JPH04190691A - Motor controller for fixed-tension suspended mechanism - Google Patents

Abstract

PURPOSE:To constitute the title motor controller in a winding type by calculating the torque command value of the drive motor of a winding mechanism which reduces the tension of the mechanism by the fluctuated amount of the detected output of a tension sensor on the basis of the detected output of the sensor. CONSTITUTION:A control circuit 106 reduces the fluctuation of the tension on a suspension wire 103 and maintains the wire 103 at a fixed tension level by generating a torque command value on the basis of the detected output of a tension sensor 104 and driving a drive motor 105 on the basis of the torque command value. A statical friction offset generating means 107 generates a statical friction offset and eliminates the influence of statical friction when the motor 105 is started from a stopped state by adding the statical friction offset to the detected output of the sensor 104. In addition, a kinetic friction offset generating means 108 generates a kinetic friction offset and eliminates the influence of kinetic friction when the motor 105 is rotated by adding the kinetic friction offset to the detected output of the sensor 104.

Description

[Detailed description of the invention] [Table of contents] Overview Industrial field of application Conventional technology Problems to be solved by the invention Means for solving the problems (Fig. 1) Working examples C Figs. 2 to 46) 1. Basic method of zero-gravity simulation system 1. Offline zero gravity simulation experiment system 11, 1.1. Overall configuration a, simulation data creation department, multi-jointed structure (space robot) C0 zero gravity simulator 1.1.2. Element parts a, constant tension suspension mechanism part ■, vertical movement mechanism a, suspension tension adjustment method.

Mouth, hanging position and gimbal, counterbalance server, wire winding mechanism part 2, tension sensor and control device, operating mechanism in horizontal plane (hanging arm) ■1 Main body support drive part, etc. 1.1. :1. Offline data creation method 1.1.4. Cooperative control operation based on online data 1.2゜Offline zero gravity simulation experiment system ■1.2.
1. Overall configuration 1.2.1 Element part a, angle detection sensor ■, optical angle detection sensor ■1 component force detection sensor 1.2.3. Control operation 1.3. Online zero-gravity simulation system 11.3.
1. Overall configuration 1.3.2. Element part a, robot-based reaction detection mechanism 1.3.3. Control operation 1.4. Online zero-gravity simulation system■1.4.
1. Overall configuration 1, L, 2. Control operation 2, hanging method 2.19 Center of gravity vertical hanging method 2.2. Static balance method 2.3. Compromise method 3, control of suspension arm 3.1 Algorithm for suspension direction vertical maintenance method 3.28 Constant tension control by winding mechanism section 3.2.1. Torque control method ■ a. Overall configuration, configuration of each element C8 Control method 3.2.2. Torque control method ■ a. Overall configuration, control method 3.2.3 Torque control method ■ a. Overall configuration, control method 3.2.4) Torque control method rv a. Overall configuration, control method 3.2 .5 Torque control method ■ a. Overall configuration, control method 3.2.6 Torque control method ■ a. Overall configuration, control method 3.2.7. Speed control method % type % 3.3. Specific configuration of control circuit 4, simulation computer system 4.1. System configuration I 4.2. System configuration ■ Effects of the invention [Summary] Motor control of a constant tension suspension mechanism used in zero gravity simulation experiment equipment, etc. that simulates the behavior of multi-jointed structures such as space robots and large satellites in zero gravity space on the ground using hardware. Regarding the device, the purpose of the present invention is to provide a motor control device for a constant tension suspension mechanism with a simple configuration using a winding mechanism. In the constant tension suspension mechanism, which maintains the tension of the hanging wire at a constant level by unwinding the hanging wire seven times, there is a tension sensor that detects the tension applied to the hanging wire, and the variation in the detected output is reduced based on the detected output of the tension sensor. and a control circuit that calculates a torque command value for the drive motor of the winding mechanism, and is configured to apply the torque command value to the drive motor.

[Industrial Application Field] The present invention relates to a constant tension suspension mechanism used in a zero-gravity simulation experiment device, etc., which simulates the behavior of multi-jointed structures such as space robots and large satellites in zero-gravity space using a ring on the ground. The present invention relates to a motor control device.

゛ In a zero-gravity space, for example, if a robot tries to capture an object by extending its arm, the extension of the arm shifts the center of gravity of the entire system, changing the posture of the robot body and causing it to be unable to capture the object. can occur.

In addition, the reaction of the impact at the time of capture cannot be ignored.

To deal with such difficult-to-predict behavior in zero-gravity space, we have created an environment in which we can directly conduct simulation experiments using hardware in a laboratory on the ground. It is necessary to be able to perform sufficient evaluation in advance in an environment close to space, which enables highly reliable hardware configuration.

[Prior Art 1] Conventionally, the following methods have been devised and implemented as methods for directly simulating the behavior of multi-joint structures in zero-gravity space using hardware on the ground.

■Simulate two-dimensional weightless motion in a plane on a surface plate using air levitation or magnetic levitation.

■Put an object in a water tank, use buoyancy to compensate for gravity, and simulate three-dimensional weightless motion.

■Creating a zero-gravity state using an aircraft's suborbital flight and moving objects within it.

In the above ■, the motion of the object is limited to a two-dimensional plane, ■
In case 2, it is not possible to simulate operation due to viscous resistance, and in case 2, simulation is limited to a short time. There is a drawback.

Therefore, a method has been proposed in which the object is suspended with a wire and the tension applied to the wire cancels out the object's gravity to create a simulated weightless state on the ground. This method requires that the object be suspended by a wire with constant tension.

Various methods can be considered for suspending the target object. For example, as shown in Figure 8, there is a balloon method in which the arm is suspended from a balloon, a counterbalance method in which the arm is suspended by a counterbalancer, and a counterbalance method in which the arm is suspended by a counter balancer. There are constant force spring types (Conston springs) in which a wire with a suspended arm is wound around a wire winding drum and unwinded, and a winding type in which the arm is suspended using a wire winding drum and a winding motor.

[Problem to be solved by the invention] When attempting to simulate weightlessness by suspending a multi-jointed structure with a complex structure, such as a space robot, using a wire, the mechanism for suspending the wire must be large-scale. It tends to be. Therefore, in order to realize a zero-gravity simulation experiment device such as a multi-jointed structure, a constant tension suspension mechanism with a compact configuration is required. For this purpose, a winding method that can easily respond to changes in the suspension gravity and aim for compactness and high precision is suitable for a constant tension suspension mechanism.

The present invention has been made in view of the above circumstances, and its purpose is to provide a motor control device for use in a constant tension suspension mechanism using a winding method. By using this, it becomes possible to realize a zero-gravity simulation experiment device for multi-jointed structures.

[Means to solve the problem] FIG. 1 is a diagram explaining the principle of the present invention.

In a first embodiment of the motor control device for a constant tension suspension mechanism according to the present invention, as shown in FIGS. 1 [A] and [B], a target object 101 is suspended from a winding mechanism 102. Suspended by hanging line 103, suspended by winding mechanism 102
In a constant tension hanging mechanism that maintains the tension of the hanging wire 103 constant by winding it up and unwinding it seven times, the hanging wire 10
3, and control circuits 106 and 109 that calculate a torque command value for the drive motor 105 of the winding mechanism 102 to reduce fluctuations in the detected output based on the detected output of the tension sensor 104. Equipped with
It is configured to give a torque command value to the drive motor 105.

As a second form of the motor control device for a constant tension suspension mechanism according to the present invention, as shown in FIG.
In the form of , the drive motor 105 of the winding mechanism lO2
The control circuit 106 further includes a static friction offset adding means 107 for adding a static friction offset to the detection output of the tension sensor 104 to offset the static friction generated by the suspension mechanism when the rotation starts. It is configured to calculate a torque command value based on the detection output of the added tension sensor 104.

As a third form of the motor control device for a constant tension suspension mechanism according to the present invention, as shown in FIG.
Alternatively, in the second embodiment, the control circuit 106 further includes a dynamic friction offset adding means 108 that adds a dynamic friction offset to the detection output of the tension sensor 104 to offset the dynamic friction generated in the suspension mechanism when the drive motor rotates. The torque command value is calculated based on the detected output of the tension sensor 104 to which an offset has been added.

As a fourth embodiment of the motor control device for a constant tension suspension mechanism according to the present invention, as shown in FIG.
In the configuration, the drive motor 105 of the winding mechanism 102
The device further includes a static friction torque offset adding means 110 for adding a static friction torque offset to the torque command value calculated by the control circuit 109 to offset the static friction generated in the suspension mechanism when the rotation starts. is configured to give a torque command value to the drive motor 105.

As a fifth embodiment of the motor control device for a constant tension suspension mechanism according to the present invention, as shown in FIG.
Alternatively, in a fourth embodiment, dynamic friction torque offset adding means 11 adds a dynamic friction torque offset for offsetting dynamic friction generated in the suspension mechanism to the torque command value calculated by the control circuit 109 when the drive motor 105 rotates.
1, and is configured to give a torque command value to which a dynamic friction torque offset is added to the drive motor 105.

In the third or fifth embodiment, the offset value related to dynamic friction can be a value that changes depending on the rotational speed of the drive motor.

Further, as a sixth embodiment of the motor control device for a constant tension suspension mechanism according to the present invention, as shown in FIG. 1 [C],
In a constant tension suspension mechanism, a target object 101 is suspended by a hanging line 103 hanging from a winding mechanism 102, and the tension applied to the hanging line 103 is maintained constant by winding the hanging line 103 and unwinding it seven times by the winding mechanism 102. , a tension sensor 104 that detects the tension applied to the hanging wire 103, and a target object l.
Target object 1 that keeps the tension of the hanging wire 103 that hangs O1 constant
01 from the variation in the detection output from the tension sensor 104 and the mass of the target object 101, and calculates the rotation speed of the drive motor 105 of the winding mechanism 102 from the calculated speed. and is configured to rotate the drive motor 105 at the rotational speed determined by the rotational speed calculation means 112.

[Operation] In the motor control device of the first form, the control circuit 10
6 generates a torque command value based on the detected output from the tension sensor 104, and drives the drive motor 105 with the torque command value to reduce fluctuations in the tension applied to the hanging wire 103 and maintain a constant tension. ing.

In the second embodiment of the motor control device, the static friction offset generating means 107 generates a static friction offset;
When the drive motor 105 is started from zero speed, this static friction offset is added to the detection output of the tension sensor 104, thereby eliminating the influence of static friction.

In the third embodiment of the motor control device, the dynamic friction offset generation means 108 generates a dynamic friction offset, and while the drive motor 105 is rotating, this dynamic friction offset is added to the detection output of the tension sensor 104, thereby controlling the influence of the dynamic friction. is being removed.

In the motor control device of the fourth embodiment, a static friction torque offset is generated by the static friction torque offset generating means 110, and when the drive motor 105 is started from zero speed, the static friction torque offset is transferred to the control circuit 10.
By adding it to the torque command value from 9, the influence of static friction is removed.

In the motor control device of the fifth embodiment, a dynamic friction torque offset is generated by the dynamic friction torque offset generating means 111, and this dynamic friction torque offset is added to the torque command value from the control circuit 109 while the drive motor 105 is rotating. This eliminates the effects of dynamic friction.

By changing the above-described dynamic friction offset or dynamic friction torque offset in accordance with the rotational speed of the drive motor 105, it becomes possible to more accurately eliminate the influence of dynamic friction.

In the motor control device of the sixth embodiment, the rotational speed calculation means 112 calculates the rotational speed of the drive motor 105 of the winding mechanism from the fluctuation value of the detection output from the tension sensor 104 and the mass of the target object 101. is given to the drive motor 105 as a rotational speed command value. By rotating the drive motor 105 at this commanded rotational speed, the suspension line]
The tension applied to 03 can be maintained at the target value.

(Margin below) [Example] Embodiments of the invention will be described below with reference to the drawings.

Here, Section 1 describes the basic method of a zero-gravity simulation system to which the motor control device for a constant tension suspension mechanism of the present invention is applied, Section 2 describes the suspension method in the same system, and then Section 3 describes the Section 3.2 describes a motor control device for a constant tension suspension mechanism as an embodiment of the present invention, and finally Section 4 describes the configuration of a simulation computer system.

1. Basic methods of zero-gravity simulation system Here we will discuss four basic methods.

■、1. Offline Zero Gravity Zero Gravity Simulation Experiment System This is a zero gravity simulation experiment system using an offline method using a one-string constant suspension mechanism and a support drive mechanism using a robot. In this method, before directly moving the target multi-joint structure (space robot, etc.), the movement of the multi-joint structure in response to external forces received in zero gravity is calculated in advance using special software. The system uses a hardware mechanism to realize exactly the same movement of a multi-jointed structure on the ground based on the movement confirmed by calculation.

1.1.1. Overall Configuration The overall processing outline of the weightless simulation experiment system for realizing the present method is shown in FIG. The system as a whole consists of a simulation data creation section 1, a zero-gravity simulator 2, and a multi-joint structure 3. It is configured to be driven via a control section 21.31 using a value calculated in advance by a software section (motion analysis software section) 12. Further, the overall hardware concept corresponding to FIG. 2 is shown in FIG.

Further, a perspective view of the specific hardware configuration of the entire mechanical section of this zero-gravity simulation system is shown in FIG. 4, a top view thereof in FIG. 5, and a side view thereof in FIG. 6.

Hereinafter, based on these figures, the simulation data creation section 1. The configurations of the zero gravity simulator 2 and the multi-joint structure 3 will be explained.

a. Simulation data creation unit The simulation data creation unit 1 includes a zero gravity simulator 2
and a part that creates basic data for operating the hardware mechanism of the multi-joint structure 3. This part consists of a multi-joint structure mathematical model section 11 that defines a mathematical model of the multi-joint structure 3, and a motion analysis software section that analyzes the movement of the multi-joint structure 3 defined by the multi-joint structure numerical model 11. 1
2, a simulation display software section 13 that graphically displays the output results of the motion analysis software section 12, and a data software conversion software section for zero gravity simulator that converts and processes the output results of the motion analysis software section 12 for the zero gravity simulator 2. Consisting of 14.

b. Multi-joint structure (space robot) The multi-joint structure 3 consists of a multi-joint structure control section 31 and a multi-joint structure mechanism section (robot main body) 32.
constitutes a servo mechanism that uses the data created by the data conversion software section 14 as a follow-up target value. Robot body 32
As shown in FIG. 4, there are two articulated arms 303 on the left and right, each with 6 degrees of freedom, and a hand 3 attached to the tip of the arm 303.
04, it consists of a book 305 that fixes this multi-joint arm 303, and the hand 304 performs operations such as capturing objects. The multi-joint arm 303 is designed to be lightweight in consideration of ease of transportation to space, and is a flexible arm with long and thin links of 200 mm in length and with reduced rigidity against bending and dust. The total length of the arm is approximately 1400 mm. In addition, in order to extend the lifespan in space, we are developing a direct drive (hereinafter referred to as D), which has no transmission parts and fewer contact parts.
The DD arm has a structure in which a motor (abbreviated as D) and an encoder are directly connected to the joint. FIG. 7 shows the joint configuration of the arm. In the figure, 02, θ3, marked with ■. θ5 is the bending axis,
Φ marks 01, C4, and C6 are rotation axes. The zero-gravity simulation system thus simulates the complex behavior of a robot mechanism having an articulated arm 303 using flexible links and DD motors.

C9 Zero gravity simulator The zero gravity simulator 2 is divided into a zero gravity simulating control section 21 and a zero gravity simulating mechanism section 22, and the control section 21 constitutes a servo mechanism that uses data created by the data conversion software section 14 as a tracking target value. .

The zero-gravity simulation mechanism section 22 includes a main body support drive section 202 consisting of a 6-degree-of-freedom robot mechanism section 208 that supports and drives the main body 305 of the target multi-joint structure 3 with 6 degrees of freedom, and parts other than the target base 301. and a constant-tension suspension mechanism section 201 that suspends it by a wire 210. The multi-joint structure 3 includes a base portion 301 and other arms, panels,
It consists of a portion 302 such as an antenna. At this time, the base 301 of the multi-joint structure 3 is supported by a six-degree-of-freedom robot mechanism 208, and the other parts 302 are supported by a plurality of sets of constant-tension suspension mechanisms 201. The entire weightless simulator consists of a frame 203 installed to surround the entire mechanism.
Supported by

In the zero-gravity simulating mechanism section 22, when the entire robot body 32 is supported by wires 210 suspended vertically downward, the number of wires increases and the mechanism and control become complicated. Therefore, the load on the suspension mechanism by the wire 210 is limited to the arm portion 303 as shown in FIG. 4, and this is referred to as a constant tension suspension mechanism section 201. The wire 210 is moved in the vertical direction under constant tension control using the hanging mechanism section 201 to balance the tension of the wire 210 with the gravity of the arm 303 and cancel the gravity of the arm 303.

The main portion 301 of the robot body 32 is supported by the robot mechanism section 208 with six degrees of freedom, and the entire suspension mechanism is supported and driven with two degrees of freedom in the horizontal plane. This is the main body support drive section 20
It is called 2. The movement of the wire 210 in a direction other than the vertical direction (the movement of the suspension mechanism section 201 in the horizontal plane, the movement of the robot main body 32, and the main body support drive section 202) is performed according to the calculation results of the simulation software.

1.1.2 Elemental Parts The main elemental parts used in this zero-gravity simulation system are explained below.

a. Constant tension suspension mechanism The constant tension suspension mechanism section 201 includes a horizontal articulated two-degree-of-freedom robot mechanism (hanging arm) 204 that moves the hanging positions of multiple wires in a horizontal plane, and this hanging arm 204. A wire winding mechanism 205 is mounted on the wire winding mechanism 205 and includes a winding drum that winds and unwinds each of a plurality of wires, a winding drum rotation drive unit, etc., and a wire 210 that is mounted on an extension of this winding mechanism 205 and hangs an arm 302. It consists of a bow length sensor 206 that measures the tension of the wire 210 to cancel the gravitational pull, and a gimbal 207 that is connected to the arm 302 side of the wire 210 to prevent interference between the six degrees of freedom of the arm 302 and the wire 210. be done. Next, the details of the suspension mechanism section 202 will be explained by dividing it into a vertical movement mechanism that cancels gravity with the wire 210, and a movement mechanism that moves the hanging position in a horizontal plane according to the movement of the suspended arm 302. do.

(2) Vertical movement mechanism After explaining the suspension tension adjustment method, which is the key point of the zero-gravity simulating mechanism section 22, we will explain the suspension position, gimbal 207, and counter, which are the component technologies of the constant tension suspension mechanism section 201 adopted in this example. Balancer 220, winding mechanism section 2
05, the tension sensor 206 will be sequentially described.

(B) Suspension tension adjustment method Various methods can be considered as a suspension tension adjustment method that suspends the arm 302 to simulate zero gravity. For example, as shown in FIG. Counterbalance type, suspended by a counterbalancer, using a spring to suspend the arm and winding the wire onto the wire winding drum/
There are constant force spring type (Conston spring) that pays out, and winding type that suspends the arm using a wire winding drum and winding motor.

Among these, the winding type consists of a wire winding drum, a motor that drives the winding drum, and a wire tension sensor.The output of the tension sensor is fed back to the motor and the tension is kept constant through winding and unwinding operations, simulating zero gravity. It is something. This winding method is advantageous in that it can respond to changes in hanging weight and can be made smaller and more precise, and the present invention employs this winding method.

FIG. 9 shows how to deal with the vertical vibration component of the flexible link. The spring winding type [A] is a method that absorbs vibrations using a coil spring that is one order of magnitude lower than the natural frequency of the flexible link.
The winding method [B] is a method in which the band of the winding system is made one order of magnitude higher than the natural frequency of the flexible link, and winding is performed in accordance with the vibration. The winding type has the advantage that weightlessness can be simulated by winding and unwinding the wire with constant tension control even in response to the vertical vibration input of the arm of an arbitrary profile, and this example adopts this method. There is.

(b) Referring to FIG. 7, which shows the hanging position, gimbal, and joint configuration of the counter balancer arm 303, the mass of the arm 303 is the (θ1+02) axis, the (θ3+04) axis,
(θ5+06) It is concentrated in the three joints around the axis.

When the attitude of the arm 303 changes, the distribution state of the mass at the three locations changes. It is desirable that this arm 303 can be supported by as few wires 210 as possible. - On the other hand, in the case of a flexible link, if the link portion is hung intensively, the flexible link will bend due to tension, and the behavior of the arm 303 will be affected by the suspension. Therefore, in this embodiment, in FIG. 10, which schematically shows the arm, the arm is suspended at two locations, the (θ3+θ,) axis and the (θ5+06) axis, by wires 210. In FIG. 1O, θ1 represents a rotation/bending axis, R is a rotor, and S is a stator. Note that (θ1+
02) The shaft is supported by the book 305, and the main body support drive unit 202
is driven by.

Furthermore, when performing a mechanical analysis of the suspension method for a single flexible link with mass points at both ends, it was found that even if the link's posture changes or bending vibration occurs, by hanging both ends with a constant tension, weightlessness can be approximated. An example of conditions that can be simulated is revealed. This condition is that the mass at both ends is sufficiently larger than the link mass, and the linear density of the link is sufficiently high. Arm 3 of this embodiment
The structure and hanging method of 03 are adapted to these conditions.

If one wire 210 is hung at each location other than the center of gravity, a moment will be generated due to the distance to the center of gravity and gravity, and the behavior of the arm will be affected by gravity.

To prevent this, the centers of gravity of the (θ3+04) and (θ5+06) axes are each suspended by one wire 210. However, since each center of gravity is inside the joint, one point is connected to the wire 21.
It cannot be measured directly at 0. Therefore, a gimbal 207 shown in FIG. 11 is used for indirect support from the outside. This gimbal 207 is a connecting member so that the ψ2 axis, the ψ3 axis, the ψ7 axis, and the ψ8 axis, which are adjacent to each other outward from the center, that is, the center of gravity, are orthogonal to each other, and their orthogonal points coincide with the center of gravity of the suspended object. It is a mechanism that rotatably supports objects by connecting them. This gimbal 207 has ψ7 axis,
Ball bearings are used for the ψ2 axis to provide rotational freedom, and the ψ8 axis is substituted with the freedom of the dust of the wire itself, simplifying the structure.

Furthermore, as mentioned above, even if the joint angle changes, the center of gravity of the suspended object must be fixed at one point at the center of the gimbal. For example, to fix the center of gravity of the (0, +04) axis in Figure 1'O, the following sufficient conditions are required for the centers of gravity of the three members S3, (R3+54), and R1. - ■ Since the center of gravity of the entire (θ3+04) axis does not change even with the rotation of the θ4 axis, the center of gravity of R4 is on the 04 rotation axis.

(2) Since the center of gravity of the entire (θ3+04) axis does not change even with θ3-axis bending, the composite center of gravity of (R4+54+Rs) and the center of gravity of 83 are both on the intersection of the θ-bending axis and the 04 rotation axis.

■Hang the intersection with a gimbal. In order to realize (1) and (2), counter balancers are provided on the rotating shaft and the bending shaft as shown in FIG. FIG. 12 is an example regarding the (θ3+04) axis, and in FIG. 12, 220a is a counter balancer of the half-length portion 220A of the (θ4-05) link;
0b is the half length part 220B of the link between (θ2-03)
220c is a counter balancer for 220C, 220e is a counter balancer for setting the center of gravity of rotor R4 of the 4th axis on the 04 axis, and the bending axis θ3 is from the rotor R1 of the 3rd axis and the stator S. Thus, the rotation axis θ4 is composed of a fourth shaft rotor R4 and a stator S3.

A counter balancer is similarly provided on the (θ5+06) axis at the tip of the arm. The intersection of two adjacent axes is on the bending axis as shown in FIG. This is also advantageous in the following respects.

■The rotor shaft can be shared with the gimbal connecting member, simplifying the mechanism.

■There is less interference between the counter balancer and gimbal, allowing for a wider operating range.

■The hanging position can be easily calculated using control and simulation software.

(C) Wire Winding Mechanism The winding mechanism 205 needs to wind the wire around the winding drum many times in response to the vertical movement of the arm. In this case, if the winding drum is simply driven to rotate around the winding drum axis, the wires will overlap at one point on the winding drum and wind up like a mountain. If slippage occurs between the wires and the mountain collapses, the movement becomes discontinuous and unstable, and the wires are also likely to break. Therefore, the wires must be wound in a single layer on the winding drum at pitches that do not touch each other. FIG. 13 shows a winding mechanism for this purpose.

FIG. 13 is a front view of the winding mechanism section 205 including a tension sensor 206, which will be described later. FIG. 14 is a side view of this as seen from direction A, and FIG. 15 is a top view of this as seen from direction B. .

In the figure, 230 is a DC motor with a reducer, 240 is a motor mounting plate, 231 is a shaft coupling that transmits the rotation of the motor 230 to the ball spline 232, and 241 is the ball spline 2.
32 is a support bearing, 232 is a ball spline that rotates the winding drum 235 with the rotation shaft of the motor 230 and provides a degree of freedom of movement in the direction of the rotational axis of the winding drum, 233 is a ball spline nut, 234 is a ball spline shaft, 235 236 is a bearing mounting plate; 237 is a guide roller with a protrusion that regulates the axial movement of the winding cylinder 236; 238 is a guide roller support shaft; 239
is a guide roller bearing.

The winding drum 235 is provided with a spiral groove 1.5 times the diameter of the wire 210. A ball spline 232 is attached to the inside of the winding drum 235, and while the winding drum 235 is rotationally driven by a motor 230, the winding drum 235 is given a degree of freedom of movement in one direction. A guide roller 237 having a protrusion that engages with a groove in the winding drum 235 is fixed to the frame of the winding mechanism section 205. As a result, the winding drum 235 rotates and is guided by the guide roller 237 that engages with the groove thereof, and moves in the axial direction to the right/left according to the winding and unwinding of the wire 210. The wire can be wound spirally in a single layer in the groove of the winding drum 235 while the position remains fixed.

(d) Structure of Tension Sensor and Control Device There are various methods for detecting the tension applied to the wire 210, such as those shown in FIG. 16, for example. In other words, the current sensor method detects the motor current and calculates the motor output torque (Fig. 2A), the torque sensor method attaches a torque sensor to the output shaft of the reducer (Fig. 2B), and the load cell is installed in the middle of the wire. These include the load cell method (Fig. 2C), which detects the load with a wire guide, and the strain gauge method (Fig. 2D), which uses a strain gauge attached to the wire guide bobbin support.

In this example, among these wire tension detection methods,
A strain gauge method is used, which has the advantage of being able to measure tension stably. As for the structure, a first
As shown in FIG. 3 and FIG. 17, two rollers 250, 251 are provided vertically below the wire winding position, and between them there is a A third roller 252 having a support shaft at a different level is provided. This uneven roller 252 is supported by a plate spring 253, and four strain gauges 254 are supported by the plate spring 253.
to obtain differential output.

With this configuration, the wire 210 is connected to the uneven roller 2.
If you wrap it around 52 and hang it, this uneven roller 25
2 can create a horizontal component of tension, and this horizontal component P can be found as follows: T=P/2sin θ, where T is the tension and θ is the angle that the wire 210 makes with the vertical line. The amount of strain of the parallel plate spring 253 that supports the uneven roller 252 is equal to the horizontal component force P within the range of elastic deformation.
Therefore, the tension T of the wire can be detected by detecting the amount of strain with four strain gauges 254 attached to the parallel plate spring 253, two on the front and the back.

The output of this tension sensor generated by the gravity of the arm joint is transmitted to the rotational drive motor 23 of the wire winding mechanism section 205.
By feeding back to 0, winding and unwinding operations with constant tension control can be obtained.

b. Movement mechanism in the horizontal plane (hanging arm) The wires of the two arms are moved two-dimensionally in accordance with the movement of the arms in the horizontal plane that projects the hanging position. The bending axis and the adjacent linear axis are one unit of the joint configuration, and these are connected in multiple stages to form a horizontal multi-joint type joint configuration.

Specifically, as shown in FIG. 18, each arm is a hanging arm having four degrees of freedom: two straight axes and two bending axes. The next bending shaft φ2 is mounted on the table of the straight shaft d. Further, the above-mentioned winding mechanism section 205 is mounted on the table of the linear axes d0 and d5□. When the arm rises vertically downward, the two tables are moved closer to each other to reduce the distance between the wires and follow the movement of the arm. The entire suspension arm 201 is connected to a book support drive unit 202, which will be described later, through a φ1 axis. Main body support drive section 2
When 02 moves the book 301 in a straight line or rotation in a horizontal plane, the entire hanging arm also moves parallel to the arm in a horizontal plane. When the main body support drive section 202 moves the base 301 in the vertical direction.

The hanging arm 201 does not move and maintains a constant height.

A DC motor with a speed reducer is used for bending the hanging arm 201, and a DC motor, a ball screw, and a straight guide are used for straight movement. Motion detection is performed using an optical low-return encoder, and the motor is servo controlled.

■0-piece support drive unit etc. The main body support drive unit 202 is a device that moves the main body 301 to take any position and posture, and it itself has six degrees of freedom (d ax, d BY, dsz, α, β
, γ). As shown in Fig. 210, for the three degrees of freedom in position, the robot is in Cartesian coordinates with clax, dsv, and de □ axes. On the other hand, d
In the BX axis, the shaft passes through a fixed table and moves straight in the vertical direction.

Regarding posture, there are three degrees of freedom α, rotation and bending, which are orthogonal to each other.
, β, and γ axes are arranged in series. The α-axis is mounted on the dsv-axis table and all dsx-axes are rotated together. Connect the φ and axis of the hanging arm between the α axis and the dsz axis. The drive mechanism uses a motor, ball screw, and linear guide for linear movement. Since the travel distance for all three linear axes is long, the critical speed decreases when the ball screw shaft rotates, and vibration and noise are likely to occur. To prevent this, ball screws are used by fixing the screw shaft and rotating it with a nut. A motor with a speed reducer is used for rotation and bending operations. In particular, on the α-axis, the robot body and hanging arm must be rotated in synchronization, resulting in a large rotation mechanism. The entire table of the d and ax axes is surrounded by large gears, and the output of the motor with a reduction gear is transmitted to this using small gears. This makes the rotation axis of α coincide with dsx, making the mechanism simple even if it is large. Motion detection is performed using an optical low-return encoder, and the motor is servo controlled.

Although the main body support drive section 202, the suspension mechanism 201, and the robot main body 32 are large and heavy, it is necessary to obtain high positioning reproducibility. Therefore, a linear guide is attached to the frame 203, which has high rigidity (and can provide stable accuracy over the entire operating range).The ball screw shaft is also fixed to this frame 203, and the frame 203
It becomes a structure that reinforces 3. In addition, the hanging arm part 201
The arm portion 302 has a cantilever structure. If the tip thereof is made to protrude outside the frame 203, the operating range in the α and dBz axes increases. Therefore, the hanging arm part 201 and the arm part 302
is operable until it passes through the crossbeam of the frame and protrudes.

FIG. 20 shows the degree of freedom configuration of the entire mechanical section. As shown in the figure, the zero-gravity simulating mechanism section 22 is a six-degree-of-freedom robot mechanism section (dax
, dlv-d sz, a, β, γ) and four sets of constant tension suspension mechanisms 201. The four wires 210 have two arms with four degrees of freedom extending horizontally ([φR1, dsR+,
φ8□, daxzl, [φ1, dsL+, φL2, d
sLzl), and a robot mechanism with two degrees of freedom per wire book corresponds to the account. The degrees of freedom of the parts driven by the motor are 12 for the robot main body 32, 12 for the suspension mechanism section 201, and 6 for the main body support drive section 202.
There are 30 degrees of freedom in total.

1.1.3. Offline data creation A method for creating offline data based on the water injection example will be described.

A model corresponding to the robot body 32 shown in FIG. 20 is constructed in the multi-joint structure mathematical model ll. When constructing a model, it is necessary to specify the overall joint configuration, link length, mass of each link and joint, moment of inertia, and center of gravity position.

The motion analysis software 12 calculates time series of joint angles, joint angular velocities, three-dimensional coordinate values of the suspension position viewed from the inertial system, and their velocities, etc. of the defined mathematical model. FIG. 21 shows the input and output to this motion analysis software. As shown in the figure, the input is a mathematical model of a multi-joint structure, which includes ■the overall joint configuration, ■the length of each link, ■the mass and moment of inertia of each link and joint, and ■the position of the center of gravity of each link and joint. In addition, as the initial values of the multi-jointed structure, ■ initial values of the main center of gravity position and posture angle, ■ initial values of the differential of the main center of gravity position and differential of the posture angle, ■ initial values of each joint angle and its derivative, ■ each interjoint. This is a time series of torque. In addition, the output time series is: ■ Time series of main center of gravity position, attitude angle, and their derivatives,
■Time series of each joint angle and its derivative; ■Time series of coordinate values of multi-joint structure sensor points and their derivatives.

The simulation display software 13 displays the output results in graphs, calculates the maximum and minimum values, automatically checks whether the calculated values are within the valid range, and avoids calculating values that are outside the valid range. In such cases, the user is notified of this in the form of a warning.

The zero-gravity simulator data conversion software 14 converts the time series of the position and posture of the center of gravity of the book 301 of the multi-jointed structure 3 calculated by the motion analysis software 12 into the six-degree-of-freedom robot mechanism (dax, da) of the zero-gravity simulator 2. ma, da□, a, β, γ)
In addition, the three-dimensional time series of the hanging positions and speeds for the four wires 210 are converted into the time series of the two hanging arms ([
φRI.

dllR1% φ0, d, , ], [φL1, d II
L+, φL2, D8L□]) is converted into a time series of each joint angle and angular velocity. The joint angle data for the two hanging arms always follow the movement of the multi-joint structure 3 and
10 is created so that it remains vertical. Specifically, the four hanging points (T*+, T
R2, T R3, “T,,” are the respective hanging positions (θ
Solve the inverse kinematics so that it lies on the vertical line of R3, θR5, θL3, θLli).

1.1.4 Cooperative control operation 2nd based on offline data
FIG. 2 is a diagram cut out from FIG. 2 with a focus on the control system in particular. As shown in the figure, each control unit 21.31 has the functions of communication control, command interpretation/selection, and servo control.
Synchronous control is performed by communication such as PIO (parallel input/output).

The joint angle data time series of the multi-joint structure 3 and the zero-gravity simulator 2 created based on the offline data creation method described above are stored on the memory of the control unit 31.21 in FIG. The two control units 31.21 each have a PI
Servo follow-up control is performed on each mechanism section 32.22 using the stored data as a target value while taking timing through communication such as O. Each mechanism part 32.22 cooperates under the control of the control part 31.21 to extend the suspension wire 210.
It works to keep it vertical. As shown in Figure 22,
This system can be said to be a servo mechanism that holds the suspension wire vertically in a feedforward manner using the data calculated by the motion analysis software 12 and the zero gravity simulator data conversion software 14 as target values.

Also, the wire winding mechanism 20 of the constant tension hanging mechanism section 201
5, to cancel the gravity of the wire 302 and realize a simulated weightless state, the wire 210 is wound and unwound seven times to keep the tension applied to the wire 210 constant.

1.2. Off-line zero-gravity simulation system ■This method is an offline zero-gravity experimental system that uses a constant tension suspension method and support drive by a robot, and improves the followability of the suspension arm by providing a feedback loop using an angle detection sensor. .

The offline zero-gravity simulation system I described above consists of a multi-joint structure 3, a two-degree-of-freedom robot mechanism section 204.6, and a multi-joint structure 3.
This is a method in which the degree-of-freedom robot mechanism section 208 is servo controlled. In this method, the target trajectory for servo control is created based on a software model built on a software simulator, so if the model is inaccurate, it may cause the wire 210 to tilt, and complete gravity compensation may not be possible. It may be damaged. In order to compensate for this, the method proposed here attaches an angle detection sensor to the base supporting each wire 210, and transmits the inclination of the wire 210 detected by this sensor to the two-degree-of-freedom robot mechanism 204 suspending the wire 210. This is to perform local position control by varying the feed, thereby improving the vertical holding ability of the wire 210.

1.2.1. Overall configuration FIG. 23 shows an outline of the processing of the offline zero-gravity simulation experiment system (2). The difference from the above-mentioned system I is that the zero gravity simulation control section 21 is provided with an angle detection sensor and local feedback control is inserted.

1.2.2. Elemental parts We will explain the new elemental parts used in this offline zero-gravity simulation system ■.

a. Angle detection sensor Two examples of angle detection sensors that detect the inclination of the hanging wire 210 will be described: an optical angle detection sensor and a component force detection sensor.

1. Optical angle detection sensor FIG. 24 explains the principle of an optical angle detection sensor. As shown in the figure, the gimbal 207 that rotatably holds the arm 302 is suspended by a wire 21O, and the gimbal 207
A ball bearing is provided on the wire 210 so that the rotation of the gimbal is not transmitted to the wire 210. Wire 2 for hanging gimbal 207
10, a light emitter 331 such as an LED is placed as close to the position as possible. The wire 210 is attached to the hanging arm side at a position where a two-dimensional photoreceptor 332 for detecting the position of the light emitter 331, such as COD, PSD, etc.
(position detection device) or a camera, etc.

Now, with the wire 210 suspended vertically, the position (δ8°, δ, .) of the light emitter 331 detected as a two-dimensional screen by the photoreceptor 332 is expressed as the XY of the two-dimensional screen.
Set as the origin of coordinates. If the position of the light emitter 331 in the two-dimensional screen detected by the photoreceptor 332 when the wire 210 is tilted from the vertical direction is (6 points, δy), then the deviation of the wire 210 from the vertical line, i.e. Angle with vertical line △
θ is.

Δθ=sin-'(δ/h)#δ/h δ='5,"+8,". Note that h in the above formula is the length of the wire 210.

10-Component Force Detection Sensor The principle of a component force detection sensor that detects the inclination of a hanging wire using a horizontal component force is shown in FIG. 25, and a more specific configuration example is shown in FIG. 26. As shown in the figure, this component force detection sensor 340 is attached to the lower part of the tension sensor mechanism 206, and includes a ball bearing portion 342 through which the wire 210 passes;
A 6-degree-of-freedom sensor (or a 2-degree-of-freedom sensor) that detects the deviation of this ball bearing 342 in the horizontal plane
341, a cover 343 surrounding the component force detection sensor, and the like. Note that the six-degree-of-freedom force sensor 341 is fixed to the base of the tension sensor mechanism 206.

With such a configuration, when the wire 210 is tilted from the vertical direction, the wire 210 causes the ball bearing portion 342 to be displaced in the horizontal direction, and the magnitude of the displacement is detected by the six-degree-of-freedom force sensor 341. , whereby the inclination Δθ of the wire 210 from the vertical line can be measured. For example, if the horizontal component force detected by the force sensor 341 is Tc, the slope 80 is △θ=S 1 n −' (T c / T
)T. It is determined by =FT-"+Tc,". Here, T is the bowing force applied to the wire 210, T
cx and T ey are X of the component force TC in the horizontal plane
direction and the X direction component.

When detecting the inclination Δθ of the wire 210 using such a component force detection sensor 340, accurate measurement can be performed without being affected by the rotation of the gimbal 207 or the twisting of the wire 210.

1.2.3. Control operation The basic control operation of the offline zero-gravity simulation system H is the same as the offline zero-gravity experiment system I, but
The difference is that when a tilt Δθ occurs in the wire 210, the tilt Δθ is detected by an angle detection sensor, and this is fed back to the two-degree-of-freedom robot mechanism 204 so that the wire 210 returns to its original vertical direction. The wire hanging position is controlled.

1.3. The online zero-gravity simulation system 1 is a wooden sword system that uses a compromise between a constant tension suspension system, an angle sensor to hold the suspension direction vertically, and a support drive using a robot.

1.3.1. Overall configuration FIG. 27 shows a processing outline of the online zero gravity simulation experiment system I.

This method separates the software simulation section and the hardware simulation section, and uses the values detected by the angle detection sensor and the 6-degree-of-freedom force sensor that detects the force and moment exerted on the base of the articulated structure to create a zero-gravity simulator. The feature is the feedback control system that corrects each joint angle in real time.

That is, the zero-gravity simulation control unit 26 has a sensor for detecting the angle of the hanging wire 210 and a 6-degree-of-freedom force sensor that detects the reaction (force and moment) acting on the multi-joint structure base 301, and the multi-joint structure control unit 36 has A joystick 35 for controlling the multi-joint structure is connected.

1.3.2. Element Part a: Robot-based Reaction Detection Mechanism FIG. 28 shows a reaction detection mechanism that detects the reaction acting on the base 301 of the multi-joint structure 3 as the joints of the multi-joint structure 3 are driven. As shown in the figure, when attaching the two left and right arms 302 to the base, a six-degree-of-freedom sensor 35 is attached to the base, and the six-degree-of-freedom sensor 351 is attached to a cover 352.
It is surrounded by This six-degree-of-freedom sensor 351 can detect the reaction (force and moment) acting on the base 301.

1.3.3. Control Operation The basic processing flow of this method will be explained based on FIG. 27.

In the system I shown in FIG. 2, the entire system is driven offline based on data prepared in advance, but in this system, it is possible to drive the multi-joint structure 3 in real time using the joystick 35. be. The multi-joint structure control circuit 36 is a servo mechanism that uses a command value from the joystick 35 as a target value, and can remotely operate the multi-joint structure mechanism 32.

The multi-joint structure 3 changes its base 301 as the joints are driven.
is subjected to a reaction (force and moment) and the suspension wire 2
10 causes a deviation from the vertical line.

Zero gravity I! The pseudo sub-control unit 6 detects this reaction with the 6-degree-of-freedom force sensor 351, and sends this value to the 6-degree-of-freedom robot mechanism [d, , d, , d8□, α, β,
γ] (see FIG. 20) into the next step command value for each joint angle. 6 degrees of freedom robot mechanism [d,,,
d□, d, □, α, β, γ], the servo drive is repeated one after another using the calculated next step command value as the target value.

FIG. 29 shows an algorithm for causing the base 301 of the space robot 3 to move in the same way as in zero gravity space using the detected value of the force sensor 351. As shown in the figure, the output values (force and moment) of the left and right 6-degree-of-freedom force sensors 351 are detected (step S1), and from Newton's equation and Euler's equation for the base 301 of the space robot, the base center of gravity position of the next step is determined. Calculate the time derivative of the attitude angle (Step S2), and 6 degrees of freedom robot (
d, , d, , dsz, a, β, γ) is calculated in step S3), the 6-degree-of-freedom robot mechanism is servo driven (step S4), and this process is repeated.

Similarly, the deviation of the hanging wire from the vertical line is detected by the angle detection sensor of the zero-gravity simulation control unit 25, and the hanging arm (two-degree-of-freedom robot mechanism) 204 is servo-driven so as to always correct this deviation to zero.

With the control method described above, the base 301 of the multi-joint structure 32 always behaves in a way that takes into account the influence of reaction.
The hanging arm 204 moves following the multi-joint structure 3 so that the wire 210 always remains vertical. This control system has real-time properties on a time scale longer than the time required for the dynamic calculations of the zero-gravity simulation control section 26 and the sampling time of the servo drive.

This online zero-gravity simulation system I also allows hardware-based zero-gravity simulation experiments to be displayed using software. That is.

The initial value and drive command value of the multi-joint structure 3 are sent directly to the motion analysis software 12 of the real-time software simulator 10 through the joystick 35. The kinematic analysis software 12 substitutes the obtained values into the equation of motion and displays them on the simulation display software 13 for each step.

With this method, real-time response is possible to movements on a time scale longer than the basic step interval in the movement analysis section 12 and display section 13.

As mentioned above, in this system configuration, joystick 35
It is possible to check the behavior of both hardware and software in real time based on the command values.

1.4. Online Zero Gravity Simulation Experiment System ■This method uses only a constant tension suspension method and a method of holding the suspension direction vertically using an angle sensor.

1.4.1. overall structure FIG. 30 shows an outline of the processing according to this method.

The feature of this method is that in the zero-gravity simulation mechanism section 22 of the online zero-gravity simulation experiment system I mentioned above, the center of gravity of the base 301 is suspended instead of the six-degree-of-freedom robot mechanism section that supported the base 301 of the multi-jointed structure 3. A constant tension suspension mechanism is newly provided, and the entire multi-joint structure is suspended by wires 211, as shown in FIG.

Therefore, the zero gravity simulation control section 27 of this system includes an angle detection sensor, and the mechanism section 22 includes a constant-tension suspension mechanism section that suspends the center of gravity of the base instead of the 6-degree-of-freedom robot mechanism section. The rationale for setting the hanging position on the center of gravity of the base will be explained in detail in the next section.

1.4.2. Control operation In this method, the entire multi-joint structure is suspended by multiple wires 210 and 211 with a constant tension, and each suspension arm constantly holds the wire vertically by checking the value detected by the angle detection sensor. The influence of gravity is canceled with respect to any movement of the multi-joint structure 3.

2. Hanging method The basic method for determining the hanging position and tension value of the hanging mechanism, which is one of the important elemental technologies in this zero-gravity simulation system, is described below. A feature of either method is that the influence of gravity on the multi-joint structure can be completely canceled by using a constant tension value that does not depend on the posture.

2.10 Center-of-Gravity Lifting Method This method is a constant tension hanging method in which the center of gravity is placed above the vertical line. That is, this method is a method in which the target multi-joint structure 3 is disassembled into links and suspended on a vertical line passing through the center of gravity of each link. The tension value is taken as the weight of each link.In the one-bokuto style, the number of wires is equal to the number of disassembled links. Therefore, when supporting the base 301 of the multi-joint structure 3 using a six-degree-of-freedom robot mechanism, the number of wires is naturally reduced.

FIG. 32 shows the principle of this system.

In FIG. 32, each link is a completely rigid body, and each hanging position P1. It is assumed that a gimbal mechanism as shown in FIG. 11 is connected to P2 and P3. When lifting a single unit, hang the unit on the vertical line of its center of gravity. Tension T is equal to the gravity of a single body. In addition, when suspending a multi-jointed structure, the center of gravity of each unit shall be suspended on the vertical line. Each tension T1, T2, T3 is equal to the gravity of each unit.

As can be seen from Fig. 32, each component is suspended on a vertical line passing through its center of gravity, so the tension values TI and T2
, T3 are always constant regardless of the posture. The influence of gravity is canceled for each component, so there is no influence of gravity on the entire assembled multi-jointed structure.

2.2. Static balance method This method is a method to determine the suspension position and tension value by establishing a static balance equation for the target multi-jointed structure.

The principle of this system will be explained by taking a serial mechanism as shown in FIG. 33 as an example. Hanging position PI, P of arm 303
2 is the bending axis of each joint, and the hand 304 alone is also 1
Hang with book wire. As mentioned above, a gimbal is used for hanging. It is assumed that the center of gravity of each link can be aligned on a straight line as shown. At this time, the equation of static balance is as follows.

T here. are the tensions supported through the left fixed body, T, , T2. T3 is the tension applied to each hanging wire, and 121 to β6 are the distances from the center of gravity.

From the above formula (5-11), the values of tension T., T, , T2.T3 are determined as follows.

The balance equation (5-1) holds true for any posture other than that shown in FIG. 33. Therefore, each tension value in (5-2) is constant regardless of the posture of the object.

2.3. Compromise method This method is a compromise between the above-mentioned center-of-gravity vertical suspension method and static balance method in the constant tension suspension method.
This is a method to reduce the number of fishing wires as much as possible.

Now 1. Consider that the entire system shown in FIG. 33 is supported by two wires with tensions T, , T2. To do this, the hand part that was previously supported by the wire with tension T must be supported with the wires TI and T2, but since the center of gravity of the link CD has shifted to the right of the hanging position C, the wire with tension T3 has to be supported. If you remove it, the link CD will sag due to gravity. In order to avoid this, a counter balancer is installed to compensate for the gravity W3 of the link CD, as shown in FIG.

The static balance equation in this case is as follows.

From the above equation (6-11), find the value of tension T., T,, T, and as before, □In this case, each tension value in equation (6-2) is always independent of the object's posture. By hanging with this constant tension, it is possible to simulate a state of zero gravity.

3. Control of hanging arm 3.1. Algorithm for hanging direction vertical holding method The deviation of the hanging wire from the vertical line determined using the angle detection sensor described above is fed back to the position control of the hanging arm by an algorithm as shown in FIG. That is, the sensor output values of the angle detection sensors 330 and 340 (
δ8, δy), (Te,, TC,) (step 511), and based on this sensor output, use inverse kinematics regarding the hanging arm to calculate the target joint angle of the hanging arm (step 512), and servo drive the hanging arm. is performed (step 513), and this process is repeated. detecting the length of the wire from the encoder output value of the winding motor that changes according to the servo drive of the hanging arm; step 514);
This is used when calculating the target joint angle of the suspension arm in the next cycle.

3.26 Constant tension control by winding mechanism section Tension sensor 2
06 is used to detect the tension applied to the hanging wire 21.0, and the winding mechanism unit 205 adjusts the tension so that the tension becomes constant.
A motor control method for controlling the winding motor 230 will be described below.

3.2.1. Torque Control Method I This torque control method I is a method in which the winding motor 230 is torque-controlled based on the detection output of the tension sensor 206, and the tension applied to the hanging wire is kept constant to simulate a zero-gravity environment.

a. Overall configuration FIG. 36 shows a system configuration for realizing the present torque control method (2). As shown in the figure, the system configuration includes a mechanism section 22 of the zero gravity simulating device that is connected with a wire to an object to simulate zero gravity, and a control section 21 that controls this mechanism section.

Here, the zero gravity simulating mechanism section 2 related to the present torque control method 1
The components of 2 are the winding mechanism 205 and tension sensor 206 described above in FIG. 13, and in FIG.
A winding motor 230 of the winding mechanism section 205 that lowers the winding, and a tension sensor 2 that detects the tension of the hanging wire 210.
06 is shown.

In its internal circuit, this tension sensor 206 detects the tension value actually applied to the wire and the tension value (based on the gravity value of the object) for realizing a zero-gravity environment determined by each suspension method in the second section above. (preset in the internal circuit) and outputs it as a tension fluctuation value. Therefore, in an equilibrium state where the object is suspended by a wire with a constant tension equal to its gravity value, this tension fluctuation value becomes zero.

On the other hand, the zero gravity simulation control section 21 includes control software 261 and a control circuit 260. The control software 261 outputs the gravity value acting on the object to be suspended as a tension command value, but the tension sensor 206 detects when the tension value applied to the wire 210 is the gravity value of the object and outputs zero. Since the gravitational force value of the target object is set as the zero point, the tension command value is also set to zero in the above example. It has become. Control circuit 260
This circuit calculates the difference between the tension fluctuation value detected by the tension sensor 206 and the tension command value of the control software using an internal comparison section, calculates and outputs the torque command value for the winding motor 230 based on the difference. be.

b. Structure of each element The specific structure of the winding mechanism section 205 and the tension sensor 206 is the same as that explained with reference to FIG.

How to control the winding mechanism for simulating C0 zero gravity As mentioned in the article, in order to cancel the gravity acting on an object and simulate zero gravity, the wire 210 is used to constantly pull the object vertically upward with a tension equal to gravity. Just raise it. Therefore, the control software 261 uses a tension command value (in this example, zero = gravity value) that instructs the tension to be applied to the wire in order to cancel the gravity acting on the object and create a zero-gravity environment.
is output to the control circuit 260.

When the object is suspended by the wire 210 with a constant tension equal to the gravity value of the object, the detection output from the tension sensor 206 is zero as described above. In this state, the control circuit 260 generates a torque command value commensurate with the gravity value of the suspended object, and the winding motor 230 generates a torque equal to the torque command value. This maintains the object in a state where its gravity and the tension of the wire are balanced, ie, in a state of weightlessness.

The tension applied to the wire 210 varies depending on the movement of the object and the external force acting on the object. This tension fluctuation value is measured by the tension sensor 206, and the measured tension fluctuation value is input to the control circuit 260 together with the tension command value (=zero) from the control software 261. The control circuit 260 calculates the difference between these tension fluctuation values and the tension command value using a comparison section, calculates the torque value of the winding motor 230 that can quickly reduce this difference (that is, the tension fluctuation value), and sets this as the torque value. It is given to the winding motor 230 as a command value.

The winding motor 230, which has received the torque command value, is driven by the torque of that magnitude, and thereby winds the wire 210 and unwinds it seven times by the winding drum 235 so as to cancel out the tension fluctuation of the wire. That is, when the tension fluctuation value increases in the positive direction, the wire 210 is unwound to decrease the wire tension, whereas when it decreases in the negative direction, the wire 210 is wound up to increase the wire tension, thereby The tension of the wire 210 is maintained constant at all times.

As described above, the tension of the wire 210 suspending the object is kept constant with the gravitational force acting on the object as the target value, thereby realizing a simulated zero-gravity environment for the object.

In the torque method I explained above, the detected output from the tension sensor 206 is taken as the tension fluctuation value, but of course, the wire 21
It is also possible to configure so that the tension value applied to 0 is output as is, and in that case, the tension command value from the control software 261 will be the gravity value of the object.

In addition, when simulating a completely weightless state of an object (for example, when simulating a low gravity state on the moon surface), the tension command value from the control software 261 should be changed appropriately. It can be handled by

3.2.2. Torque control method ■ This torque control method eliminates the influence of static friction of the take-up motor, etc., and then uses a tension sensor to control the torque of the take-up motor to maintain a constant tension on the hanging wire. This method simulates a zero gravity environment.

The winding mechanism (winding motor, winding drum) and tension sensor generally have static friction, and when the winding mechanism is suddenly rotated from zero speed, this static friction causes the wire to The winding and unwinding operations are not smooth (as a result, the constant tension control becomes unstable and the simulation of a zero gravity environment becomes inaccurate. This torque control method (■) eliminates the influence of static friction, This improves the responsiveness of constant tension control.

a. Overall configuration FIG. 37 shows a system configuration for realizing torque control method (2). In this torque control method (2), in addition to the configuration of the above-mentioned torque control method I, the tension fluctuation value detected by the tension sensor 206 is also input to the control software 262,
The control software 262 is capable of detecting the occurrence of static friction based on this tension fluctuation value and changing the tension command value to the control circuit 260 so as to eliminate the influence thereof.

), control method The winding mechanism section 2 corresponds to the movement of the object (space robot).
If a static friction force is generated when 05 is activated, the tension applied to the wire changes from a constant tension value by the influence of the static friction. That is, if static friction occurs in the winding mechanism or the like when winding up the wire in response to the upward movement of the object, the tension applied to the wire decreases by the amount of static friction. Similarly, if static friction occurs when the wire is wound down in response to the downward movement of the object, the tension on the wire increases by the amount of static friction.

Therefore, the fluctuation value of the tension actually applied to the wire is detected by the tension sensor 206 and input into the control software 262. The control software 262 has a tension command value for canceling the gravity of the object in a normal equilibrium state. Hereinafter, this tension command value will be referred to as a reference tension command value. In this example, the reference tension command value is zero because the gravity value of the object is taken as the zero point as explained in the above-mentioned torque control method I. The control software 262 compares the input tension fluctuation value with its own reference tension command value. As a result of this comparison, if the tension fluctuation value is greater than the standard tension command value (that is, if the tension fluctuation value is a positive value), it can be determined that static friction has occurred during wire winding; If the tension command value (that is, the tension fluctuation value is a negative value), it can be determined that static friction has occurred during wire winding.

When the control software 262 detects the occurrence of static friction based on the tension fluctuation value, it gives a static friction offset to the reference tension command value so as to remove the influence of this static friction. This static friction offset value is selected to be a value corresponding to the static friction force that is measured in advance when the winding mechanism section etc. is activated.

In other words, the tension fluctuation value is a positive value (static friction during lowering)
In this case, the control software 262 makes the tension command value given to the control circuit 260 smaller than the reference tension command value by the static friction offset value. As a result, the tension applied to the wire 210 is reduced by the amount of static friction effort, and as a result, the tension applied to the wire 210 remains constant despite the occurrence of static friction. On the other hand, when the tension fluctuation value is a negative value (static friction during winding), the control software 262 sets the tension command value given to the control circuit 260 to be thicker than the reference tension command value by the static friction offset value ( As a result, the tension applied to the wire 210 increases by the amount of static friction effort, so that the tension applied to the wire 210 remains constant despite the occurrence of static friction.

Note that in the torque control method described above, static friction is detected based only on the tension fluctuation value from the tension sensor 206, but the invention is not limited to this, and for example, rotational speed information of the winding motor 230 can also be controlled. By configuring the software 262 to input the information and determining the occurrence of static friction based on the tension fluctuation value and the motor rotation speed, detection of the occurrence of static friction becomes more reliable.

3.2.3. Torque control method ■ This torque control method eliminates the effects of dynamic friction as well as the effects of static friction on the winding motor, etc., and then uses a tension sensor to control the torque of the winding motor to control the torque applied to the hanging wire. This method simulates a zero-gravity environment by keeping the tension constant.

In addition to the above-mentioned static friction, dynamic friction that occurs during the operation of the suspension mechanism also exists in the winding mechanism (winding motor, winding drum), tension sensor, and the like. This torque control method (2) is designed to eliminate the influence of dynamic friction in addition to eliminating the influence of static friction in the torque control method (2) described above.

a. Overall configuration FIG. 38 shows a system configuration for realizing torque control method m. In this torque control method (■), in addition to the configuration of the torque control f wholesale method (■) described above, the winding motor 230
The rotation speed is input to the control software 263, and the control software 263 can change the tension command value based on the tension value from the tension sensor 206 and the rotation speed from the take-up motor 230. .

b. Control Method When the winding mechanism section 205 performs one winding and seven rewinding operations, the rotational speed value of the first winding motor is taken into the control software 263. In the control software 263, in addition to adding/subtracting the static friction offset value to the reference tension command value when starting the suspension mechanism as explained in the torque control method -■ above, the control software 263 adds/subtracts the static friction offset value to the reference tension command value when the suspension mechanism is activated. Add dynamic friction offset value to! / subtracted, and this dynamic friction offset value is changed according to the motor rotation speed.

That is, since kinetic frictional force is generally proportional to speed, a proportionality constant that determines the relationship between rotational speed and kinetic frictional force generated at that rotational speed is determined in advance through experiments. When the rotational speed is zero, a static friction offset for static friction is given to the tension command value as in torque control method H, but when the rotational speed is not zero, the rotational speed is multiplied by the proportionality constant described above. The dynamic friction is calculated and added to the tension command value as a dynamic friction offset.

In this case, if the reference tension command value is zero, when the rotation is unwinding (unwinding), the tension command value is the value of negative kinetic friction and effort, and when one revolution is winding (winding up), the tension command value is the value of positive kinetic friction and effort. value.

As a result, the output of the comparison section in the control circuit 260 (the difference between the tension fluctuation value and the tension command value) changes in magnitude by the dynamic friction force, and a torque command value that cancels out this dynamic friction force is generated, thereby causing the above-mentioned torque command value to be generated. In combination with the effect of the torque control method H, it is possible to realize an effect of eliminating the influence of frictional force.

3.2.4. Torque control method ■ This torque control method ■ is the same as the torque control method described above.
Similarly, this method controls the torque of the winding motor 230 based on the detection output of the tension sensor 206.

a. Overall configuration FIG. 39 shows a system configuration for realizing the torque control method tV. Comparing this with torque control method I, the configurations of the take-up motor 230 and the tensioner sensor 206 are the same. Therefore, the tension sensor 206 takes the detected value when the tension is balanced with the gravity value of the object as the zero point, and outputs the fluctuation value of the tension from there as the tension fluctuation value.The difference is that the tension command value is generated. Control software 2
61, and only the tension fluctuation value from the tension sensor 206 is input to the control circuit 264.

b. Control method Torque control method! In (2), the tension fluctuation value from the tension sensor 206 is input to the control circuit 264;
64 calculates a torque command value such that this tension fluctuation value quickly decreases and supplies it to the winding motor 230. By generating a torque corresponding to this torque command value, the winding motor 230 winds and unwinds the wire 21O of the winding mechanism section by seven windings, thereby suppressing fluctuations in the tension applied to the wire 210. The tension exerted on the object (will be kept constant with gravity as the target value).

1.2.5. ) - Torque control method V Torque control method (2), like torque control method (2), is a method for controlling the torque of the winding motor while eliminating the influence of static friction.

a. Overall configuration FIG. 40 shows a system configuration for realizing torque control method V. In the figure, the winding motor 23Ω, the tension sensor 206, and the control circuit 264 are the same as in the torque control method (2). The difference is that the torque command value output from the control circuit 264 is input to the static friction torque offset generation means 265 and the adder 266, and the rotation speed value from the winding motor 230 is input to the static friction torque offset generation means 265. It has been entered. Static friction torque offset generation means 265 generates a static friction torque offset value based on the torque command value from control circuit 264 and the rotational speed value from take-up motor 230, and outputs it to adder 266. Adder 266 adds this static friction torque offset value to the torque command value from control circuit 264 and sends it to winding motor 230.

b. Control method When the suspension mechanism is started, the control circuit 264 calculates a torque command value according to the tension fluctuation value from the tension sensor 206 and gives it to the winding motor 230. However, the torque command value as it is is not enough. , there is a time delay before the winding motor 230 starts moving due to the influence of static friction.

Therefore, the static friction torque offset generating means 265 monitors the motor rotation speed, and only when the rotation speed is zero,
A static friction torque offset value for canceling the static friction force is output, and added to the torque command value from the control circuit 264 by an adder 266 and applied to the winding motor 230.

The magnitude of this static friction torque offset is determined in advance by an experiment as a torque value that offsets the static friction force.
It is set in the static friction torque offset generation means 265. Static LI: The polarity of the friction torque offset value is set in the static friction torque offset generation means 265 using a torque value that balances the gravity of the object as a reference torque value, and this reference torque value is set as a torque command value from the control circuit 264. Determine by comparing with.

In other words, if the reference torque value is greater than the torque command value, it can be determined that static friction has occurred during lowering, and in this case, the static friction torque offset value is made negative to reduce the torque command value accordingly. The effect of static friction is removed by increasing the tension applied to the wire in the lowering direction.

In addition, if the reference torque value < torque command value, it can be determined that static friction has occurred during hoisting, and in this case, the static friction torque offset value is made positive, the torque command value is increased by that amount, and the The effect of static friction is removed by increasing the tension applied to the wire in the direction.

3.2.6. Torque Control Method ■ This torque control method vI, like the torque control method ■, is a method of controlling the torque of the winding motor after removing the influence of dynamic friction in addition to the influence of static friction.

a. Overall configuration FIG. 41 shows a system configuration for realizing the torque control method Vl. The difference between this torque control method ■ and the torque control method ■ described above is that torque control method V
In T, a torque offset generation means 267 is used instead of the static friction torque offset generation means 265, and this torque offset generation means 267 generates the rotational speed value of the winding motor 230 and the torque command value of the control circuit 264. When the rotation speed is zero, a static friction torque offset value is output, and when the rotation speed is not zero, a dynamic friction torque offset value is output.

b. Control method When starting the winding motor 230 from a state where the rotational speed is zero, the torque offset generating means 267 outputs a static friction torque offset value to the adder 266, similar to the torque control method (2) described above. Then, the static friction torque offset value is added to the torque command value from the control circuit 264 and sent to the winding motor 230 to eliminate the influence of static friction.

On the other hand, when the rotational speed is not zero, dynamic friction occurs in the suspension mechanism, so the torque offset generating means 267
generates a dynamic friction torque offset value to offset this dynamic friction force, adds this to the torque command value from the control circuit 264 in an adder 266, and sends it to the winding motor 230. This eliminates the influence of dynamic friction during wire winding 7 and unwinding.

As the magnitude of the dynamic friction torque offset value, a torque value that cancels the dynamic friction force, which is determined in advance through experiments, is used. Further, as described in the above-mentioned torque control method (2), since the magnitude of the dynamic frictional force changes depending on the rotational speed of the motor, the magnitude of the dynamical friction torque offset value is made to change in accordance with the rotational speed. As in the case of static friction, the polarity of the dynamic friction torque offset value is positive when winding the wire,
Negative when lowering.

3, 2.7. Speed control method This speed control method calculates the speed of the object from the tension detected by the tension sensor and the mass of the target object, and controls the speed of the winding motor based on that value, controlling the tension on the wire to a constant value to create a zero-gravity environment. Figure 42 shows a system configuration for realizing this speed control method. The difference from the torque control method paths I to ■ described above is that the zero-gravity simulation control section 21 uses the control software 26.
This control software 268 consists of only tension sensor 8.
06 and outputs a speed command to the winding motor 230.

b. Control method The feature of this speed control method is that the movement of the object in a weightless environment is calculated directly from the external force F applied to the object, and a speed command is given to the winding motor.

That is, the external force F acting on the object is detected from the variation in the output of the tension sensor 206, and the mass m of the object measured in advance is
Using the following equation, the control software 268 sequentially calculates the velocity ■ that the object should have in order to simulate a zero-gravity environment.

仝" Fdt=m (V Vo ) In the formula, △t is the sampling time, and Vo is the initial speed. The speed ■ calculated by this formula is output as a speed command to the winding motor 230. This causes the winding Zero-gravity simulation can be realized while eliminating the influence of friction of the take-up motor 230, winding drum, tension sensor 206, etc.

3.3. Specific configuration of control circuit Control circuit 26 used in torque control method I, ■, ■ mentioned above
The specific configuration of 0 will be described. - FIG. 43 shows an example of the configuration of this control circuit 260. As shown,
This control circuit 260 constitutes a closed loop together with the winding motor 230 and the tension sensor 206, and performs a fixed price system @ (regulator) with the tension command value as the target value.

In this configuration example, the tension sensor 206 is zero-adjusted in the tension sensor amplifier so that a zero value is given to the control circuit 260 as a tension value when the gravity of the object and the target tension applied to the wire are balanced. . Therefore, the output value from the tension sensor 206 is a fluctuation value from zero (ie, a fluctuation range from the target tension). Correspondingly, the tension command value from the control software 261.262.263 is a fluctuation value (offset value) from zero when compensating for friction etc.
It becomes.

As shown in FIG. 43, the control circuit 260
70, an observer 271, a feedback gain adjustment section 272, integrators 273, 274, and the like. The comparison unit 270 compares the detection value detected by the tension sensor 206 and the tension command value from the control software, and outputs a tension fluctuation value that is the difference between the two. The observer 271 is designed based on the transfer function of the winding motor 230 and the tension sensor 206, and estimates the acceleration of the winding motor 230, which is an important state variable, from the tension fluctuation value and the feedback torque command value.

The feedback gain adjustment unit 272
The tension fluctuation value and the estimated acceleration from 1 are also input to integrator 2.
The torque command value is input as a feed pack from 73, and based on these values, the feedback gain of the torque command value, motor acceleration, and tension fluctuation value is adjusted, and the responsiveness of the entire system is adjusted using modern control theory. Constructed so that stability can be adjusted.

The integrator 273 receives the output signal of the feedback gain adjustment section 272, eliminates the steady-state deviation, and outputs a torque command value.
The torque command value is inputted, and the differential element originally possessed by the controlled object is canceled and applied to the winding motor 230.

Next, the specific configuration of the control circuit 264 used in the torque control methods IV, V, and VI described above will be described.

FIG. 44 shows an example of the configuration of this control circuit 264.

In this configuration example, the winding motor 230 and tension sensor 206 that are to be controlled are expressed differently from those shown in FIG. 43, and accordingly the control circuit (inside the dotted line in the figure)
have different configurations. Also in this configuration example, the tension sensor 206 has an output value set to zero when gravity and tension are balanced. Therefore, the detection output is output as a fluctuation value from zero.

The feedback gain adjustment section 275 adjusts the feedback gain of the torque command value and the tension fluctuation value, and can adjust the responsiveness and stability of the entire system using modern control theory. The integrator 273 is for eliminating steady-state deviation due to disturbance.

4. Simulation calculation system In the zero-gravity simulation experiment system described above, the arm of the space robot is suspended with a constant tension using multiple wires to cancel gravity, thereby simulating a zero-gravity state.

How the base of a space robot moves requires consideration of the reaction of the arm's movement, and it is necessary to generate movement data for the base through calculations while taking this into account. Furthermore, since the wire for suspending the arm must always be oriented in the vertical direction, the suspension arm section must move to follow the movement of the arm.

The amount of calculation required for the operation data of these arms, base portions, hanging arm portions, etc. is extremely large. Furthermore, since such a space robot has a large number of motors that move the base, arm, suspension arm, etc., the amount of calculation required to control them is extremely large. Therefore, a simulation computer system that can efficiently perform these calculations is required.

4.1. System configuration I System configuration (2) is a system configured using a high-performance computer and a plurality of control computers, and is applied to the offline zero-gravity simulation experiment system described above.

This system configuration I is shown in FIG. In the figure, 41
is a high-performance computer such as a workstation, and
This corresponds to the simulation data creation section in the figure.

42.43 is a control computer such as a personal computer suitable for real-time processing, and is a high-performance computer 41.
and R5-232C to exchange data,
The control computers 42 and 43 are connected to each other by a PIO (parallel input/output) channel, making it possible to perform synchronous communication.

The control computer 42 is designed to control the servo mechanisms of the left and right arms of the space robot and the servo mechanisms of the left and right suspension arms of the weightless simulator. The control computer 43 also controls the servo mechanism of the main body support drive section 202 of the weightless simulator that supports the base of the space robot and the control section of the tension mechanism that performs constant tension control.

This system operation is explained below. First, the high-performance computer is used to create a model of the space robot, perform motion analysis, create data that allows the robot to actually operate, and create motion data for the suspension arm based on these data. , 3,) Performs processing equivalent to the simulation data creation section described in offline data creation method.

This high-performance computer 41 will generate all operational data regarding the space robot and the zero-gravity simulator.

These operating data are transmitted to each control computer 42,43 using R3-232C. Each control computer 42.
43 converts the received data into instruction values for the servo control boards of each joint of the robot, and sends them to each servo control board to operate the robot. The computers 42 and 43 are synchronized and controlled so that the base section, arm, and suspension arm section operate in an accentuated manner. Also, the control computer 43
independently controls the tension mechanism.

4.2. System configuration ■ System configuration ■ uses multiple control computers.

This system configuration (2) is shown in FIG. In the figure, 44
．． 45 and 46 are all control computers, and in this configuration, the three control computers 44 to 46 share control of the robot and creation of operation data for each part, distributing the load associated with data creation. I'm trying. Here, the left arm control computer 44 controls the left arm servo mechanism and the left hanging arm servo mechanism, and the right arm control computer 46 controls the right arm servo mechanism and the right hanging arm servo mechanism. The main body control computer is designed to control the main body servo mechanism and tension control section.

Main body control computer 45 and arm control calculation 144.
It is possible to send and receive RS-232C data between 46 and 50 meters, and synchronous communication can be performed using parallel input/output channels.

The operation of this system (2) will be explained below.

First, a computer that controls the control system for the space robot's base creates a model of the space robot, performs motion analysis, and determines the movement of the base. Unless the operation of the base section is determined, the operation related to the hanging arm cannot be determined, so the arm control computers 44 and 46 wait for the base control computer 45 to determine the operation of the base section.

Once the operation of the base is determined, the data is transferred to R3-2
32C to the arm control computers 44 and 46. The arm control computer generates motion data regarding the arm and the hanging arm based on the received base motion data.

When arm control computers 44 and 46 finish generating arm-related operation data, each control computer 44 to 4
6 operates the space robot and the weightless simulator while synchronizing each other via parallel input/output channels.

At this time, each of the computers 44 to 46 performs a process of converting each operation data into an instruction value to the servo mechanism section.

The zero-gravity simulation experiment system described above was intended to simulate the behavior of a multi-jointed structure in zero-gravity space on the ground, but it is of course not limited to this, and the tension value of the constant-tension suspension mechanism can be adjusted appropriately. If chosen, it can also simulate the behavior of articulated structures in low-gravity environments, such as the lunar surface.

[Effects of the Invention] As described above, according to the present invention, it is possible to provide a motor control device for a constant tension hanging mechanism using a winding method.

By using a constant tension suspension mechanism using this motor control device, it is possible to realize a zero-gravity simulation experiment device that creates an environment close to a three-dimensional zero-gravity space in a laboratory on the ground. This makes it possible to repeatedly conduct simulation experiments of the target's operation in zero-gravity space over a long period of time, allowing for careful preliminary evaluation of the performance of the sensor system, actuator system, control system, etc. in zero-gravity space that makes up the target. This makes it possible to construct more reliable hardware such as space robots.

[Brief explanation of the drawing]

Fig. 1 is a diagram explaining the principle of the present invention; Fig. 2 is a diagram explaining the processing outline of the offline zero-gravity simulation system I; Fig. 3 is a diagram explaining the hardware concept of the system shown in Fig. 2; The figure is a perspective view of the specific hardware configuration of the system shown in Figure 2, Figure 5 is a top view of the device shown in Figure 4, Figure 6 is a side view of the device shown in Figure 4, and Figure 7 shows the joint configuration of the arm. Figure 8 is an explanatory diagram of various suspension tension adjustment methods, Figure 9 is an explanatory diagram illustrating how to deal with the vertical vibration component of the link, Figure 1O is a diagram showing the schematic configuration of the arm, and Figure 11 is an explanatory diagram of various suspension tension adjustment methods. The figure shows the configuration of the gimbal, Figure 12 illustrates the configuration of the counter balancer, Figure 13 is a front view of the winding mechanism, Figure 14 is a side view of the winding mechanism, and Figure 15 is a diagram explaining the configuration of the counter balancer. A top view of the winding mechanism, FIG. 16 is a diagram explaining various tension detection methods, FIG. 17 is a diagram explaining the principle of the tension detection mechanism, FIG. 18 is a diagram explaining the joint configuration of the hanging arm, Figure 19 is a diagram explaining the degree of freedom configuration of the main body support drive part, Figure 20 is a diagram explaining the degree of freedom of the entire mechanical part, Figure 21 is a diagram explaining input and output to the motion analysis software, The figure is a diagram explaining the processing concept of the control system that maintains the hanging direction vertically in a feedforward manner, Figure 23 is a diagram explaining the processing concept of the offline zero-gravity simulation system H, and Figure 24 is the optical angle detection sensor. Figures 25 and 26' are diagrams explaining the component force detection sensor, Figure 27 is a diagram explaining the processing concept of the online zero gravity simulation system I, Figure 28 is a multi-joint Figure 29 is a diagram explaining the algorithm for moving the articulated structure base. Figure 30 is a diagram of the online zero-gravity simulation experiment system. Figure 31 is a diagram explaining the processing concept, Figure 30 is a perspective view showing the specific hardware configuration of system H, Figure 32 is a diagram explaining the principle of the hanging method suspended on the vertical line of the center of gravity, and Figure 33 is Figure 34 is a diagram explaining the method for determining the hanging position and tension value based on the principle of static balance. Figure 34 is a diagram explaining the eclectic suspension method using a counter balancer. Figure 35 is an algorithm for maintaining the hanging direction vertically. Figures 36 to 41 are system outline diagrams explaining the winding motor torque control methods ■ to ■, respectively; Figure 42 is a system outline diagram explaining the winding motor speed control method; 44 are block diagrams showing an example of the configuration of a control circuit for controlling the winding motor, FIG. 45 is a diagram showing the computer system configuration I of the zero-gravity simulation experiment device, and FIG. 46 is a block diagram showing a configuration example of the control circuit for controlling the winding motor. Device computer system configuration ■
FIG. In the figure, l... Simulation data creation unit, 2... Zero gravity simulator, 3... Multi-joint structure, 11... Multi-joint structure numerical model, 12... Motion analysis software, 13...
...Simulation display software, 14.. Data conversion software for zero gravity simulator, 21.. Zero gravity simulation control section, 22.. Zero gravity simulation mechanism section, 31.. Multi-joint structure control section, 32.- Multi-joint structure mechanism section, 41...
High-performance computer, 42-46...control computer, 201
... Constant tension hanging mechanism section, 202 ... Main body support drive section, 203 ... Frame, 204 ... 2 degrees of freedom robot mechanism, 205 ... Wire winding mechanism, 206 ...
・Tension sensor, 207...-Gimbal, 208...6
Degree of freedom robot mechanism, 210... wire, 230...
・DC motor with reducer, 231...Shaft coupling, 232
...Ball spline, 233; ...Ball spline nut, 234...Ball spline shaft, 235
- Winding drum, 236... Bearing mounting plate, 237... Guide roller with protrusion, 238... Guide roller support shaft, 239... Bearing for guide roller, 240... Motor mounting plate, 241 ...Support bearing, 250.251・
・-Wire guide roller, 252... Uneven roller, 2
53...Plate spring, 254...-Strain gauge, 301.3
05--Multi-joint structure base part, 302--Multi-joint mechanism part, 303--Arm, 304--Hand,
305... Base, 330... Optical angle detection sensor, 331... Light emitter, 340... Component force detection sensor, 332... Photoreceptor, 341... 6 degrees of freedom force sensor, 342...・・Ball bearing part, 343・・・Cover, 351
...6 degree of freedom sensor, 260.264...Control circuit, 261-263.268...Control software, 265
...- Static friction torque offset generation means, 266.-
Adder, 267... Torque offset generation means, 27
0... Comparison section, 271... Observer, 272.2
75...Feedback adjustment section, 273.274...
・Constitution of integrator hardware metal body g + formation 3rd section + disseminated liver body ゛ (top view) Memory η 4 part metal I book's claw (I rule ■ diagram) Sonfu's forceps top pricking claw Fig. 9: 浚'i: m no. arm Enoki Fig. 16 Tension V history if! + Wounded Kawa Riden 17FK N Domito 7 @ @ K Double y meeting I'I analysis software input/output n Fig. 21 Principle diagram of light'f bar angle return detection senwa Fig. 24 Shinai main Shiyama sensa Fig. 25 Part force pinching se ΣFirst η'■ Algorithm cylinder 29 Diagram hanging 41 [991!Fl cost misfortune (year's hanging)] Figure 35 Rainbow 177 Flights of Agony Eel Natsushi Tsubobbe The yoke is forced upon her, the summer system, the evil Torf is sueed, and the banal formula ゛I 绂 36Figure 1=,!7+ff*1llLII
177 Compound Obfuscation NIm'] Shobu Torufuji IIM
He゛■ 輌会normkiatsuki structure section 餠豐n greetings covering the summer system review county toru 7昂・j9 Crazy Man8゛■ Box 3B figure Torque system OP SaikyuU Figure 39 Torque printing valley P power formula■ Figure 40 Torque 11 production p method Figure 43 Ko: γjn system ke ν mouth kiss A pig "1 Figure 44 Edited summer orange system side-high I Figure 45 Drunken system structure M'II Figure 46

Claims (1)

[Claims] 1. The target object (101) is suspended by a hanging wire (103) hanging from a winding mechanism (102), and the hanging wire is wound/unwound by the winding mechanism to suspend the object (101). In a constant tension suspension mechanism that maintains the tension of the wire at a constant level, a tension sensor (104) detects the tension applied to the suspension wire.
and a control circuit (106, 109) that calculates a torque command value for the drive motor of the winding mechanism that reduces the variation in the detection output based on the detection output of the tension sensor, and calculates the torque command value for the drive motor of the winding mechanism. A motor control device for a constant tension suspension mechanism configured to apply a constant tension to a drive motor. 2. Further comprising static friction offset adding means (107) for adding a static friction offset to the detection output of the tension sensor to offset the static friction generated in the hanging mechanism when the drive motor of the winding mechanism starts rotating. 2. The motor control device for a constant tension suspension mechanism according to claim 1, wherein the control circuit is configured to calculate the torque command value based on the detected output of the tension sensor to which the static friction offset is added. 3. Dynamic friction offset adding means (108) for adding a kinetic friction offset for offsetting the kinetic friction generated in the suspension mechanism to the detection output of the tension sensor when the drive motor rotates;
), wherein the control circuit is configured to calculate the torque command value based on the detected output of the tension sensor to which the dynamic friction offset is added. Device. 4. Static friction torque offset adding means for adding a static friction torque offset for offsetting static friction generated in the suspension mechanism to the torque command value calculated by the control circuit when the drive motor of the winding mechanism starts rotating. 2. The motor control device for a constant tension suspension mechanism according to claim 1, further comprising: (110) and configured to give a torque command value to which the static friction torque offset is added to the drive motor. 5. Further comprising a dynamic friction torque offset adding means (111) for adding a dynamic friction torque offset for offsetting the dynamic friction generated in the suspension mechanism to the torque command value calculated by the control circuit when the drive motor rotates; 5. The motor control device for a constant tension suspension mechanism according to claim 1, wherein the motor control device is configured to provide the drive motor with a torque command value to which a dynamic friction torque offset is added. 6. The motor control device for a constant tension suspension mechanism according to claim 3 or 5, wherein the offset value related to the dynamic friction is a value that changes depending on the rotational speed of the drive motor. 7. Suspend the target object (101) with a hanging wire (103) hanging from a winding mechanism (102), and keep the tension on the hanging wire constant by winding/unwinding the hanging wire with the winding mechanism. A tension sensor (104) that detects the tension applied to the suspension wire in the constant tension suspension mechanism that maintains the constant tension.
Calculate the vertical speed of the target object to keep the tension of the hanging wire that suspends the target object constant based on the variation in the detection output from the tension sensor and the mass of the target object, and perform winding from the calculated speed. a constant tension suspension mechanism, comprising a rotational speed calculation means (112) for determining the rotational speed of a drive motor (105) of the mechanism, and configured to rotate the drive motor at the rotational speed determined by the rotational speed calculation means. motor control device. 8. The control circuit includes an observer that estimates the acceleration of the drive motor of the winding mechanism from fluctuations in the detected output of the tension sensor, and a feedback gain that adjusts the torque command value of the drive motor from the acceleration estimated by the observer. and an integrating circuit that eliminates a steady-state deviation of the torque command value and cancels a differential element originally possessed by the controlled object and outputs the same to the drive motor. A motor control device for a constant tension suspension mechanism according to any one of the above. 9. The control circuit includes a feedback gain adjustment section that generates a torque command value of the drive motor of the winding mechanism based on fluctuations in the detected output of the tension sensor while adjusting a feedback gain, and eliminates the steady deviation of the torque command value. 7. The motor control device for a constant tension suspension mechanism according to claim 1, further comprising an integrating circuit for outputting the same to the drive motor.