value among the M1 racket data.

[0144] Similarly, when M2≧2, data on the attributes (gender, age group, competitive level, etc.) of the performers who swung the racket are different are input as M2 pieces of movement trajectory data.

[0145] In the racket design support system 250b, an evaluation value q can be calculated for each combination of M1 pieces of movement trajectory data and M2 pieces of racket data (a total of M1 × M2), and input to the design support information generation unit 107.

[0146] Fig. 18 is a flowchart explaining a first example of the control process by the racket design support system shown in Fig. 17. The control process shown in Fig. 18 can also be executed by the CPU 220 of the information processing device 201 activating a racket diagnosis program stored in the memory 230.

[0147] FIG. 18 shows a control process for obtaining a simulation result when M2=1 and M1≧2, that is, when the person performing the racket swing is fixed and the racket data is changed.

[0148] As shown in FIG. 18, in S310a, CPU 220 receives, from movement trajectory data input unit 104, an input of movement trajectory data corresponding to the user for whom the racket is designed (M2=1).

[0149] Furthermore, the CPU 220 receives the input of M1 pieces of racket data (M1 ≥ 2) according to S320a. Among the M1 pieces of racket data, at least one of the plurality of design parameter values of the racket 20 (for example, at least one of the stiffness values of the segments SGr2 to SGr5 constituting the racket shaft) is set to a different value from each other. Also, in S320a, a count value i for simulation calculation is set to an initial value (i = 1).

[0150] In S325a, the CPU 220 selects, as the input to the analysis device 100, the racket data of the i-th model (i is a natural number from 1 to M1) among the M1 pieces of racket data input in S320a.

[0151] According to S330a, the CPU 220 constructs the motion equation model 101 of the analysis device 100 in the same manner as S230 (FIG. 15), reflecting the racket data selected in S325a. Furthermore, in S340a, in the same manner as S240 (FIG. 15), a simulation calculation is executed using the motion trajectory data input in S310a and the motion equation model constructed in S330a. In S350a, the simulation output value (swing start to end) when the racket 20 represented by the i-th racket data is swung is obtained by the simulation calculation in S340a. The CPU 220 extracts or calculates a predetermined evaluation amount from the simulation output value obtained in S350a in the same manner as S260 (FIG. 15) according to S360a.

[0152] In S370a, the CPU 220 determines whether the count value i is M1 or more. When i < M1 (when the NO determination in S370a), the count value i is incremented by 1 (S375), and the process returns to S325a. As a result, until the count value i becomes from 1 to M1, by repeating the processes of S325a to S360a, the evaluation amount q is obtained from the simulation results for each of the M1 pieces of racket data.

[0153] When i≧M1 (YES in S370a), CPU 220 stores the evaluation amount calculated for each of M1 racket data in S380a. Furthermore, in S390a, design support information indicating the correspondence between design parameter values ​​and evaluation amounts can be generated. The design support information corresponds to information for comparing multiple simulation output values ​​(or evaluation amounts calculated from multiple simulation output values) under different racket data or movement trajectory data.

[0154] FIG. 20 shows an example of the design support information obtained in the control process of FIG. The horizontal axis of Fig. 20 indicates design parameter values ​​that differ among the M1 racket data items, while the vertical axis indicates the above-mentioned q1 (head speed) and q2 (joint load moment) as the evaluation quantity q.

[0155] In the example of Figure 20, the values ​​of evaluation quantity q1 (head speed) and evaluation quantity q2 (joint load moment) obtained by simulation change with respect to changes in design parameter value P. The head speed (q1) reaches a maximum value when P = p1, and the joint load moment (q2) reaches a minimum value when P = p2. By generating the values ​​of p1 and p2 related to the design parameter value P as part of the design support information, it is possible to use them in optimal racket design.

[0156] Alternatively, it is also possible to generate the design support information in a form showing the change in the evaluation amount q* weighted according to the equation (14) relative to the change in the design parameter value P.

[0157] Next, Fig. 19 is a flowchart explaining a second example of the control process by the racket design support system shown in Fig. 17. The control process shown in Fig. 19 can also be executed by the CPU 220 of the information processing device 201 activating a racket diagnosis program stored in the memory 230.

[0158] FIG. 19 shows a control process for obtaining simulation results in which, contrary to FIG. 18, M1=1 and M2≧2, i.e., the racket data is fixed and performances are compared when different actors swing the racket.

[0159] As shown in FIG. 19, in S310b, the CPU 220 accepts input of racket data to be used for performance comparison by different users (M1=1).

[0160] Furthermore, CPU 220 accepts input of motion trajectory data for M2 people (M2≧2) through S320b. The attributes of the racket swingers differ among the M2 pieces of motion trajectory data. For example, the attributes can be defined by information such as gender, competitive level, age group, etc., such as "athlete-female," "female-general," "beginner-female," and "senior." In S320b, count value i for simulation calculation is set to an initial value (i=1).

[0161] In S325b, CPU 220 selects, as input to analysis device 100, the i-th (i=a natural number from 1 to M2) motion trajectory data from among the M2 pieces of motion trajectory data input in S320b.

[0162] CPU 220 constructs equation of motion model 101 of analysis device 100 in S330b, reflecting the racket data input in S310b. Furthermore, in S340b, a simulation calculation is performed using the equation of motion model constructed in S330b and the movement trajectory data received in S325b, similar to S340a (FIG. 18). In S350b, a simulation output value (from swing start to swing end) when racket 20 is swung according to the i-th movement trajectory data is acquired by the simulation calculation in S340b. In S360b, CPU 220 extracts or calculates a predetermined evaluation amount from the simulation output value acquired in S350b, similar to S360a (FIG. 18).

[0163] In S370b, the CPU 220 determines whether the count value i is greater than or equal to M2. When i < M2 (when the NO determination is made in S370b), the count value i is incremented by 1 (S375), and the process returns to S325b. As a result, until the count value i reaches from 1 to M2, by repeating the processes of S325b to S360b, the evaluation quantity q is obtained from the simulation results of each of the M2 operation observation data. In the control process of FIG. 19, since M1 = 1, the motion equation model is fixed. Therefore, in the loop process when i ≥ 2, the process of S330b may be skipped.

[0164] When i ≥ M2 (when the YES determination is made in S370b), the CPU 220 saves, by S380b, the evaluation quantity obtained for each of the M2 operation trajectory data. Further, in S390b, design support information indicating the correspondence between the operator (attribute) and the evaluation quantity can be generated.

[0165] FIG. 21 shows an example of the design support information obtained by the control process of FIG. 19. In the example of FIG. 21, the design support information can be generated so as to show the comparison of the evaluation quantities q1 and q2 for each user attribute. Thereby, information such as whether the racket design is realized for a desired target layer can be obtained. Also in FIG. 21, it is also possible to generate the design support information using the evaluation quantity q* (Equation (14)) obtained by integrating a plurality of evaluation quantities q1 and q2 by weighting.

[0166] Also, by combining the processes of FIGS. 18 and 19, even when M1 ≥ 2 and M2 ≥ 2, by obtaining and comparing the simulation output values and evaluation values for each combination of one operation trajectory data out of M1 and one racket data out of M2, it is also possible to obtain the design support information.

[0167] In the second embodiment, for the sake of simplicity, an example has been described in which racket selection diagnostic information and design support information are generated by comparing evaluation quantities q (q1, q2). However, any value based on the simulation output values ​​(S250, S350a, S350b) obtained by the simulation calculation of the motion equation model 101 can be used to generate racket selection diagnostic information and design support information without being limited to evaluation values.

[0168] In this embodiment, a badminton racket is assumed to be the racket being swung, but it is also possible to analyze rackets for other sports (e.g., squash) that are swung by the movement of the upper limbs from the shoulder joint.

[0169] The embodiments disclosed herein should be considered to be illustrative and not restrictive in all respects. The scope of the present disclosure is defined by the claims, not the above description, and is intended to include all modifications within the meaning and scope of the claims. [Explanation of symbols]

[0170] 10 Upper limb, 20 Racket, 30 Measurement stand, 100 Analysis device, 101 Motion equation model, 102 Three-dimensional motion analysis device, 103 Racket data input unit, 104 Motion trajectory data input unit, 105 External force calculation unit, 106 Diagnostic information generation unit, 107 Design support information generation unit, 110, 110X Joint torque calculation unit, 115 Inverse dynamics calculation unit, 120 Restoring torque calculation unit, 130 Evaluation amount calculation unit, 150 Correction calculation unit, 152, 157 Subtraction unit, 154 Proportional control calculation unit, 155 Differential control calculation unit, 156 Addition unit, 201, 203 Information processing device, 202 Information display device, 204 Cloud, 220 CPU, 230 Memory, 240 Circuit, 245 Bus, 250a Racket selection diagnosis system, 250b Racket design support system, CG center of gravity, Fext shoulder joint force (vector), JN, JNh0~JNh3, JNhj, JNr1~JNr5, JNrj joint, MRh1~MRh8, MRr1~MRr10 marker, SG, SGh1~SGh3, SGr1~SGr6, SGrj segment, SGF1 segment, Thuman joint torque (vector), Tshaf restoring torque (vector), ftp measurement force (rigidity), i parameter value, kd differential gain, kp proportional gain, q evaluation quantity, q1 head speed, q2 joint load moment, wg1, wg2 weighting coefficient.

Claims

1. It is an analytical device, The racket and the upper limbs of the person swinging the racket are integrated and represented by a three-dimensional equation of motion model consisting of a collection of multiple segments and multiple joints. The plurality of segments include a plurality of first segments which are components of the upper limb and a plurality of second segments which virtually divide the racket. The plurality of joints include a first joint corresponding to each of the shoulder joints and joints connecting adjacent first segments, which correspond to the end of the upper limb opposite to the racket, and a second joint which is a virtual joint having a rotational spring and connects adjacent second segments. The input to the aforementioned equation of motion model includes the joint torque that the operator applies to the first joint, the restoring torque that acts on the second joint in accordance with the deformation of the racket, motion-dependent terms representing centrifugal force and Coriolis force, the translational force that the operator applies to the end of the upper limb opposite to the racket, and the acceleration due to gravity. The output of the aforementioned equation of motion model includes the acceleration and angular acceleration of each segment, the constraint force acting on each joint, and the constraint moment acting on the joint. The aforementioned analysis device is An external force calculation unit that calculates the translational force from motion trajectory data when the operator swings the racket, A first torque calculation unit calculates the joint torque from the aforementioned motion trajectory data, The analysis apparatus further comprises a second torque calculation unit that calculates the restoring torque using the rotational behavior of each of the second joints calculated using the output of the equation of motion model and constant parameters of the racket.

2. The first torque calculation unit is, A basic torque calculation unit calculates the basic torque of the joint torque from the motion trajectory data according to inverse dynamics calculations, The system includes a correction calculation unit that calculates a torque correction amount to reduce the deviation between the first rotation angle based on the motion trajectory data for each of the first joints and the second rotation angle of each of the first joints calculated using the output of the equation of motion model, The analysis apparatus according to claim 1, wherein the joint torque included in the input of the equation of motion model is calculated using the basic torque and the torque correction amount.

3. The analysis apparatus according to claim 2, wherein the correction calculation unit calculates the torque correction amount by proportional-derivative control calculation with respect to the deviation of the first rotation angle and the second rotation angle.

4. The aforementioned equation of motion model is configured to calculate the output vector that constitutes the output by multiplying the input matrix that constitutes the input by a predetermined coefficient matrix, The constant parameters include the stiffness parameters of the virtual rotational springs of each of the second joints. The analysis apparatus according to claim 1, wherein the components of the coefficient matrix include the weight and moment of inertia of each of the second segments constituting the racket.

5. The system further comprises an evaluation quantity calculation unit that extracts or calculates an evaluation quantity based on a portion of the output included in the output vector, accompanied by multiplication of the output vector by a predetermined selection matrix. The analysis apparatus according to claim 4, wherein the evaluation quantity includes at least one of the velocity of the second segment corresponding to the head of the racket and the constraint moment acting on the first joint, which corresponds to the elbow or wrist among the plurality of joints.

6. The analysis apparatus according to claim 5, wherein the evaluation quantity calculation unit further calculates a quantitative breakdown of the effects of the joint torque, the restoring torque, the motion-dependent term, and the translational force on the evaluation quantity extracted by the selection matrix, by multiplying the input matrix by the selection matrix.

7. A racket selection diagnostic system, An analytical apparatus according to any one of claims 1 to 6, A first data input unit for selectively inputting one racket data from among multiple racket data for each of multiple racket models into the analysis device, The system comprises a second data input unit for inputting motion trajectory data into the analysis device when the user of the racket selection diagnostic system swings the racket as the operator, Among the plurality of racket data, at least some of the constant parameters of the racket differ from one another. The analysis device calculates the simulation output value, which is the output of the equation of motion model, using the equation of motion model constructed by reflecting the single racket data input by the first data input unit and the motion trajectory data input by the second data input unit. The aforementioned racket selection diagnostic system is A racket selection diagnostic system further comprising a diagnostic information generation unit that generates information for comparing the simulation output values ​​when each of the plurality of racket data is input to the analysis device.

8. A racket selection diagnostic system, The analytical apparatus according to claim 5 or 6, A first data input unit for selectively inputting one racket data from among multiple racket data for each of multiple racket models into the analysis device, The system comprises a second data input unit for inputting motion trajectory data into the analysis device when the user of the racket selection diagnostic system swings the racket as the operator, Among the plurality of racket data, at least one of the stiffness parameter of the second joint, the weight of the second segment, and the moment of inertia differs from one another. The analysis device calculates the simulation output value, which is the output of the equation of motion model, using the equation of motion model constructed by reflecting the single racket data input by the first data input unit and the motion trajectory data input by the second data input unit. The aforementioned racket selection diagnostic system is A racket selection diagnostic system further comprising a diagnostic information generation unit that generates information for comparing the evaluation quantities obtained from the simulation output values ​​when each of the plurality of racket data is input to the analysis device.

9. A racket design support system, An analytical apparatus according to any one of claims 1 to 6, A first data input unit for inputting one racket data from M1 (M1: natural number) racket data to the analysis device, The device comprises a second data input unit for inputting one of the M2 (M2: natural number) motion trajectory data into the analysis device, At least one of M1 and M2 is 2 or more. When M1 is 2 or more, at least some of the constant parameters of the rackets differ among the M1 racket data. When M2 is 2 or more, the swing operator of the M2 motion trajectory data is different. The analysis device calculates the simulation output value, which is the output of the equation of motion model, using the equation of motion model constructed by reflecting the single racket data input by the first data input unit and the motion trajectory data input by the second data input unit. The aforementioned racket design support system is A racket design support system further comprising a design support information generation unit that generates information for comparing the simulation output values ​​when each combination of one of the M1 racket data and one of the M2 motion trajectory data is input to the analysis device.

10. A racket design support system, The analytical apparatus according to claim 5 or 6, A first data input unit for inputting one racket data from M1 (M1: natural number) racket data to the analysis device, The device comprises a second data input unit for inputting one of the M2 (M2: natural number) motion trajectory data into the analysis device, At least one of M1 and M2 is 2 or more. When M1 is 2 or more, at least one of the stiffness parameter of the second joint, the weight of the second segment, and the moment of inertia differs from one another among the M1 racket data. When M2 is 2 or more, the swing operator of the M2 motion trajectory data is different. The analysis device calculates the simulation output value, which is the output of the equation of motion model, using the equation of motion model constructed by reflecting the single racket data input by the first data input unit and the motion trajectory data input by the second data input unit. The aforementioned racket design support system is A racket design support system further comprising a design support information generation unit that generates information for comparing the evaluation quantity obtained from the simulation output value when each combination of one of the M1 racket data and one of the M2 motion trajectory data is input to the analysis device.

Images