JPH0448598B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION Field of Industrial Application The present invention relates to an absolute position detection device for a robot joint that can immediately recognize the exact position of a robot joint after power is turned on without returning to the origin.

BACKGROUND OF THE INVENTION Currently, many robots using reduction gears require a return to the origin when the power is turned on. In other words, since the motor rotates within the range of motion of the robot joint,
If you do not know how many revolutions the motor is rotating, such as when turning on the power, it is necessary to install a sensor such as a limit switch at the rotational position of the joint, and perform a return-to-origin operation in which the joint is rotated automatically or manually until the signal is detected. becomes. However, before returning to the origin, it is not known at what position the robot is stopped, and depending on the position, there is a possibility that the robot will collide with another obstacle when returning to the origin. Therefore, there are several ways to eliminate the need for returning to the origin by installing an absolute position detection device that detects the position of the joints in addition to the motor's rotational position detector, so that the rotation of the motor can be known even when the power is turned on. It is thought that

A conventional example will be explained with reference to FIG. FIG. 3 is a diagram showing the vicinity of the drive motor of a robot joint using a speed reducer, where 1 is a motor, 2 is a first absolute position detector that detects the rotational position of motor 1, 3, 4, 5, 6 is a gear train that reduces the rotation of motor 1 by 100:99. Reference numeral 7 denotes a second absolute position detector, which is rotated by gear trains 3, 4, 5, and 6 with a delay of 100:99 relative to the rotation of the first absolute position detector. In the above configuration, since the motor 1 rotates many times within the operating range of the robot joint, the first
The absolute position of the joint cannot be determined using only the absolute position detector. Therefore, by setting the rotation difference of the first and second absolute position detectors 2 and 7 to 0 at the origin position of the joint, and detecting the rotation difference from the difference between the signals of the two absolute type position detectors 2 and 7. , the absolute position of the joint can be known. In this example, the absolute position of the motor can be known up to 100 revolutions. In this way, gear trains 3, 4, 5, and 6 are approximately 1 (in the above example, 100
With a reduction ratio of 99), the absolute position of the joint can be detected, so the reduction gear can be achieved with a small number of gears.
(For example, Japanese Patent Application Laid-Open No. 60-214019) In addition, as a method other than the above, gear trains 3, 4, 5,
There is also a method in which the reduction ratio of 6 is not approximately 1, but a high reduction ratio of 100:1, and the absolute position of the joint is detected using only the second absolute position detector. (For example, "System and Control" Vol. 27, No. 11, PP. 711-720, 1983) Problems to be Solved by the Invention In the above conventional example, the reducer provided for absolute position detection is Conventional example No. 2 requires several stages of gear trains due to its high reduction ratio. On the other hand, in the first conventional example, the reduction ratio is approximately 1, so it can be configured with a small number of gears, but as shown in the above row, the ratio is 100:99, which is very close to 1, so such a reduction gear can be constructed with one gear. It is difficult to configure, and as in the above example, at least 2
A gear train of stages (4 gears) is required.

The present invention provides an absolute position detection device for a robot joint that has a reduced number of parts, is inexpensive, and has a compact configuration by omitting a speed reducer for absolute position detection.

Means for Solving the Problems In order to solve the above problems, the absolute position detection device for a robot joint of the present invention is rotatably driven by a first member, a motor attached to the first member, and a speed reducer. A first rotational position detector attached to the motor or the first member to detect the rotational position of the motor, and a second rotational position detector attached to the second member to detect the rotational position of the motor. The rotational position detector includes a rotational position detector, and a signal processing unit that detects a rotation difference due to a signal difference between the first and second rotational position detectors and determines the position of the second member with respect to the first member.

Effect The application of this technical means is as follows.

That is, the first and second rotational position detectors rotate at high speed in accordance with the rotation of the motor, but as the second member rotates, a phase difference occurs between the rotations of the first and second rotational position detectors. Then, this phase difference is detected by the signal processing unit, and the rotational position of the second member relative to the first member can be determined. Here the first and second
A compact, commercially available, solid type rotational position detector can be used as the rotational position detector, and since it does not use a reducer for absolute position detection, it is a compact absolute position detection device for robot joints with a small number of parts and low cost. can be configured.

Embodiment An absolute position detection device for a robot joint according to an embodiment of the present invention will be described below with reference to the drawings.

In FIG. 1, 8 is a first member, 9 is a second member rotatably supported by the first member 8 by a bearing 12, 10 is a fixed part attached to the first member 8, and an output part is attached to the second member. A harmonic reducer 9 is connected to the harmonic reducer 9, a motor 11 is attached to the first member 8 and whose output shaft is connected to the input part of the harmonic reducer 10;
1, a resolver as a first rotational position detector that detects the rotational position of the output shaft of motor 11; Reference numeral 15 denotes a signal processing unit that processes the signals from the resolvers 13 and 14 to determine the rotational position of the motor and the rotational position of the joint.

Next, a method for determining the rotational position of the second member 9 with respect to the first member 8 will be explained. In FIG. 1, the rotation of the motor 11 is decelerated by a speed reducer 10, causing the second member 9 to rotate. Here, the rotational position of the second member 9 is assumed to be θ ₂ clockwise when looking from the direction from the first member 8 to the second member 9. Further, the rotational position of the resolver 13 is θ _r1 , and the rotation direction is the same as θ _a , and when θ _a =0, θ _r1 =0. Similarly, the rotational position of the resolver 14 is θ _r2 , and the rotation direction is the same as θ _a , and when θ _a =0, θ _r2 =0. Here, the range of motion of the joint is 360°. That is, 0≦θ _a <2π [rad] - (1) Also, as the second member rotates, a rotation difference occurs between the resolver 13 and the resolver 14, and the following relationship exists between θ _r1 and θ _r2 . be.

θ _r1 −θ _r2 = θ _a −(2) Furthermore, θ _r1 and θ _r2 are expressed in the following form.

θ _r1 =2πn ¹ +θ _s1 -(3) θ _r2 =2πn ² +θ _s2 -(4) However, n 1 =0, 1, 2, 3,...... n ₂ =0, 1, ² , 3,... ...0≦θ _s1 <2π ―(5) 0≦θ _s2 <2π ―(6) Here, n ¹ and n ² represent how many times the resolvers 13 and 14 have rotated from the origin, and when the power is turned on. It is unclear. θ _s1 and θ _s2 are quantities that the resolvers 13 and 14 can actually measure.

Substituting equation (2) into equation (1), 0≦θ _r1 −θ _r2 <2π[rad] －(7) Furthermore, substituting equations (3) and (4) into equation (7), 0≦2π (n ¹ − ^{n 2} ) + θ _s1 − θ _s2 <2π − (8) When θ _s1 ≧θ _s2 , since n ¹ and n ² are integers, conditional expressions (5), (6), and (8)
From the formula, n ¹ = n ² must be satisfied. Therefore (2),
From equations (3) and (4), the joint position θ _a is determined by θ _a = θ _s1 − θ _s2 − (9).

On the other hand, when θ _s1 < θ _s2 , from equations (5) and (6), -2π<θ _s1 −θ _s2 <0 - (10) At this time, in order to satisfy equation (8), n ¹ −n ² =1 −(11). Therefore, from equations (2), (3), and (4), θ _a =2π+θ _s1 −θ _s2 −(12) and the joint position θ _a can be determined.

In this way, the joint position θ _a is calculated as θ _s1 - θ _s2 , and if the result is positive, use that value as θ _a , and if it is negative, add 2π [rad] to the result to calculate θ _a . You can ask for it.

Note that in the above example, in order to obtain θ _a , it is necessary to determine whether θ _s1 −θ _s2 is positive or negative to obtain θ _a , but when obtaining digital θ _a , there is no need to make this determination. FIG. 2 shows the signal processing method. FIG. 2 shows the signal processing section in FIG. 1. In FIG. 2, 16 and 17 are R/D conversion units that convert the signals of the resolvers 13 and 14 into binary numbers, respectively, and 18 is a subtraction unit that takes the difference between the binary numbers of the R/D conversion units 16 and 17. For example, let's say we want to represent 0 to 2π [rad] using a 4-digit binary number.
When θ _s1 = π/2 [rad] and θ _s2 = π/4 [rad], θ _s1 = “1000” and θ _s2 = “0100”.
In this case, θ _a remains as is, θ _s1 −θ _s2 = π/4 [rad] =
"0100" is required. On the other hand, θ _s1 =π/4 [rad],
When θ _s2 = π/2 [rad], θ _s1 = “0100”, θ _s2 =
It is expressed as "1000". In this case, since θ _s1 < θ _s2 , it is necessary to add 2π [rad] to θ _s1 - θ _s2 , but when handling it in binary numbers, "0100" - "1000" = "1100"
= 3π/4 [rad] and θ _a is automatically calculated, and 2π [rad]
There is no need to add

As described above, the absolute position θ _a of the joint can be easily determined by the signal processing unit 15, but the resolution of θ _a is equal to the resolution of the resolvers 13 and 14, and is quite coarse as a resolution for the rotational position of the motor 11. Become something. Therefore, θ _a is used to detect how many rotations the motor 11 is from the origin position, and a highly accurate rotational position of the motor can be obtained by directly using the signal from the resolver 13.

In the embodiment, resolvers of absolute position detectors are used as the first and second rotational position detectors, but similar effects can be obtained by using incremental position detectors.

Effects of the Invention As described above, the present invention provides a first
a rotational position detector, a second rotational position detector attached to the second member that detects the rotational position of the motor, and detecting a signal difference due to a rotational difference between the first and second rotational position detectors, By providing a signal processing unit that determines the position of the second member relative to the first member,
Since the absolute position of the robot joint can be detected without a conventional speed reducer for absolute position detection, it is possible to obtain a robot joint with a small number of parts, a low cost, and a compact configuration.

[Brief explanation of the drawing]

FIG. 1 is a sectional view of an absolute position detection device for a robot joint according to an embodiment of the present invention, FIG. 2 is a block diagram of the signal processing section 15 in FIG. 1, and FIG. 3 is a diagram of a robot using a conventional speed reducer. FIG. 3 is a cross-sectional view showing the vicinity of a joint drive motor. 8... First member, 9... Second member, 10... Harmonic reducer, 11... Motor, 13... First rotational position detector (resolver), 14... Second rotational position detector (resolver), 15 ...Signal processing section.

Claims (1)

[Claims] 1 a first member, a second member rotatably supported by the first member, a reducer attached to the first member and having an output portion connected to the second member, and the first member A robot joint comprising a motor attached to the motor and having its output shaft connected to an input part of the reducer, a first rotational position detector attached to the motor and detecting the rotational position of the motor; A second rotational position detector is attached to two members and detects the rotational position of the motor; What is claimed is: 1. An absolute position detection device for a robot joint, comprising: a signal processing unit that determines the position of a member.