Robot, drive mechanism, reducer, torque sensor, and torque detection method

Abstract

[Problem] To provide a robot, a drive mechanism, a reduction gear, a torque sensor, and a torque detection method, all of which make it possible to improve accuracy in detecting a torque. [Solution] This robot 1 has: a torque sensor 2 that comprises a reduction gear 4 for outputting a second torque obtained by increasing a first torque outputted from a motor 3, an outer peripheral part 5 for outputting the second torque outputted from the reduction gear 4, and a sensor part 8 for detecting a torque that has been applied at least to the outer peripheral part 5; and an attenuation part 6 that is positioned in a transmission path through which the second torque is transmitted to the torque sensor 2 from the reduction gear 4 and that attenuates the transmission of the second torque.

Description

[Technical field]

[0001] The disclosed embodiments relate to a robot, a drive mechanism, a reducer, a torque sensor, and a torque detection method. [Background technology]

[0002] Patent Document 1 describes a torque sensor. This torque sensor includes a first member and a second member, a connecting member that connects the first member and the second member to be relatively rotatable about a first axis by an amount according to the magnitude of the torque and also connects the first member and the second member to be relatively rotatable about a second axis perpendicular to the first axis by a force other than the torque, a scale arranged on the first member, a light-emitting unit arranged on the second member for emitting light from a position on the second axis to the scale, and a light-receiving unit arranged on the second member for receiving light emitted from the light-emitting unit and reflected by the scale. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] International Publication No. 2019 / 021442 Summary of the Invention [Problem to be solved by the invention]

[0004] When the torque sensor is disposed near the reducer, it is desired to reduce the influence of vibrations of the reducer and improve the detection accuracy.

[0005] The present invention has been made in consideration of such problems, and has an object to provide a robot, a drive mechanism, a reducer, a torque sensor, and a torque detection method that are capable of improving the torque detection accuracy. [Means for solving the problem]

[0006] In order to solve the above problem, according to one aspect of the present invention, a robot is provided having a reducer that increases a first torque output from a motor and outputs a second torque, an output unit that outputs the second torque from the reducer, and a torque sensor having a sensor unit that detects at least the torque applied to the output unit, and a damping unit that is located in a transmission path of the second torque from the reducer to the torque sensor and damps the transmission of the second torque.

[0007] According to another aspect of the present invention, a drive mechanism is provided having a reducer that increases a first torque output from a motor and outputs a second torque, an output section that outputs the second torque from the reducer, and a torque sensor having a sensor section that detects at least the torque applied to the output section, and a damping section that is located in a transmission path of the second torque from the reducer to the torque sensor and damps the transmission of the second torque.

[0008] According to another aspect of the present invention, a reducer is provided having an output section that increases a first torque output from a motor and outputs a second torque to a torque sensor, and a damping section that is formed as part of the output section and attenuates the transmission of the second torque to the torque sensor.

[0009] According to another aspect of the present invention, a torque sensor is applied having an input section that inputs torque output from a reducer, an output section that outputs the torque input by the input section, a sensor section that detects at least the torque applied to the output section, and a damping section that is formed as part of the input section and damps the transmission of the torque from the reducer.

[0010] According to another aspect of the present invention, a torque detection method is applied, which includes increasing a first torque output from a motor by a reducer to output a second torque, outputting the second torque output from the reducer by an output section of a torque sensor, detecting at least the torque applied to the output section by a sensor section of the torque sensor, and attenuating the transmission of the second torque in a transmission path of the second torque from the reducer to the torque sensor. Effect of the Invention

[0011] According to the robot etc. of the present invention, the torque detection accuracy can be improved. [Brief description of the drawings]

[0012] [Figure 1] FIG. 1 is an external view illustrating an example of a configuration of a robot according to an embodiment. [Diagram 2] 1 is a cross-sectional view showing an example of a cross-sectional structure of a tip portion of a robot arm. [Diagram 3] FIG. 4 is a perspective view illustrating an example of the appearance of a damping portion. [Figure 4] FIG. 2 is a plan view illustrating an example of a configuration of the torque sensor as viewed from the Z-axis positive direction side. [Diagram 5] FIG. 4 is a side view illustrating an example of a configuration of the torque sensor as viewed from the negative Y-axis direction. [Figure 6] 1 is a plan view illustrating an example of a configuration of an outer circumferential portion, an inner circumferential portion, a connecting portion, and the like of a torque sensor as viewed from the positive direction of the Z axis. FIG. [Figure 7] 7 is a cross-sectional view showing an example of a cross-sectional structure of the substrate taken along the line VII-VII in FIG. [Figure 8] 11 is a plan view illustrating an example of a configuration of a detection unit as viewed from the negative Z-axis direction. FIG. [Figure 9] FIG. 2 is an explanatory diagram conceptually illustrating an example of an overall configuration of a substrate of a torque sensor. [Figure 10] FIG. 2 is a block diagram illustrating an example of a functional configuration of a torque calculation unit that executes a torque calculation process. [Figure 11]13 is a graph for explaining how the detection error caused by the torque ripple that is specific to a reducer equipped with a strain wave gear mechanism is cancelled by adding the output of the first system and the output of the second system. [Figure 12] 10 is a graph showing an example of measurement results of an output error of a first system and an output error of a second system in a comparative example in which a damping section is not provided. [Figure 13] 11 is a graph showing an example of a measurement result of an output error between an average value of an output of a first system and an output of a second system in a comparative example in which a damping section is not provided. [Figure 14] 11 is a graph showing an example of measurement results of an output error of a first system and an output error of a second system in an embodiment provided with a damping section. [Figure 15] 13 is a graph showing an example of a measurement result of an output error between an average value of an output from a first system and an output from a second system in an embodiment provided with a damping section. [Figure 16] 13 is a cross-sectional view showing an example of the cross-sectional structure of the tip of a robot arm in a modified example in which the damping section is formed as part of the output section of the reducer. FIG. [Figure 17] 11 is a cross-sectional view showing an example of the cross-sectional structure of the tip of a robot arm when the damping portion is formed as part of the input portion of the torque sensor. FIG. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] Hereinafter, an embodiment will be described with reference to the drawings.

[0014] <1. Robot configuration> An example of the configuration of a robot 1 according to this embodiment will be described with reference to Fig. 1. As shown in Fig. 1, the robot 1 is a so-called dual-arm robot, and has a base 101, a body 102, and two arms 103L and 103R configured as separate bodies. Note that the robot 1 does not necessarily have to be a dual-arm robot, and may be a robot having a single arm.

[0015] The base 101 is fixed to a surface (e.g., a floor surface) on which the robot 1 is to be installed, for example, by anchor bolts or the like. The base 101 may be fixed to a surface other than the floor surface (e.g., a ceiling surface, a wall surface, a stand, etc.). The body 102 is supported on the upper end of the base 101 so as to be rotatable.

[0016] The arm 103L is rotatably supported on one side of the torso 102. The arm 103L includes a shoulder 104L, an upper arm A section 105L, an upper arm B section 106L, a lower arm 107L, a wrist A section 108L, a wrist B section 109L, and a torque sensor 2 as a flange section. Note that, although the present embodiment describes a case where a torque sensor is provided on the flange section, a torque sensor having a configuration similar to that of the torque sensor 2 may be provided on some or all of the joint sections other than the flange section.

[0017] The shoulder 104L is rotatably supported on one side of the body 102. The shoulder 104L is driven to rotate relative to one side of the body 102 by driving an actuator (not shown) provided at a joint between the shoulder 104L and the body 102.

[0018] The upper arm A portion 105L is supported rotatably on the tip side of the shoulder portion 104L. The upper arm A portion 105L is driven to rotate with respect to the tip side of the shoulder portion 104L by driving an actuator (not shown) provided at a joint between the upper arm A portion 105L and the shoulder portion 104L.

[0019] The upper arm B section 106L is rotatably supported at the tip of the upper arm A section 105L. The upper arm B section 106L is driven to rotate relative to the tip of the upper arm A section 105L by driving an actuator (not shown) provided at a joint between the upper arm B section 106L and the upper arm A section 105L.

[0020] The lower arm portion 107L is rotatably supported on the tip side of the upper arm B portion 106L. The lower arm portion 107L is driven to rotate with respect to the tip side of the upper arm B portion 106L by driving an actuator (not shown) provided at a joint between the lower arm portion 107L and the upper arm B portion 106L.

[0021] The wrist A portion 108L is rotatably supported on the tip side of the lower arm portion 107L. The wrist A portion 108L is driven to rotate relative to the tip side of the lower arm portion 107L by driving an actuator (not shown) provided at a joint between the wrist A portion 108L and the lower arm portion 107L.

[0022] The wrist B section 109L is supported rotatably on the tip side of the wrist A section 108L. The wrist B section 109L is driven to rotate with respect to the tip side of the wrist A section 108L by driving an actuator (not shown) provided at a joint between the wrist B section 109L and the wrist A section 108L.

[0023] The torque sensor 2 is rotatably supported on the tip side of the wrist B section 109L. The torque sensor 2 is driven to rotate relative to the tip side of the wrist B section 109L by driving an actuator 110L provided at a joint between the torque sensor 2 and the wrist B section 109L. The actuator 110L has a motor 3 and a reducer 4. Between the torque sensor 2 and the reducer 4, a damping section 6 (see Figs. 2 and 3 described later) is provided that damps the transmission of torque from the reducer 4 to the torque sensor 2.

[0024] A hand 120L is attached to the tip side of the torque sensor 2. The hand 120L rotates together with the torque sensor 2. The hand 120L includes a pair of claw members 130 that are movable in directions toward and away from each other.

[0025] On the other hand, the arm 103R has a structure symmetrical to the arm 103L, and is rotatably supported on the other side of the torso 102. The arm 103R includes a shoulder 104R, an upper arm A section 105R, an upper arm B section 106R, a lower arm 107R, a wrist A section 108R, a wrist B section 109R, and a torque sensor 2 as a flange section. Note that, although the present embodiment describes a case where a torque sensor is provided on the flange section, a torque sensor having a similar configuration to the torque sensor 2 may be provided on some or all of the joint sections other than the flange section.

[0026] The shoulder 104R is rotatably supported on the other side of the torso 102. The shoulder 104R is driven to rotate relative to the other side of the torso 102 by driving an actuator (not shown) provided at a joint between the shoulder 104R and the torso 102.

[0027] The upper arm A portion 105R is rotatably supported on the tip side of the shoulder portion 104R. The upper arm A portion 105R is driven to rotate with respect to the tip side of the shoulder portion 104R by driving an actuator (not shown) provided at a joint between the upper arm A portion 105R and the shoulder portion 104R.

[0028] The upper arm B section 106R is rotatably supported at the tip of the upper arm A section 105R. The upper arm B section 106R is driven to rotate relative to the tip of the upper arm A section 105R by driving an actuator (not shown) provided at a joint between the upper arm B section 106R and the upper arm A section 105R.

[0029] The lower arm section 107R is rotatably supported on the tip side of the upper arm B section 106R. The lower arm section 107R is driven to rotate with respect to the tip side of the upper arm B section 106R by driving an actuator (not shown) provided at a joint between the lower arm section 107R and the upper arm B section 106R.

[0030] The wrist A portion 108R is rotatably supported on the distal end of the lower arm portion 107R. The wrist A portion 108R is driven to rotate relative to the distal end of the lower arm portion 107R by driving an actuator (not shown) provided at a joint between the wrist A portion 108R and the lower arm portion 107R.

[0031] The wrist B section 109R is supported rotatably on the tip side of the wrist A section 108R. The wrist B section 109R is driven to rotate relative to the tip side of the wrist A section 108R by driving an actuator (not shown) provided at a joint between the wrist B section 109R and the wrist A section 108R.

[0032] The torque sensor 2 is rotatably supported on the tip side of the wrist B section 109R. The torque sensor 2 is driven to rotate relative to the tip side of the wrist B section 109R by driving an actuator 110R provided at a joint between the torque sensor 2 and the wrist B section 109R. The actuator 110R has a motor 3 and a reducer 4. Between the torque sensor 2 and the reducer 4, a damping section 6 (see Figs. 2 and 3 described later) is provided that damps the transmission of torque from the reducer 4 to the torque sensor 2.

[0033] A hand 120R is attached to the tip side of the torque sensor 2. The hand 120R rotates together with the torque sensor 2. The hand 120R includes a pair of claw members 140 that are movable in directions toward and away from each other.

[0034] Each of the actuators includes, for example, a motor and a reducer, etc. In the above, the rotation about the rotation axis along the longitudinal direction (or extension direction) of the arms 103L, 103R is called "rotation", and the rotation about the rotation axis approximately perpendicular to the longitudinal direction (or extension direction) of the arms 103L, 103R is called "turning", to distinguish between the two.

[0035] The "perpendicular" in the above explanation is not strict, and actual tolerances and errors are allowed. The "perpendicular" in the above explanation does not mean that virtual axes intersect, but includes the case of a twisted position as long as the directions of the virtual axes intersect.

[0036] <2. Configuration of torque sensor, motor, reducer, and damping unit> An example of the configuration of the torque sensor 2, motor 3, reducer 4, and damping unit 6 included in the robot 1 will be described with reference to Figures 2 and 3. Since the torque sensor 2, motor 3, reducer 4, and damping unit 6 in each of the arms 103L and 103R are similar in configuration, the configuration in the arm 103R will be described here as an example. Figure 2 is a cross-sectional view showing an example of the cross-sectional structure of the tip of the arm 103R. Figure 3 is a perspective view showing an example of the appearance of the damping unit 6. Note that the configuration of the sensor unit of the torque sensor 2 and the like is omitted in Figure 2.

[0037] 2, a hand 120R is rotatably supported at the tip of a wrist B portion 109R of an arm 103R via a motor 3, a reducer 4, and a torque sensor 2. The hand 120R is driven to rotate by driving an actuator 110R equipped with a motor 3 and a reducer 4. The hand 120R rotates together with the torque sensor 2.

[0038] The motor 3 has a motor frame 301, a bracket 303, a bearing 305, a motor shaft 307, a rotor 309, and a stator 311. The motor frame 301 and the bracket 303 are fixed with bolts (not shown) and are attached to the tip of the wrist B part 109R. For example, two bearings 305 are provided on the inner circumference of the bracket 303, and the hollow motor shaft 307 is supported rotatably by protruding to the base end side (the opposite side to the hand 120R; the same applies below). A rotor 309 is provided on the outer circumference of the motor shaft 307 and is disposed opposite a stator 311 provided on the inner circumference of the motor frame 301 with a radial gap therebetween. The motor 3 drives the motor shaft 307 to rotate and output a first torque. A cylindrical sleeve 313 fixed to the motor frame 301 is inserted inside the motor shaft 307, and a cable or the like can be wired inside.

[0039] The reducer 4 increases the first torque output from the motor 3 to output a second torque. The reducer 4 is, for example, a reducer equipped with a strain wave gear mechanism. The reducer 4 has a sliding bearing 401, a circular spline 403, a flexspline 405, and a wave generator 407. The sliding bearing 401 has an outer ring 409, a sliding body 411, and an inner ring 413. The outer ring 409 is fixed to the bracket 303 of the motor 3 together with the flexspline 405 by a bolt 415. The inner ring 413 is fixed to the circular spline 403, and is rotatably supported by the outer ring 409 together with the circular spline 403 via the sliding body 411. The inner circumferential side of the wave generator 407 is connected to the motor shaft 307 and rotates together with the motor shaft 307. The wave generator 407 is an example of a first input unit that inputs the first torque output from the motor 3. When the wave generator 407 rotates once, the circular spline 403 moves in the same rotational direction as the wave generator 407 by the difference in the number of teeth between the circular spline 403 and the flex spline 405 (for example, 2) because the flex spline 405 is fixed. The circular spline 403 and the flex spline 405 are an example of a speed reducer that increases the first torque input by the first input unit. The circular spline 403 is an example of a first output unit that outputs the second torque increased by the speed reducer. The speed reducer 4 has a bearing cover 419 that covers a bearing 417 of the wave generator 407.

[0040] The torque sensor 2 has a ring-shaped outer peripheral portion 5, a ring-shaped inner peripheral portion 7, and a sensor portion 8 (not shown in FIG. 2; see FIG. 4 described later). The sensor portion 8 is disposed between the outer peripheral portion 5 and the inner peripheral portion 7, and detects at least the torque applied to the outer peripheral portion 5 based on the relative displacement between the outer peripheral portion 5 and the inner peripheral portion 7. The outer peripheral portion 5 is fixed to the hand 120R by a plurality of bolts (12 in this example) (not shown) inserted into bolt holes 11 (see FIG. 4 described later). The inner peripheral portion 7 is fixed to the circular spline 403 of the reducer 4 by a plurality of bolts (12 in this example) 201 inserted into bolt holes 13 (see FIG. 4 described later). The inner peripheral portion 7 is an example of a second input portion that inputs the second torque output from the first output portion. The outer peripheral portion 5 is an example of an output portion that outputs the second torque output from the reducer 4, or a second output portion that outputs the second torque input by the second input portion. The bolt 201 is an example of a fixing member that fixes the second input portion to the first output portion.

[0041] A cylindrical seal member 203 protruding toward the reducer 4 is fixed to the base end side of the outer circumferential portion 5 by a bolt 205. The seal member 203 provides a seal while sliding against the outer circumferential surface of an outer ring 409 of a sliding bearing 401 of the reducer 4, preventing the intrusion of water and dust.

[0042] A recess 207 is formed on the outer periphery side of the base end side of the inner periphery portion 7. The outer periphery surface of the circular spline 403 of the reducer 4 fits into the recess 207 for positioning. A seal groove 209 is formed on the inner periphery side of the base end side of the inner periphery portion 7. A seal member 211 that provides a seal while sliding against a bearing cover 419 of the reducer 4 is provided within the seal groove 209. A bearing 213 that rotatably supports a sleeve 313 is provided further inner than the seal groove 209.

[0043] The damping portion 6 is located in a transmission path of the second torque from the reducer 4 to the torque sensor 2, and damps the transmission of the second torque. The damping portion 6 is located in a transmission path of the second torque from the circular spline 403 of the reducer 4 to the inner periphery 7 of the torque sensor 2. The damping portion 6 is interposed between the circular spline 403 and the inner periphery 7, and is a member formed separately from the reducer 4 and the torque sensor 2. The damping portion 6 is interposed between a first transmission surface 421 that transmits the second torque of the circular spline 403, and a second transmission surface 215 that transmits the second torque of the inner periphery 7. The second transmission surface 215 is formed inside the recess 207. The damping portion 6 is interposed between the first transmission surface 421 and the second transmission surface 215 at a position fixed by the bolt 201. The damping section 6 is not particularly limited in terms of its material, shape, etc., as long as it has a property capable of damping the transmission of the second torque from the reducer 4 to the torque sensor 2. For example, the damping section 6 may be an elastic shim member sandwiched between the first transmission surface 421 and the second transmission surface 215. In this case, the number of shim members may be one or more.

[0044] The damping portion 6 may be a metal gasket made of metal such as mild steel, pure iron, stainless steel, copper, aluminum, titanium, or a special alloy. In this case, a damping structure with excellent sealing properties, heat resistance, and pressure resistance can be realized, and is suitable for high temperature and pressure conditions or when a high level of sealing properties is required. Also, the damping portion 6 may be made of a material containing, for example, resin or rubber.

[0045] Fig. 3 shows an example of the appearance of the damping section 6. As shown in Fig. 3, the damping section 6 is a ring-shaped plate member having a predetermined thickness. The damping section 6 is formed with a plurality of bolt holes 601 (12 in this example) through which the bolts 201 are inserted.

[0046] The reducer 4, the torque sensor 2, and the attenuation unit 6 constitute a drive mechanism 10. The drive mechanism 10 executes a torque detection method including: increasing a first torque output from the motor 3 by the reducer 4 to output a second torque; outputting the second torque output from the reducer 4 by an outer circumferential portion 5 of the torque sensor 2; detecting the torque applied to at least the outer circumferential portion 5 by a sensor unit 8 of the torque sensor 2; and attenuating the transmission of the second torque in a transmission path of the second torque from the reducer 4 to the torque sensor 2.

[0047] The above-described configuration of the drive mechanism 10 is an example, and is not limited to the above. For example, in the above-described configuration, the flexspline 405 in the reducer 4 is fixed to the motor 3, and the circular spline 403 is the output part, but the opposite may be true, that is, the circular spline 403 is fixed to the motor 3, and the flexspline 405 is the output part. For example, in the above-described configuration, the inner periphery 7 in the torque sensor 2 is connected to the reducer 4 to serve as the input part, and the outer periphery 5 is connected to the hand 120R to serve as the output part, but the opposite may be true, that is, the outer periphery 5 is connected to the reducer 4 to serve as the input part, and the inner periphery 7 is connected to the hand 120R to serve as the output part. For example, in the above-described configuration, the reducer 4 is a reducer equipped with a strain wave gear mechanism, but the reducer may be a reducer equipped with a strain wave gear mechanism other than a strain wave gear mechanism, such as a planetary gear mechanism.

[0048] <3. Overall configuration of torque sensor> An example of the overall configuration of the torque sensor 2 will be described with reference to Figures 4 and 5. Figure 4 is a plan view of the torque sensor 2 as viewed from the positive direction of the Z axis, and Figure 5 is a side view of the torque sensor 2 as viewed from the negative direction of the Y axis.

[0049] 4 and 5, the torque sensor 2 has a ring-shaped outer circumferential portion 5, an inner circumferential portion 7 that is arranged radially inside the outer circumferential portion 5 and has a ring shape with a smaller diameter than the outer circumferential portion 5, and a plurality of connecting portions 9A to 9D (four in this example) that connect the outer circumferential portion 5 and the inner circumferential portion 7. Note that the inner circumferential portion 7 may have a solid structure (such as a disk-shaped member) instead of a hollow structure.

[0050] As shown in FIG. 4, the outer peripheral portion 5 and the inner peripheral portion 7 are arranged concentrically in the radial direction centered on the Z axis. The outer peripheral portion 5 is formed with a plurality of bolt holes 11 (12 in this example) through which a plurality of bolts (not shown) for fixing the outer peripheral portion 5 to external devices such as the hands 120L and 120R are inserted. The bolt holes 11 are arranged in a concentrated manner near each of the connecting portions 9A to 9D so that the positions corresponding to the connecting portions 9A to 9D in the circumferential direction around the Z axis are closer together than the positions not corresponding to the connecting portions 9A to 9D. The inner peripheral portion 7 is formed with a plurality of bolt holes 13 (12 in this example) through which a plurality of bolts 201 for fixing the inner peripheral portion 7 to an external device such as the reducer 4 are inserted. The bolt holes 13 are arranged at approximately equal angular intervals (30 degree intervals in this example) in the circumferential direction around the Z axis. The bolt holes 13 may also be arranged in a concentrated manner near each of the connecting portions 9A to 9D, similar to the bolt holes 11. Furthermore, bolt holes 11 may be arranged at approximately equal angular intervals in the circumferential direction around the Z axis, similar to bolt holes 13.

[0051] 5, the thickness of the inner circumferential portion 7 in the Z-axis direction and the thickness of the outer circumferential portion 5 in the Z-axis direction are formed to be approximately the same. The outer circumferential portion 5 and the inner circumferential portion 7 are arranged so that their bottom surfaces (end surfaces on the negative Z-axis direction side) and top surfaces (end surfaces on the positive Z-axis direction side) are approximately flush with each other. The outer circumferential portion 5 and the inner circumferential portion 7 may be formed to have different thicknesses.

[0052] The connecting parts 9A to 9D are arranged at approximately equal angular intervals (90 degree intervals in this example) in the circumferential direction around the Z axis in the space S (see FIG. 5) between the outer peripheral part 5 and the inner peripheral part 7. The connecting parts 9A to 9D connect the outer peripheral part 5 and the inner peripheral part 7 so as to be relatively rotatable around the Z axis by a minute amount corresponding to the magnitude of the torque (torsion moment Mz around the Z axis) to be detected by the torque sensor 2. For convenience of explanation, in the space S, the space between the connecting parts 9A and 9B in the circumferential direction is referred to as S1, the space between the connecting parts 9B and 9C is referred to as S2, the space between the connecting parts 9C and 9D is referred to as S3, and the space between the connecting parts 9D and 9A is referred to as S4. Note that the number and arrangement of the connecting parts may be other than those described above as long as the outer peripheral part 5 and the inner peripheral part 7 can be connected to each other with a predetermined rigidity.

[0053] As shown in FIG. 4, a plurality of (four in this example) board units PB1-PB4 are arranged in the space S between the outer peripheral portion 5 and the inner peripheral portion 7. Each of the board units PB1-PB4 has a plurality of boards and is configured as a set of boards (see FIG. 7 described later). Each of the board units PB1-PB4 is arranged in a plurality of spaces S1-S4 between the plurality of connecting portions 9A-9D in the circumferential direction around the Z axis. That is, the board unit PB1 is arranged in the space S1, the board unit PB2 in the space S2, the board unit PB3 in the space S3, and the board unit PB4 in the space S4. The board units PB3 and PB4 are provided with external connectors 15 and 17, respectively, for inputting power and transmitting / receiving signals to / from external devices.

[0054] In this example, the board units adjacent in the circumferential direction, that is, the board unit PB4 and the board unit PB1, the board unit PB1 and the board unit PB2, and the board unit PB2 and the board unit PB3, are electrically connected by the connection parts 19, 21, and 23 across the connecting parts 9A, 9B, and 9C. The connection parts 19, 21, and 23 are, for example, flexible printed circuits (FPCs). Note that the connection parts may be anything that connects the boards together so that signals can be transmitted and received, and may be, for example, lead wires, cables, connectors, etc., other than FPCs. In addition, in consideration of the convenience of signal transmission and reception, four FPCs may be used to electrically connect across the connecting parts 9A, 9B, 9C, and 9D.

[0055] In this embodiment, among the axes perpendicular to the Z axis, the axis passing through approximately the center position of the substrate unit PB2 in the circumferential direction (the direction that halves the angular interval between connecting portions 9B and 9C) is defined as the X axis, and the axis perpendicular to the X axis, i.e., the axis passing through approximately the center position of the substrate unit PB1 in the circumferential direction (the direction that halves the angular interval between connecting portions 9A and 9B), is defined as the Y axis.

[0056] <4. Configuration of outer periphery, inner periphery, connecting portion, etc.> An example of the configuration of the outer circumferential portion 5, the inner circumferential portion 7, the connecting portions 9A to 9D, etc. of the torque sensor 2 will be described with reference to Fig. 6. Fig. 6 is a plan view of the outer circumferential portion 5, the inner circumferential portion 7, the connecting portions 9A to 9D, etc., as viewed from the Z-axis positive direction. Fig. 6 shows the torque sensor 2 shown in Fig. 4 with the board units PB1 to PB4 and the connecting portions 19, 21, 23, etc. removed. To avoid complication, Fig. 6 omits the illustration of the bolt holes 13 of the inner circumferential portion 7. Fig. 6 also shows the detection portions D1 to D4 installed on the board units PB1 to PB4 with dashed lines to show their positional relationship with the scales SC1 to SC4.

[0057] As shown in FIG. 6, the outer circumferential portion 5 and the inner circumferential portion 7 are connected by connecting portions 9A-9D, which are arranged at approximately 90 degree intervals in the circumferential direction. The inner circumferential portion 7 has a plurality of (four in this example) scale fixing portions SF1-SF4 that protrude radially outward in the space S. The scale fixing portions SF1-SF4 are arranged at approximately 90 degree intervals in the circumferential direction. Scales SC1-SC4 are fixed to the upper surfaces of the scale fixing portions SF1-SF4, respectively, by, for example, screws (not shown). Although not shown, each of the scales SC1-SC4 has a plurality of reflective slits arranged in the circumferential direction.

[0058] The scale fixing part SF1 protrudes in the positive direction of the Y-axis and arranges the scale SC1 at a midpoint between two fixing positions of the substrate unit PB1 (pin holes 25, 25 of substrate fixing parts PF1, PF1 described later) in the circumferential direction. The scale fixing part SF2 protrudes in the positive direction of the X-axis and arranges the scale SC2 at a midpoint between two fixing positions of the substrate unit PB2 (pin holes 25, 25 of substrate fixing parts PF2, PF2 described later) in the circumferential direction. The scale fixing part SF3 protrudes in the negative direction of the Y-axis and arranges the scale SC3 at a midpoint between two fixing positions of the substrate unit PB3 (pin holes 25, 25 of substrate fixing parts PF3, PF3 described later) in the circumferential direction. The scale fixing part SF4 protrudes in the negative direction of the X-axis and arranges the scale SC4 at a midpoint between two fixing positions of the substrate unit PB4 (pin holes 25, 25 of substrate fixing parts PF4, PF4 described later) in the circumferential direction.

[0059] Each of the scale fixing parts SF1 to SF4 is configured as a separate piece from the inner circumferential part 7. Each of the scale fixing parts SF1 to SF4 has a base 29 in which a bolt hole 27 is formed. The base 29 is attached to and detached from a fixing portion of the inner circumferential part 7 (for example, a recess formed on the bottom surface side) by a fixing bolt (not shown) inserted through the bolt hole 27. In this way, each of the scale fixing parts SF1 to SF4 is detachably fixed to the inner circumferential part 7. The scale fixing parts SF1 to SF4 and the inner circumferential part 7 may be configured as an integral part. Also, the scale fixing parts SF1 to SF4 may be connected to form a single piece.

[0060] The outer peripheral portion 5 has a plurality of sets (four sets in this example) of substrate fixing portions PF1 to PF4 that protrude inward in the space S. Each set of substrate fixing portions PF1 to PF4, i.e., the set of substrate fixing portions PF1, PF1, the set of substrate fixing portions PF2, PF2, the set of substrate fixing portions PF3, PF3, and the set of substrate fixing portions PF4, PF4, are arranged at approximately 90 degree intervals in the circumferential direction. The substrate fixing portions PF1, PF1 are arranged to sandwich the scale fixing portion SF1 in the circumferential direction, the substrate fixing portions PF2, PF2 are arranged to sandwich the scale fixing portion SF2 in the circumferential direction, the substrate fixing portions PF3, PF3 are arranged to sandwich the scale fixing portion SF3 in the circumferential direction, and the substrate fixing portions PF4, PF4 are arranged to sandwich the scale fixing portion SF4 in the circumferential direction. On the upper surface of each set of substrate fixing portions PF1 to PF4, substrate units PB1 to PB4 having detection portions D1 to D4, respectively, are fixed at a plurality of fixing positions (two in this example) by stud pins 31, 31 (see FIG. 7 described later).

[0061] The board fixing parts PF1, PF1 each protrude in the negative direction of the Y-axis, and each has a pin hole 25 into which a stud pin 31 for fixing the board unit PB1 is press-fitted. The board fixing parts PF1, PF1 fix the board unit PB1 such that the detection part D1 is disposed facing the scale SC1 in the Z-axis direction at a midpoint between the two fixing positions (pin holes 25, 25) of the board unit PB1 in the circumferential direction. The board fixing parts PF2, PF2 each protrude in the negative direction of the X-axis, and each has a pin hole 25 into which a stud pin 31 for fixing the board unit PB2 is press-fitted. The board fixing parts PF2, PF2 fix the board unit PB2 such that the detection part D2 is disposed facing the scale SC2 in the Z-axis direction at a midpoint between the two fixing positions (pin holes 25, 25) of the board unit PB2 in the circumferential direction.

[0062] The board fixing parts PF3, PF3 each protrude in the Y-axis positive direction, and each has a pin hole 25 into which a stud pin 31 for fixing the board unit PB3 is press-fitted. The board fixing parts PF3, PF3 fix the board unit PB3 such that the detection part D3 is disposed facing the scale SC3 in the Z-axis direction at a midpoint between the two fixing positions (pin holes 25, 25) of the board unit PB3 in the circumferential direction. The board fixing parts PF4, PF4 each protrude in the X-axis positive direction, and each has a pin hole 25 into which a stud pin 31 for fixing the board unit PB4 is press-fitted. The board fixing parts PF4, PF4 fix the board unit PB4 such that the detection part D4 is disposed facing the scale SC4 in the Z-axis direction at a midpoint between the two fixing positions (pin holes 25, 25) of the board unit PB4 in the circumferential direction.

[0063] Each of the substrate fixing parts PF1 to PF4 is configured as a separate piece from the outer periphery 5. Each of the substrate fixing parts PF1 to PF4 has a base 35 in which a bolt hole 33 is formed. The base 35 is attached to and detached from a fixing portion of the outer periphery 5 (for example, a recess formed on the bottom surface side) by a fixing bolt (not shown) inserted into the bolt hole 33. In this way, each of the substrate fixing parts PF1 to PF4 is detachably fixed to the outer periphery 5. The substrate fixing parts PF1 to PF4 and the outer periphery 5 may be configured as an integral unit.

[0064] The above-described configuration is an example and is not limiting. For example, scale fixing parts SF1-SF4 may be provided on the outer circumferential part 5, and scales SC1-SC4 may be fixed to the outer circumferential part 5, and substrate fixing parts PF1-PF4 may be provided on the inner circumferential part 7, and detection parts D1-D4 may be fixed to the inner circumferential part 7.

[0065] <5. Configuration of each board unit> An example of the configuration of each of the board units PB1 to PB4 will be described with reference to Fig. 7. Fig. 7 is a cross-sectional view showing an example of the cross-sectional structure of the board units PB1 to PB4 taken along the VII-VII cross-sectional line in Fig. 4 (also shown in Fig. 6 for reference). Note that Fig. 7 omits illustration of the connection parts 19, 21, 23 and the circuit components on each board.

[0066] Since the board configurations of the board units PB1 to PB4 are similar, the board unit PB1 will be described here as an example. As shown in FIG. 7, the board unit PB1 has a plurality of boards PB1a to PB1c (three in this example). The boards PB1a to PB1c are each substantially the same shape and are arranged in multiple stages in the Z-axis direction. The boards PB1a to PB1c are supported by board fixing parts PF1, PF1 and two stud pins 31, 31 press-fitted into them with a predetermined gap therebetween in the Z-axis direction. The boards PB1a to PB1c are arranged in this order from the negative side of the Z-axis to the positive side.

[0067] The substrate PB1a is a substrate that mainly performs signal processing of analog signals. A detection unit D1 is disposed on the lower surface (surface on the Z-axis negative side) of the substrate PB1a so as to face a scale SC1 fixed to the scale fixing unit SF1 with a predetermined gap in the Z-axis direction. The detection unit D1 and the scale SC1 form a first optical sensor OS1, which detects relative displacement between the outer periphery 5 and the inner periphery 7. The substrate PB1a is provided with an amplifier circuit for the analog signal output from the detection unit D1, an offset removal circuit, and the like.

[0068] The board PB1b is a board that mainly performs signal processing of digital signals. The board PB1b is provided with an A / D conversion circuit, an arithmetic circuit that functions as a torque calculation unit 37 described later, a power supply monitoring processing circuit, and the like. The functions of these circuits may be implemented by a program executed by a CPU, or a part or all of them may be implemented by actual devices such as ASIC, FPGA, other electric circuits, and the like.

[0069] The board PB1c is a board that mainly performs processing related to communication functions. The board PB1c is provided with a power supply circuit, a communication circuit for transmitting and receiving signals between other devices and boards, and the like. In the board unit PB3 and the board unit PB4, the above-mentioned external connectors 15, 17 (shown by dashed lines in FIG. 7) are provided on the top surfaces (surfaces on the positive side of the Z axis) of the boards PB3c and PB4c.

[0070] The board configurations of the other board units PB2 to PB4 are the same as that of the board unit PB1 described above, and therefore will not be described. The scale SC2 fixed to the scale fixing portion SF2 and the detection unit D2 arranged on the board PB2a of the board unit PB2 constitute the second optical sensor OS2. Similarly, the scale SC3 fixed to the scale fixing portion SF3 and the detection unit D3 arranged on the board PB3a of the board unit PB3 constitute the third optical sensor OS3. Similarly, the scale SC4 fixed to the scale fixing portion SF4 and the detection unit D4 arranged on the board PB4a of the board unit PB4 constitute the fourth optical sensor OS4. The sensor unit 8 has the first optical sensor OS1, the second optical sensor OS2, the third optical sensor OS3, and the fourth optical sensor OS4.

[0071] The board configuration of the board units PB1 to PB4 described above is an example, and is not limited thereto. For example, if the functions of the boards described above can be implemented on a single board, each of the board units PB1 to PB4 may be configured as a single board. Furthermore, when processing is shared among a plurality of boards, it is not limited to the above-mentioned examples of sharing analog signal processing, digital signal processing, and communication processing, and may be shared among a smaller number of boards (for example, two boards), or the functions may be further subdivided and shared among a larger number of boards (four or more boards). Furthermore, a ring-shaped board may be configured by extending the board units on a circumference, and a configuration without a connection part such as an FPC may be used.

[0072] <6. Configuration of the detection unit> An example of the configuration of the detection units D1 to D4 will be described with reference to Fig. 8. Fig. 8 is a plan view of the detection units D1 to D4 as viewed from the negative direction of the Z axis.

[0073] Since the detectors D1 to D4 have the same configuration, the detector D1 will be described as an example here. As shown in Fig. 8, the detector D1 has a light source 39 and two light receiving units 41 arranged so as to sandwich the light source 39 in the radial direction centered on the Z axis.

[0074] The light source 39 emits light to the scale SC1. The light source 39 is not particularly limited as long as it is a light source capable of irradiating light to an irradiation area, and for example, an LED (Light Emitting Diode) can be used. The light source 39 is configured as a point light source in which no optical lens or the like is arranged, and emits diffused light. By using a point light source, the light source 39 can irradiate light almost evenly to the scale SC1 arranged at the opposing position. In addition, since no focusing or diffusion is performed by optical elements, errors due to optical elements are unlikely to occur, and it is possible to increase the straightness of the light to the scale SC1.

[0075] Each light receiving section 41 has a plurality of (16 in this example) light receiving elements 43 arranged at equal intervals along the circumferential direction around the Z axis. The plurality of light receiving elements 43 are formed to have an incremental pattern. The number of light receiving elements 43 constituting the light receiving section 41 may be other than the above.

[0076] The first optical sensor OS1 irradiates light from the light source 39 of the detection unit D1 onto the scale SC1, and the light reflected by the scale SC1 is received by each light receiving element 43 of the light receiving unit 41. As a result, the first optical sensor OS1 detects the amount of relative rotation between the outer circumferential portion 5 and the inner circumferential portion 7, and transmits the amount of rotation to the torque calculation unit 37 (see FIG. 10 described later).

[0077] The other detectors D2 to D4 have the same configuration as the detector D1 described above, and therefore their explanation will be omitted. The second optical sensor OS2, the third optical sensor OS3, and the fourth optical sensor OS4 also detect the amount of relative rotation between the outer circumferential portion 5 and the inner circumferential portion 7 in the same manner as described above, and transmit the amount of rotation to the torque calculation unit 37. The torque calculation unit 37 calculates the torque value acting between the outer circumferential portion 5 and the inner circumferential portion 7 based on the amount of rotation received from the optical sensors OS1 to OS4 and the elastic coefficients of the connecting portions 9A to 9D, etc.

[0078] The above-described configuration of the detection unit D1 is merely an example, and is not limited thereto. For example, in the above, the two light receiving units 41 are arranged so as to sandwich the light source 39 in the radial direction, but the two light receiving units may be arranged so as to sandwich the light source 39 in the circumferential direction.

[0079] <7. Overall configuration of the board unit> An example of the overall configuration of the board units PB1 to PB4 of the torque sensor 2 will be described with reference to Fig. 9. Fig. 9 is an explanatory diagram conceptually showing an example of the overall configuration of the board units PB1 to PB4 of the torque sensor 2.

[0080] As shown in FIG. 9, the torque sensor 2 has four board units PB1 to PB4. The boards PB1a to PB1c that make up board unit PB1 are connected by a stacking connector CN1 so as to be able to send and receive power and signals to each other. The boards PB2a to PB2c that make up board unit PB2 are connected by a stacking connector CN2 so as to be able to send and receive power and signals to each other. The boards PB3a to PB3c that make up board unit PB3 are connected by a stacking connector CN3 so as to be able to send and receive power and signals to each other. The boards PB4a to PB4c that make up board unit PB4 are connected by a stacking connector CN4 so as to be able to send and receive power and signals to each other.

[0081] The boards PB4c and PB1c are connected by a connection part 19 via stacking connectors CN4 and CN1 so that they can send and receive power and signals to each other. The boards PB1c and PB2c are connected by a connection part 21 via stacking connectors CN1 and CN2 so that they can send and receive power and signals to each other. The boards PB2c and PB3c are connected by a connection part 23 via stacking connectors CN2 and CN3 so that they can send and receive power and signals to each other. The board PB4c is provided with an external connector 17 for inputting power and sending and receiving signals to and from an external device, and the board PB3c is provided with an external connector 15 for inputting power and sending and receiving signals to and from an external device.

[0082] For ease of explanation, in the following, the board unit PB1 and the first optical sensor OS1 will be referred to as the first sensor module SM1, the board unit PB2 and the second optical sensor OS2 will be referred to as the second sensor module SM2, the board unit PB3 and the third optical sensor OS3 will be referred to as the third sensor module SM3, and the board unit PB4 and the fourth optical sensor OS4 will be referred to as the fourth sensor module SM4.

[0083] The first sensor module SM1 and the third sensor module SM3 arranged at an angular interval of 180 degrees in the circumferential direction constitute a first system, with the third sensor module SM3 functioning as the master and the first sensor module SM1 functioning as the slave. The second sensor module SM2 and the fourth sensor module SM4 arranged at an angular interval of 180 degrees in the circumferential direction constitute a second system, with the fourth sensor module SM4 functioning as the master and the second sensor module SM2 functioning as the slave.

[0084] In the first system, the first sensor module SM1 transmits the detection value by the first optical sensor OS1 to the third sensor module SM3 via the second sensor module SM2, and the third sensor module SM3 calculates an average value by adding the detection value by the first optical sensor OS1 received to the detection value by the third optical sensor OS3. In the second system, the second sensor module SM2 transmits the detection value by the second optical sensor OS2 to the fourth sensor module SM4 via the first sensor module SM1, and the fourth sensor module SM4 calculates an average value by adding the detection value by the second optical sensor OS2 received to the detection value by the fourth optical sensor OS4. When only canceling the detection error due to other-axis interference (described in detail later), the third sensor module SM3 calculates a torque value using the calculated average value and transmits it to the outside via the external connector 15. Also, the fourth sensor module SM4 calculates a torque value using the calculated average value and transmits it to the outside via the external connector 17.

[0085] On the other hand, in addition to canceling the detection error due to interference from other axes, when using the torque sensor 2 in combination with a reducer of a strain wave gear mechanism, the detection error due to torque ripple can be canceled (details will be described later) as follows, for example. That is, either the third sensor module SM3 or the fourth sensor module SM4 transmits the above-mentioned calculated average value to the other sensor module, and the other sensor module receiving the average value adds these average values ​​to calculate a further average value. Then, the calculated average value is used to calculate a torque value, and the calculation result is transmitted to the outside via the external connector 15 or the external connector 17.

[0086] In the above, the calculation of the average value in each system is performed by the master, but this is not limited to this. For example, the calculation of the average value in each system may be performed by the slave, and the average values ​​may be transmitted to the master of one system, and the master may calculate the average of the average values ​​to calculate the torque value. Also, for example, the calculation of the average value in one system may be performed by the master, and the calculation of the average value in the other system may be performed by the slave, and the average value may be transmitted from the slave to the master of one system, and the master may calculate the average of the average values ​​to calculate the torque value.

[0087] As described above, by configuring the transmission and reception of signals between the master and slave in one system via the slave in the other system, the wiring can be simplified compared to when the master and slave of each system are directly connected. Also, since the third sensor module SM3 and the fourth sensor module SM4 are not connected, this also promotes wiring reduction.

[0088] The above-mentioned board configuration is an example, and is not limited thereto. For example, the third sensor module SM3 and the fourth sensor module SM4 may be connected by a connection portion, and the board units PB1 to PB4 may be connected in a circular ring shape. Also, for example, an external connector may be provided in the slave instead of or in addition to the master in each system, or may be provided in at least one of the master and the slave in only one of the systems.

[0089] <8. Torque calculation process> An example of the torque calculation process will be described with reference to Fig. 10. Fig. 10 is a block diagram showing an example of the functional configuration of the torque calculation unit 37 that executes the torque calculation process. The process by the torque calculation unit 37 shown in Fig. 10 may be shared and executed by all or some of the substrates PB1b, PB2b, PB3b, and PB4b, or may be executed by only one specific substrate.

[0090] The torque calculation unit 37 calculates a torque value based on the outputs of a plurality of optical sensors (four optical sensors OS1 to OS4 in this example). As shown in FIG. 10, the torque calculation unit 37 has conversion units 56, 58, 60, and 62, an adder unit 57, a divider unit 59, an adder unit 61, a divider unit 63, an adder unit 65, and a divider unit 67.

[0091] Each of the conversion units 56, 58, 60, 62 calculates the torque value acting between the outer circumferential portion 5 and the inner circumferential portion 7 based on the output (detection value) of each of the first optical sensor OS1, the second optical sensor OS2, the third optical sensor OS3, and the fourth optical sensor OS4, as well as the elastic coefficients, etc. of the connecting portions 9A to 9D.

[0092] The adding unit 57 adds the torque value calculated by the converting unit 56 and the torque value calculated by the converting unit 60. The dividing unit 59 divides the torque value added by the adding unit 57 by 2 to calculate an average value.

[0093] The addition unit 61 adds the torque value calculated by the conversion unit 58 and the torque value calculated by the conversion unit 62. The division unit 63 divides the torque value added by the addition unit 61 by 2 to calculate an average value.

[0094] As described above, by adding the outputs (including the torque value calculated by the conversion unit) of the optical sensors OS1 and OS3 arranged at 180 degree intervals in the circumferential direction and the outputs (including the torque value calculated by the conversion unit) of the optical sensors OS2 and OS4, the detection error due to the interference with other axes can be canceled. Here, the "interference with other axes" refers to the torsion moment Mx around the X-axis, the torsion moment My around the Y-axis, and the translational forces Fx, Fy, and Fz in the directions of the X-axis, Y-axis, and Z-axis. In addition to the rotation around the Z-axis due to the torque (torsion moment Mz around the Z-axis) to be detected, a relative rotation around the X-axis or Y-axis may occur between the outer peripheral portion 5 and the inner peripheral portion 7 of the torque sensor 2 due to a disturbance force other than the torque (torsion moment Mx, My). In addition, a relative movement in the directions of the X-axis, Y-axis, and Z-axis may occur due to a disturbance force other than the torque (translational forces Fx, Fy, and Fz).

[0095] For example, when a relative rotation occurs between the outer peripheral portion 5 and the inner peripheral portion 7 around the X-axis, in the optical sensors OS1 and OS3 constituting the first system, the interval between the scales SC1 and SC3 and the detection units D1 and D3 varies, but since a parallel positional relationship is maintained with respect to the circumferential direction around the Z-axis, the change in the detection position in the circumferential direction around the Z-axis (amount of rotation) is zero or very small. Therefore, the effect on the torque detection accuracy can be ignored. On the other hand, in the optical sensors OS2 and OS4 constituting the second system, an inclination occurs between the scales SC2 and SC4 and the detection units D2 and D4 with respect to the circumferential direction around the Z-axis, so that the detection position in the circumferential direction changes and a detection error occurs. However, in each of the optical sensors OS2 and OS4, the detection error occurs in the opposite direction (positive direction and negative direction) in the circumferential direction around the Z-axis, so that the detection error due to the torsion moment Mx can be canceled by adding the outputs of the optical sensors OS2 and OS4.

[0096] Similarly, when relative rotation occurs around the Y-axis between the outer peripheral portion 5 and the inner peripheral portion 7, the effect on the torque detection accuracy of the optical sensors OS2 and OS4 constituting the second system can be ignored. On the other hand, in the optical sensors OS1 and OS3 constituting the first system, an inclination occurs between the scales SC1 and SC3 and the detection portions D1 and D3 in the circumferential direction around the Z-axis, so the detection position in the circumferential direction changes and a detection error occurs. However, in each of the optical sensors OS1 and OS3, the detection error occurs in opposite directions (positive and negative directions) in the circumferential direction around the Z-axis, so the detection error due to the torsional moment My can be canceled out by adding the outputs of the optical sensors OS1 and OS3.

[0097] In addition, when a relative rotation other than around the X-axis and the Y-axis occurs between the outer peripheral portion 5 and the inner peripheral portion 7 on the XY-axis plane, in the first system, the effect on the torque detection accuracy for the component around the X-axis can be ignored as described above, and the detection error for the component around the Y-axis can be canceled by adding the outputs of the optical sensors OS1 and OS3. In the second system, the effect on the torque detection accuracy for the component around the Y-axis can be ignored as described above, and the detection error for the component around the X-axis can be canceled by adding the outputs of the optical sensors OS2 and OS4. Therefore, by adding the outputs of the optical sensors OS1 and OS3 and the outputs of the optical sensors OS2 and OS4, respectively, the torque sensor 2 can cancel the detection error even when a relative rotation occurs around any axis in the XY-axis plane.

[0098] Furthermore, for example, when a relative movement occurs between the outer peripheral portion 5 and the inner peripheral portion 7 in the X-axis direction, the change (amount of rotation) in the detection position in the circumferential direction around the Z axis between the scales SC2, SC4 and the detection units D2, D4 in the optical sensors OS2, OS4 constituting the second system is zero or very small, so the effect on the detection accuracy of the torque can be ignored. On the other hand, in the optical sensors OS1, OS3 constituting the first system, a detection error occurs because the detection position in the circumferential direction around the Z axis changes between the scales SC1, SC3 and the detection units D1, D3. However, in each of the optical sensors OS1, OS3, the detection error occurs in opposite directions (positive and negative directions) in the circumferential direction around the Z axis, so the detection error due to the translational force Fx can be canceled by adding the outputs of the optical sensors OS1, OS3.

[0099] Similarly, when a relative movement occurs between the outer peripheral portion 5 and the inner peripheral portion 7 in the Y-axis direction, the change (amount of rotation) in the detection position in the circumferential direction around the Z-axis between the scales SC1, SC3 and the detection units D1, D3 is zero or very small in the optical sensors OS1, OS3 constituting the first system, so the effect on the torque detection accuracy can be ignored. On the other hand, in the optical sensors OS2, OS4 constituting the second system, a detection error occurs because the detection position in the circumferential direction around the Z-axis changes between the scales SC2, SC4 and the detection units D2, D4. However, in each of the optical sensors OS2, OS4, the detection error occurs in the opposite direction (positive direction and negative direction) in the circumferential direction around the Z-axis, so the detection error due to the translational force Fy can be canceled by adding the outputs of the optical sensors OS2, OS4.

[0100] In addition, when a relative movement occurs between the outer peripheral portion 5 and the inner peripheral portion 7 in a direction other than the X-axis direction and the Y-axis direction on the XY-axis plane, in the first system, the effect on the torque detection accuracy for the Y-axis direction component can be ignored as described above, and the detection error for the X-axis direction component can be canceled by adding the outputs of the optical sensors OS1 and OS3. In the second system, the effect on the torque detection accuracy for the X-axis direction component can be ignored as described above, and the detection error for the Y-axis direction component can be canceled by adding the outputs of the optical sensors OS2 and OS4. Therefore, by adding the outputs of the optical sensors OS1 and OS3 and the outputs of the optical sensors OS2 and OS4, the torque sensor 2 can cancel the detection error no matter in which direction a relative movement occurs on the XY-axis plane.

[0101] Furthermore, when relative movement occurs in the Z-axis direction between the outer circumferential portion 5 and the inner circumferential portion 7, in any of the optical sensors OS1 to OS4, the change in the detection position (amount of rotation) in the circumferential direction around the Z-axis between the scales SC1 to SC4 and the detection portions D1 to D4 is zero or very small, so the impact on the torque detection accuracy can be ignored.

[0102] The adder 65 adds together the average value divided by the divider 59 and the average value divided by the divider 63. In other words, the adder 65 adds together the output (average value) of the first system and the output (average value) of the second system arranged at 90 degrees offset from each other in the circumferential direction. The divider 67 divides the average value added by the adder 65 by 2, and further calculates the average value of the average values.

[0103] As described above, by adding the output (average value) of the first system and the output (average value) of the second system arranged at 90 degrees offset from each other in the circumferential direction, when the torque sensor 2 is used in combination with the reducer 4 having a strain wave gear mechanism, it is possible to cancel a detection error due to torque ripple specific to the reducer 4. The strain wave gear mechanism has a circular spline 403, a flexspline 405, and a wave generator 407. The flexspline 405 is bent into an ellipse by the wave generator 407, and the teeth of the flexspline 405 mesh with the circular spline 403 at the major axis portion and the teeth of the flexspline 405 are separated at the minor axis portion. When either the circular spline 403 or the flexspline 405 is fixed and the wave generator 407 is rotated clockwise, the flexspline 405 is elastically deformed and the meshing position of the teeth of the flexspline 405 with the circular spline 403 moves sequentially. When the wave generator 407 rotates once, if the circular spline 403 is fixed, the flexspline 405 moves in the counterclockwise direction by the difference in the number of teeth (for example, 2) between the circular spline 403 and the flexspline 405. If the flexspline 405 is fixed, the circular spline 403 moves in the clockwise direction by the difference in the number of teeth (for example, 2).

[0104] With the above-described configuration of the strain wave gear mechanism, the meshing position of the teeth of the flexspline 405 and the circular spline 403 passes twice while the wave generator 407 makes one rotation (360 degrees). As a result, as shown in FIG. 11, two periods of torque ripple are generated per one rotation of the wave generator 407, which may appear as a detection error of the torque sensor 2. The detection error is caused by the torque ripple and has a period of 180 degrees. Meanwhile, the optical sensors OS1 and OS3 of the first system and the optical sensors OS2 and OS4 of the second system are arranged with a 90 degree circumferential shift, so that the detection errors of the respective outputs have a phase difference of 90 degrees. Therefore, the torque sensor 2 can cancel the detection error caused by the torque ripple that is specific to a reducer equipped with a strain wave gear mechanism by adding the output (average value) of the first system and the output (average value) of the second system.

[0105] The torque calculation process described above is an example, and is not limited thereto. For example, when a reducer without a strain wave gear mechanism is used, or when only canceling detection errors due to other axis interference is to be performed, the average value divided by the divider 59 and the average value divided by the divider 63 may be output as independent torque values ​​for the first and second systems without providing the adder 65 and the divider 67. In this case, the sensor output can be duplicated while canceling detection errors due to other axis interference.

[0106] Alternatively, the adder 57 and the divider 59 may add the output of the first optical sensor OS1 and the output of the second optical sensor OS2 (or the output of the second optical sensor OS2 and the output of the third optical sensor OS3) arranged at a 90 degree difference in the circumferential direction to calculate an average value, the adder 61 and the divider 63 may add the output of the third optical sensor OS3 and the output of the fourth optical sensor OS4 (or the output of the fourth optical sensor OS4 and the output of the first optical sensor OS1) arranged at a 90 degree difference in the circumferential direction to calculate an average value, and the adder 65 and the divider 67 may calculate the average value. In this case, too, it is possible to cancel the detection error due to interference with other axes and to cancel the detection error due to the torque ripple that is unique to the strain wave gear mechanism. In this case, when only the detection error due to the torque ripple is to be canceled, the adder 65 and the divider 67 may not be provided, and the average value divided by the divider 59 and the average value divided by the divider 63 may be output as independent torque values. In this case, the sensor output can be doubled while canceling detection errors caused by torque ripple.

[0107] The above-mentioned processes in the conversion units 56, 58, 60, 62, the addition unit 57, the division unit 59, the addition unit 61, the division unit 63, the addition unit 65, the division unit 67, etc. are not limited to the examples of division of processes, and may be processed by a smaller number of processing units (for example, one processing unit), or may be processed by further subdivided processing units. Furthermore, these functions may be implemented by a program executed by a CPU (not shown) provided on the boards PB1b, PB2b, PB3b, PB4b, or part or all of them may be implemented by actual devices such as ASIC, FPGA, other electric circuits, etc.

[0108] <9. Damping effect of the damping section> An example of the damping effect of the damping section 6 will be described with reference to Figs. 12 to 15. Fig. 12 is a graph showing an example of the measurement result of the output error of the first system and the output error of the second system in a comparative example in which a damping section is not provided. Fig. 13 is a graph showing an example of the measurement result of the output error of the average value of the output of the first system and the output of the second system in a comparative example in which a damping section is not provided. Fig. 14 is a graph showing an example of the measurement result of the output error of the first system and the output error of the second system in an embodiment in which a damping section is provided. Fig. 15 is a graph showing an example of the measurement result of the output error of the average value of the output of the first system and the output of the second system in an embodiment in which a damping section is provided. Note that Figs. 14 and 15 show the measurement results when a metal gasket is used as the damping section 6.

[0109] As shown in Fig. 12, when the damping unit 6 is not provided, the output of the first system and the output of the second system each have an output error of about 3% due to the torque ripple of the reducer 4 equipped with a strain wave gear mechanism. As shown in Fig. 13, when the damping unit 6 is not provided, the output error due to the torque ripple can be reduced by taking the average of the output of the first system and the output of the second system, but an error of about 1% remains.

[0110] As shown in Fig. 14, when the attenuation unit 6 is provided, the attenuation unit 6 significantly reduces the effect of the torque ripple caused by the reducer 4, and the output of the first system and the output of the second system are each reduced to an output error of approximately 0.8%. As shown in Fig. 15, when the attenuation unit 6 is provided, by taking the average of the output of the first system and the output of the second system, the output error caused by the torque ripple is approximately 0.2%, and the effect of the torque ripple caused by the reducer can be almost eliminated.

[0111] <10. Effects of the embodiment> As described above, the robot 1 of this embodiment has a reducer 4 that increases the first torque output from the motor 3 and outputs a second torque, an outer circumferential portion 5 that outputs the second torque from the reducer 4, a torque sensor 2 having a sensor unit 8 that detects at least the torque applied to the outer circumferential portion 5, and a damping unit 6 that is located in the transmission path of the second torque from the reducer 4 to the torque sensor 2 and damps the transmission of the second torque.

[0112] If the torque sensor 2 is disposed near the reducer 4, vibrations specific to the reducer may be detected as torque ripples, resulting in reduced torque detection accuracy. According to this embodiment, the transmission of the second torque is attenuated in the transmission path of the second torque from the reducer 4 to the torque sensor 2. This makes it possible to reduce torque ripples without changing the magnitude of the second torque itself transmitted from the reducer 4 to the torque sensor 2. As a result, the influence of vibrations of the reducer 4 can be reduced, thereby improving torque detection accuracy.

[0113] In this embodiment, the sensor unit 8 may have a first optical sensor OS1, a second optical sensor OS2, a third optical sensor OS3, and a fourth optical sensor OS4 that optically detect torque.

[0114] In this case, the sensor unit 8 becomes an optical detection system with high sensitivity, and the effect of ripples caused by the reducer 4 becomes prominent. Therefore, by suppressing this effect, it is possible to achieve highly accurate torque detection using a highly sensitive detection system.

[0115] In this embodiment, the reducer 4 may have a wave generator 407 that receives a first torque output from the motor 3, a circular spline 403 and a flex spline 405 that increase the first torque received by the wave generator 407, and a circular spline 403 that outputs the increased second torque, the torque sensor 2 may have an inner circumferential portion 7 that receives the second torque output from the circular spline 403, an outer circumferential portion 5 that outputs the second torque received by the inner circumferential portion 7, and connecting portions 9A to 9D that connect the inner circumferential portion 7 and the outer circumferential portion 5, the optical sensors OS1 to OS4 may detect the torque based on the relative displacement between the inner circumferential portion 7 and the outer circumferential portion 5, and the damping portion 6 may be located in the transmission path of the second torque from the circular spline 403 to the inner circumferential portion 7.

[0116] In this case, the transmission of the second torque can be attenuated in the transmission path of the second torque from the circular spline 403 of the reducer 4 to the inner peripheral portion 7 of the torque sensor 2. This reduces the torque ripple and reduces the influence of vibration of the reducer 4, thereby improving the torque detection accuracy.

[0117] In this embodiment, the damping portion 6 is interposed between the circular spline 403 and the inner circumferential portion 7, and may be a member formed separately from the reducer 4 and the torque sensor 2.

[0118] In this case, a member formed separately from the reducer 4 and the torque sensor 2 can be interposed as the damping section 6 between the circular spline 403 of the reducer 4 and the inner periphery 7 of the torque sensor 2. By adopting a structure in which a separate member is interposed between the reducer 4 and the torque sensor 2 in this way, a simple structure can be achieved, and it becomes easy to add functions by retrofitting to an existing structure. As a result, it becomes possible to improve the performance of an existing production line in a short period of time.

[0119] In this embodiment, the damping portion 6 may be interposed between the first transmission surface 421 of the circular spline 403 that transmits the second torque and the second transmission surface 215 of the inner circumferential portion 7 that transmits the second torque.

[0120] In this case, the second torque is transmitted to the damping unit 6 via the first transmission surface 421 of the circular spline 403 of the reducer 4, and the second torque is transmitted to the inner circumferential portion 7 of the torque sensor 2 via the second transmission surface 215 of the inner circumferential portion 7 of the torque sensor 2. In this way, the damping unit 6 can be interposed in the transmission path of the second torque, so that the influence of vibration of the reducer 4 can be reduced more reliably.

[0121] In this embodiment, the robot 1 may further have a bolt 201 that fixes the inner circumference 7 of the torque sensor 2 to the circular spline 403 of the reducer 4, and in that case, the damping part 6 may be interposed between the first transmission surface 421 and the second transmission surface 215 at a position fixed by the bolt 201.

[0122] In this case, the damping unit 6 can be firmly in contact with the first transmission surface 421, and the damping unit 6 can be firmly in contact with the second transmission surface 215. This improves the reliability of the transmission of the second torque between the damping unit 6 and the first transmission surface 421 and the second transmission surface 215. As a result, the effectiveness of the function of the damping unit 6 can be improved, and the influence of vibration of the reducer 4 can be more reliably reduced. Furthermore, the damping unit 6 can be positioned and fixed to both transmission surfaces 421, 215 without any additional parts for positioning and fixing the damping unit 6.

[0123] In this embodiment, the damping portion 6 may be a shim member having elasticity and sandwiched between the first transmission surface 421 and the second transmission surface 215 .

[0124] In this case, the torque ripple can be reduced by elastic deformation of the shim member, and therefore the influence of vibration of the reduction gear 4 can be more reliably reduced. In addition, by using a metal gasket, for example, as the shim member, it is possible to provide a structure with excellent sealing properties and heat resistance.

[0125] <11. Variations> The disclosed embodiment is not limited to the above, and various modifications are possible without departing from the spirit and technical concept of the present invention. Such modifications will be described below.

[0126] (11-1. When the damping section is formed as a part of the output section of the reducer) In the above embodiment, the damping unit 6 is a member formed separately from the reducer 4 and the torque sensor 2, but the damping unit is not limited to being a separate member. For example, the damping unit may be formed as a part of the output unit of the reducer.

[0127] With reference to FIG. 16, an example of the configuration of the damping section of this modified example will be described. In FIG. 16, the same components as those in FIG. 2 are denoted by the same reference numerals, and the description will be omitted. As shown in FIG. 16, the damping section 6A in this modified example is formed as a part of the circular spline 403 (an example of an output section) of the reducer 4. The damping section 6A is a ring-shaped part formed on the tip side (hand 120R side) of the circular spline 403. The damping section 6A only needs to have a property capable of damping the transmission of the second torque from the reducer 4 to the torque sensor 2, and the material, shape, and the like are not particularly limited. For example, the damping section 6A may be formed as a part having a lower Young's modulus than other parts of the circular spline 403. In this case, for example, the damping section 6A may be formed of a material that is softer than other parts. In addition to or instead of this, for example, the damping section 6A may be formed to have a structure that is softer than other parts by forming a plurality of recesses or protrusions in the damping section 6A.

[0128] In this modification, the surface on the tip side of the damping portion 6A serves as a first transmission surface 421, which is in contact with and fixed to the second transmission surface 215 of the inner circumferential portion 7 of the torque sensor 2 by the bolt 201.

[0129] According to this modification, the damping portion 6A is formed as a part of the circular spline 403 of the reducer 4. In this case, the transmission of the second torque can be damped by the circular spline 403 of the reducer 4 itself. This eliminates the need to interpose a separate member between the reducer 4 and the torque sensor 2, and therefore the drive mechanism 10 or the robot 1 that can reduce the influence of vibration of the reducer 4 and improve the torque detection accuracy can be realized with a compact structure. In addition, the number of parts can be reduced, making assembly easier. Furthermore, by integrating the damping portion 6A with the reducer 4, the relative positional accuracy between the reducer 4 and the damping portion 6A is improved, and vibration can be reduced more reliably.

[0130] In this modification, the damping portion 6A may be formed as a portion having a lower Young's modulus than the other portions. In this case, since the damping portion 6A has a lower Young's modulus than the other portions, the portion itself can be elastically deformed to reduce torque ripple. Therefore, the influence of vibration of the reduction gear 4 can be more reliably reduced.

[0131] (11-2. When the damping section is formed as a part of the input section of the torque sensor) In the above embodiment, the damping unit 6 is a member formed separately from the reducer 4 and the torque sensor 2, but the damping unit is not limited to being a separate member. For example, the damping unit may be formed as a part of the input unit of the torque sensor.

[0132] With reference to FIG. 17, an example of the configuration of the damping section of this modified example will be described. In FIG. 17, the same components as those in FIG. 2 are denoted by the same reference numerals, and the description will be omitted. As shown in FIG. 17, the damping section 6B in this modified example is formed as a part of the inner circumferential section 7 (one example of an input section) of the torque sensor 2. The damping section 6B is a ring-shaped section formed on the base end side (opposite side to the hand 120R) of the inner circumferential section 7. The damping section 6B only needs to have a property capable of damping the transmission of the second torque from the reducer 4 to the torque sensor 2, and the material, shape, and the like are not particularly limited. For example, the damping section 6B may be formed as a section having a lower Young's modulus than the other sections of the inner circumferential section 7. In this case, for example, the damping section 6B may be formed of a material that is softer than the other sections. In addition to or instead of this, for example, the damping section 6B may be formed to have a structure that is softer than the other sections by forming a plurality of recesses or protrusions in the damping section 6B.

[0133] In this modification, the surface on the base end side of the damping portion 6B serves as the second transmission surface 215, which is in contact with and fixed to the first transmission surface 421 of the circular spline 403 of the reducer 4 by the bolt 201.

[0134] According to this modification, the damping portion 6B is formed as a part of the inner circumferential portion 7 of the torque sensor 2. In this case, the transmission of the second torque can be damped by the inner circumferential portion 7 of the torque sensor 2 itself. This eliminates the need to interpose a separate member between the reducer 4 and the torque sensor 2, and thus the drive mechanism 10 or the robot 1 that can reduce the influence of vibration of the reducer 4 and improve the torque detection accuracy can be realized with a compact structure. In addition, the number of parts can be reduced, making assembly easier. Furthermore, by integrating the damping portion 6B with the torque sensor 2, the relative positional accuracy between the torque sensor 2 and the damping portion 6B is improved, and vibration can be reduced more reliably.

[0135] In this modification, the damping portion 6B may be formed as a portion having a lower Young's modulus than the other portions. In this case, since the damping portion 6B has a lower Young's modulus than the other portions, the portion itself can be elastically deformed to reduce torque ripple. Therefore, the influence of vibration of the reduction gear 4 can be more reliably reduced.

[0136] (11-3.Others) The problems and effects that the above-described embodiments and modifications are intended to solve are not limited to those described above. That is, the embodiments and modifications may solve problems or achieve effects that are not described above, or may solve only some of the problems or achieve only some of the effects that are described.

[0137] In the above description, when terms such as "vertical," "parallel," and "plane" are used, the terms are not used in their strict sense. In other words, "vertical," "parallel," and "plane" mean "substantially vertical," "substantially parallel," and "substantially plane," allowing for design and manufacturing tolerances and errors.

[0138] In addition, in the above description, when the external dimensions, size, shape, position, etc. are described as "same", "equal", "different", etc., the description does not mean in the strict sense. In other words, "same", "equal", "different" means "substantially the same", "substantially the same", "substantially equal", or "substantially different", allowing for design and manufacturing tolerances and errors.

[0139] In addition to the above, the methods according to the above-mentioned embodiments and each modified example may be appropriately combined and used. Although not illustrated individually, the above-mentioned embodiments and each modified example may be implemented with various modifications without departing from the spirit thereof. [Explanation of symbols]

[0140] 1. Robot 2 Torque Sensor 3 Motor 4 Reducer 5 Outer periphery (output section, second output section) 6 Damping section 6A Attenuation section 6B Damping section 7 Inner circumference (input section, second input section) 8 Sensor section 9A~9D Connecting part 10 Drive mechanism 201 Volts 215 Second Transmission Surface 403 Circular spline (reduction section, output section, first output section) 405 Flexspline (reduction section) 407 Wave Generator (First Input) 421 First Transmission Surface OS1 1st optical sensor OS2 Second optical sensor OS3 3rd optical sensor OS4 4th optical sensor

Claims

1. a reducer having a first output unit that increases a first torque output from a motor and outputs a second torque; a torque sensor including: a second input section that receives the second torque output from the first output section; a second output section that outputs the second torque received by the second input section; and a sensor section that detects at least the torque applied to the second output section; a damping unit located in a transmission path of the second torque from the first output unit to the second input unit and configured to damp the transmission of the second torque; A robot having the above configuration.

2. The sensor unit includes: An optical sensor that optically detects the torque. The robot according to claim 1.

3. The reducer includes: a first input unit that inputs a first torque output from the motor; a speed reducer that increases the first torque input by the first input unit; the first output unit that outputs the second torque increased by the speed reducer; having The torque sensor is a coupling portion coupling the second input portion and the second output portion; The optical sensor includes: detecting the torque based on a relative displacement between the second input portion and the second output portion; The robot according to claim 2.

4. The attenuation portion is A member interposed between the first output portion and the second input portion and formed separately from the reducer and the torque sensor. The robot according to claim 3.

5. The member is a first transmission surface that transmits the second torque of the first output portion and a second transmission surface that transmits the second torque of the second input portion; The robot according to claim 4.

6. a fixing member that fixes the second input portion to the first output portion, The member is The fixing member is disposed between the first transmission surface and the second transmission surface, and is fixed by the fixing member. The robot according to claim 5.

7. The member is a shim member having elasticity and sandwiched between the first transmission surface and the second transmission surface; The robot according to claim 5 or 6.

8. The attenuation portion is formed as part of the first output section; The robot according to claim 3.

9. The attenuation portion is formed as part of the second input section; The robot according to claim 3.

10. The attenuation portion is The Young's modulus is lower than that of other parts. The robot according to claim 8 or 9.

11. a reducer having a first output unit that increases a first torque output from a motor and outputs a second torque; a torque sensor including: a second input section that receives the second torque output from the first output section; a second output section that outputs the second torque received by the second input section; and a sensor section that detects at least the torque applied to the second output section; a damping unit located in a transmission path of the second torque from the first output unit to the second input unit and configured to damp the transmission of the second torque; A drive mechanism having

12. a first output section that increases a first torque output from the motor and outputs a second torque to a second input section of the torque sensor; a damping section formed as a part of the first output section, located in a transmission path of the second torque from the first output section to the second input section, and damping the transmission of the second torque; A reducer having

13. a second input section for inputting a torque output from the first output section of the reducer; a second output unit that outputs the torque inputted by the second input unit; A sensor unit that detects at least a torque applied to the second output unit; a damping section formed as a part of the second input section, located in a torque transmission path from the first output section to the second input section, and configured to damp the transmission of the torque; The torque sensor has

14. a first torque output from the motor is increased by a reducer to output a second torque from a first output portion of the reducer; inputting the second torque output from the first output portion through a second input portion of a torque sensor, and outputting the second torque input through the second input portion through a second output portion of the torque sensor; Detecting at least the torque applied to the second output section by a sensor section of the torque sensor; Attenuating transmission of the second torque in a transmission path of the second torque from the first output portion to the second input portion; A torque detection method comprising:

Images