JPH0677910B2 - Control method for industrial robot - Google Patents

Description

The present invention relates to an industrial robot control method applied to a SCARA type robot.

[Conventional technology]

SCARA robots used for mounting odd-shaped parts
In the (XY) plane, the first arm and the second arm move from point to point. This SCARA robot has a problem that it moves very slowly depending on the position when moving from a certain point to a certain point.

In the conventional SCARA robot, the angular acceleration of each axis is always constant without considering the difference in position. However, according to the analysis of the inventor of the present application, it has been found that the required torque of each axis is significantly different depending on the position of the point even when the moving distance is the same.

[Problems to be solved by the invention]

A SCARA robot generally includes a first motor, a first arm pivotally attached at one end to a rotation shaft of the first motor, and
A second motor attached near the other end of the first arm;
A second arm having one end pivotally attached to the rotation shaft of the second motor.

Here, focusing on the moment of inertia about the rotation axis of the second motor, it changes depending on the weight of the work carried by the robot and the weight of the arm. Further, focusing on the moment of inertia about the rotation axis of the first motor, the first arm and the second arm can be changed in addition to those that change depending on the weight of the work carried by the robot and the weight of the arm, like the second motor. Is greatly affected by the angle formed by, that is, the position of the robot.

According to the equation of motion of the rotating body that "the torque required for rotation is the multiplication of the moment of inertia and the angular acceleration", if the robot servo motor is driven and controlled so as to output the same angular acceleration for each motor, the moment of inertia Servo control is performed to change the motor torque in proportion to.

Therefore, in the case where the angular acceleration of each axis of the SCARA robot is made uniform, it is necessary to set the angular acceleration to a low value in accordance with the movement pattern that requires the most torque in the series of operations. As a result, there is a drawback that the movement time is 30% to 80% longer than the original ability in the movement pattern without applying torque.

Therefore, the present invention and an object thereof are to determine the optimum acceleration of each axis when moving from a certain point (that is, the angular acceleration such that the torque of one axis becomes an allowable torque value), and move at this optimum acceleration. It is an object of the present invention to provide a method for controlling an industrial robot, which can be controlled so as to reduce the transfer time.

Another object of the present invention is to provide a control method for an industrial robot, which is a simple analysis method for obtaining an optimum acceleration and which enables real-time calculation during operation of the robot.

[Means for solving problems]

This invention is a SCARA type robot in which a first arm 2 which is rotated by the θ1 axis in FIG. 1 and a second arm 6 which is rotated by a second θ2 axis move in an (XY) plane. It is a control method of an industrial robot for controlling.

According to the present invention, the moving time required when all the motors are operated with the maximum allowable torque is calculated from the position information in the (XY) plane, and the first θ1 axis and the second θ2 are calculated. The step of analyzing, for each moving operation between two points, which one of the axes has a long moving time, and one of the first θ1 axis or the second θ2 axis having the longer moving time is set as an allowable maximum torque, First θ1 axis or second θ2
Set the first θ so that the other side of the shaft is below the maximum allowable torque.
And a step of setting the respective angular accelerations of the first axis and the second θ2 axis for each moving operation between two points.

[Action]

One of the θ1 axis and the θ2 axis has an allowable torque, and the other has an allowable torque or less. In this case, in order to satisfy the condition that the two axes operate synchronously, the axis having the longer movement time is determined and the axis having the longer movement time is set as the allowable torque. This analysis is performed for each moving operation in the series of operations. Therefore, one axis always has an allowable torque for each moving operation, and the moving time can be shortened.

〔Example〕

An embodiment of the present invention will be described below. FIG. 1 shows an example of a scalar type robot to which the present invention can be applied.

In this robot, a base portion of a first arm 2 is rotatably supported on a base portion 1, and a drive portion 5 composed of a servo motor 3 and a speed reducer 4 arranged on the base portion 1 causes a first arm to move. The second axis (referred to as the θ1 axis) is rotated. A second arm 6 is rotatably supported at the tip of the first arm 2. A shaft (referred to as a θ2 axis) of the second arm 6 is rotated by a drive unit 9 including a servo motor 7 and a speed reducer 8 arranged on the tip of the first arm 2. A hand 10 is attached to the tip of the second arm 6. In the scalar type robot, the first arm 2 and the second arm 6
Both move on the (X-Y) plane. The operation at this time is controlled by instructing the positions of the start point and the end point of the movement and indicating what speed is to be accelerated between these two points.

A DC servo motor or an AC servo motor is used as the servo motors 3 and 7. FIG. 2 shows an example of the configuration of a servo circuit for the servomotors 3 and 7.

In FIG. 2, 11 indicates a microcomputer. This microcomputer 11 supplies a speed command 13 and a direction command 14 to a speed program circuit 12 according to a program for operation control. The speed program circuit 12 forms an acceleration / deceleration curve based on a command from the micro computer 11. The output signal of the speed program circuit 12 is supplied to the SERVOPACK 15. The speed signal from the speed detector 16 and the position signal from the position detector 17 are fed back and supplied to the servo pack 15.

In the servo pack 15, the servo motor 18 is controlled by the feedback detection signal so that the arm moves according to a predetermined acceleration / deceleration curve determined for each position. FIG. 3 shows an example of the acceleration / deceleration curve.

In FIG. 3, points A to B are constant acceleration sections, points B to C are constant velocity sections, and points C to D are constant acceleration sections. By this speed control from point A to point D, the axis is moved to the vicinity of the target point, and the positioning operation is switched from point D. Then, at the point E, the target position is reached. Therefore, the time from point A to point E is the travel time.

The present invention, which aims at shortening the movement time, increases the acceleration when the movement is stopped, and realizes an acceleration / deceleration curve as shown by a broken line in FIG.

The configuration shown in FIG. 2 is provided for each of the servomotor 3 or the servomotor 7 of the industrial robot shown in FIG. In this case, of the respective axes of the arms 2 and 6, the speed of the other axis is adjusted so that the moving time of the longer axis and the moving time of the other axis match.
The axes are adapted to perform synchronous operation.

Further, the memory of the microcomputer 11 stores teaching data. This teaching data relates to the order, position, and time of movement of the robot. This teaching data is read out, the position, speed, acceleration, etc. of each axis are calculated by the microcomputer 11, and the speed command 13 and the direction command 14 are sent to the speed program circuit 12.
Is output to.

The operation analysis of the above-mentioned SCARA robot will be described with reference to FIG. This motion analysis is easy to calculate,
It is not an equation but has a feature that can secure a considerable accuracy.

This analysis method basically has the following features. First, each of the first arm 2 and the second arm 6 is represented by one or more mass points. Secondly, the torque of the θ1 axis for rotating the first arm 2 is determined by the moving angle of the mass point of the first arm 2, the moving angle of the mass point of the second arm as seen from the θ1 axis, and the radius. From each of them, and obtained as a combined torque of the torques. Thirdly, by setting the ratio of the moving angle of the θ1 axis and the moving angle of the θ2 axis as a threshold value, θ1 that affects the torque of the θ2 axis
The equation is prevented by selecting the angular acceleration of the axis. More on this below.

(A) Modeling is performed.

The mass of each of the first arm 2 and the second arm 6 is represented by one point.

Mass (m ₁ , m ₂ ) ＝ Σmi Distance from joint (lm ₁ , lm ₂ ) ＝ (Σmi × ri ² / Σmi) ^1/2 where mi is the mass of the fine section and ri is the distance from the center Is.

(B) Motion Analysis of θ1 Axis Rotating Arm 2 First, the torque Treq of θ1 axis required to move from point P1 to point P2 is calculated.

As it can be seen from Figure 4, the system, which is configured of two mass m _1, m _2, m _1, the movement of each of m ₂ (in practice, stop) calculating a torque required for the Then add it. Here, when the angular acceleration of the θ1 axis at the time of stop is 1, the torque T1lreq required when the mass m ₁ is stopped is expressed by the following equation.

T11req = force x arm length = (mass x acceleration) x arm length = {mass x (angular acceleration x arm length)} x arm length = m ₁ x 1 x lm ₁ ² mass Next, how to do with respect to m ₂ is described.

First, because it is a problem of the movement stop from point P1 to point P2, the length of the arm is Lm at point P2, the mass is m ₂ is evident. However, ₁ cannot be calculated as an angular acceleration. This is because (Δθm ≠ Δθ ₁ ). Therefore, here, the angular acceleration is {1
It is calculated as being substantially equal to × (Δθm / Δθ1)}.

T12req = m ₂ × 1 × (Δθm / Δθ1) × Lm ₂ Therefore, the torque required to stop the θ1 axis of the mass m ₁
T1req is calculated as T1req = T11req + T12req = ₁ (m ₁ × lm ₁ ² + m ₂ × (Δθm / Δθ1) × Lm ² }.

Here, to minimize the travel time from point P1 to point P2,
The value of 1 may be changed so that T1req becomes equal to the maximum torque of the servo motor. Here, if the optimum value of ₁ is _1R and the allowable torque of the servo motor used is T1max, then _1R = 1 x (T1max / T1req). This (T1max / T1req) is called the utilization rate of the torque of the θ1 axis.

(C) Motion analysis of θ2 axis for rotating the arm 6 First, as shown in FIG. 4, the force generated by the mass m ₂ is the force Fθ2 generated by the acceleration accompanying the stop of the θ2 axis and the acceleration accompanied by the stop of the θ1 axis. Is the resultant force Fθ1.

First, considering Fθ2, in the case of FIG. 4, the θ2 axis is Δ
It moves by θ2, but when viewed in absolute coordinates, only ΔΔθ2 moves. Since the acceleration naturally occurs with respect to the change in the absolute coordinates, assuming that the acceleration at the time of the stop of θ2 is 2, Fθ2 is expressed as follows.

Fθ2 = m ₂ × 2 × lm ₂ × (ΔΔθ ₂ / Δθ2) Next, assuming that the angular acceleration of the θ1 axis is 1, Fθ1 is Fθ1 = m ₂ × 1 × Lm, where Fθ1 is the first arm 2 Since it is a vector orthogonal to, the required torque T2req of the θ2 axis is as follows, adding the effective ratio cosθ22 of Fθ1 to the θ2 axis.

T2req = lm ₂ {Fθ2- (Fθ1 × cossθ22)} After that, similarly to the θ1-axis, _2R = 2 × (T2max / T2req) The torque utilization rate RA2 of the θ2-axis is RA2 = T2max / T2req.

The actual calculation is basically the same as the above-mentioned principle, but θ2
Since there are some problems in setting the axis acceleration, this point will be described.

This problem occurs because 1 for calculating Fθ1 is not constant. In other words, 1 should always be _1R , but in reality, when the movement amount of the θ2 axis is larger than a certain amount, (the movement time of the θ2 axis> the movement time of the θ1 axis), and the θ1 axis and the θ2 axis Since the axis is synchronously operated, the acceleration of the θ1 axis is small.

In order to solve this problem accurately, it is necessary to repeat the calculation using a method such as Newton's method, so real-time calculation is difficult with a 16-bit microcomputer.

By the way, in the θ1 axis servomotor, both the arms 2 and 6 serve as loads, and therefore, generally, the torque is stronger than that of the θ2 axis servomotor. Further, the velocity and acceleration are usually smaller on the θ1 axis than on the θ2 axis.
When the ratio of the speed and the acceleration is both K, the following relationship holds.

(A) In the case of | K × Δθ1 / Δθ2 |> 1, the movement time is generally longer on the θ1 axis. That is, the θ1 axis is dominant. In this case, (1 = _1R ), and the θ1 axis is the allowable torque value.

(B) In the case of | K × Δθ1 / Δθ2 | <1, the traveling time is generally longer on the θ2 axis. That is, the θ2 axis is dominant.
In this case, the maximum allowable acceleration 1 _L imit when only the θ1 axis is moved is used as 1 = 1 _L imit × | K × Δθ1 / Δθ2 |, and the reduced 1 is used.

In this way, it is possible to perform the synchronization operation that matches the longer moving time. Actually, as described above, after one shaft is set to the allowable torque value, the calculation is performed again and the check is performed.

The above analysis is performed for each moving operation to determine the optimum angular acceleration of each acceleration / deceleration curve.

〔The invention's effect〕

According to the present invention, the optimum acceleration can be set for each pattern of the moving operation of the SCARA robot, and the moving time from point to point can be shortened. Further, since the respective axial torques of the θ2 axis performed on the θ1 axis can be analyzed by a simple calculation, it is possible to obtain the optimum acceleration for each operation in real time from the data of the position information.

[Brief description of drawings]

FIG. 1 is a front view of an example of a SCARA robot to which the present invention can be applied, FIG. 2 is a block diagram of an example of a servo circuit, and FIG. 3 is a schematic diagram for explaining an acceleration / deceleration curve. FIG. 4 is a schematic diagram for explaining the operation analysis of the embodiment of the present invention. Description of main symbols in the drawings 2,6: arm, 3,7,18; servo motor, 12: speed program circuit.

Claims (1)

[Claims] 1. A first rotation drive unit, a first arm pivotally attached at one end to a rotation shaft of the first rotation drive unit, and near the other end of the first arm. A second rotation drive unit attached thereto, and a second arm having one end pivotally attached to the rotation shaft of the second rotation drive unit, in the (X-Y) plane In the robot control method for controlling the SCARA robot that moves with, the minimum moving time required when operating all axes with the maximum allowable torque from the position information in the (XY) plane is described. Calculate which axis has the longest required minimum movement time,
The step of analyzing each movement operation between two points, and for each movement operation between the two points, the axis having a longer minimum movement time becomes the permissible maximum torque of the axis, and the other axis the permissible maximum torque of the axis. A robot control method comprising the steps of: setting the angular acceleration of each axis as follows.