KR100238983B1 - Method and apparatus for position control of robot - Google Patents

Abstract

The present invention relates to a position control method and a position control apparatus of a robot, comprising: assuming experimental information about a bias feed forward gain between a pair of support shafts and storing the result in a memory; PID control by obtaining a position error between a target position and an actual position of the load shaft; Detecting a current position value of the header portion; Calculating feed forward gains of the respective support shafts based on the detected current position values of the header portion and the assumed bias feed forward gain information; And applying each feed forward gain to the pair of support shafts by position compensation. As a result, mechanical damage and noise can be reduced, and excellent high precision control and high speed operation can be performed.

Description

Robot position control method and position control device

The present invention relates to a position control method and a control apparatus for a robot, and more particularly, to a position control method and a control apparatus for an orthogonal robot for a chip mounter.

Among the proposed robots of various types, a chip mounter robot is a robot that requires high-speed and high-precision position control for automatically mounting components on a printed circuit board.

5 is a schematic configuration diagram of an orthogonal robot for such a chip mounter. As can be seen from this figure, the orthogonal robot 31 for a chip mounter includes a pair of support shafts 33 and 35 and a pair of support shafts 33 and 35 facing each other at a predetermined distance. It is composed of a load shaft 37 which is supported on both sides of the free end and moves in the longitudinal direction, and a header portion 39 which reciprocates in the longitudinal direction of the load shaft 37. For the sake of convenience, the support shaft on the left side of the pair of support shafts will be referred to as the first support shaft 33 and the support shaft on the right side will be referred to as the second support shaft 35. It is movable in parallel along the longitudinal direction of the support shafts 33 and 35 by a ball screw (not shown) provided on one side thereof, and the header portion 39 is also perpendicular to the moving direction of the load shaft by a ball screw (not shown). It is movable. The main controller 3 controls the movement of the load shaft 37 in the support shaft direction and the movement of the header portion 39 in the load shaft direction, respectively, so that the position of the header portion 39 can be accurately controlled. As a result, the mounting and mounting of the component by the chip mounting nozzle of the header portion 39 can be performed.

6 is a configuration diagram of a conventional position control system for an orthogonal robot for a chip mounter. The conventional position control system 41 has a pair of position controllers 5 and 13 for controlling the support shafts 33 and 35 on both sides, and an output signal? 1 from these position controllers 5 and 13, respectively. It has a motor system 11, 19, that is, a servo driver (7, 15) and a servo motor (9, 17) which are operated and controlled by τ2, respectively. Each position controller 5, 13 compares the target position command value θ from the main controller 3 with the current position values fb1 and fb2 fed back from the motor system 11, and thus the position thereof. Based on the error, proportional, integral, and derivative (hereinafter referred to as PID control) control is performed.

If the position of the load shaft 37 does not coincide with respect to the support shaft on either side of the first support shaft 33 or the second support shaft 35, an operation signal proportional to the error is transmitted through the return path. A closed-control systom or feedback-control systom 25 is provided to compensate. That is, when the positions of the two motor systems 11, 19 set by the PID control result do not coincide with each other, for example, the second support shaft 35 with respect to the first support shaft 33, which is the reference axis being controlled. If the position of the load shaft 37 set in the () does not match, this difference value is detected through an error detector or the like. Then, the main controller 3 makes the synchronous control compensation proportional to this difference value to the position controller 13 which controls the position of the load shaft 37 on the second support shaft. This synchronous control compensation can be determined by multiplying the positional deviation φ represented by the misalignment between the first support shaft 33 and the second support shaft 35 and the experimental compensation gain 30.

7 is a control flowchart of FIG. 6. In the conventional chip mounter robot control, first, the main controller 3 reads target position data from the upper reference input element (not shown) (P1), and outputs a corresponding target position command value θ. (P3). Then, the adders 27 and 29 compare the target position command value θ with the current positions fb1 and fb2 of the fed-back motor systems 11 and 19, and output signals to the position controllers 5 and 13. (e1, e2) will be provided. Accordingly, the position controllers 5 and 13 perform PID control according to the positional error between the target position command value θ and the current position values fb1 and fb2 (P4). At this time, in the case of the controlled first support shaft 33, the position control (or the speed control of the servomotor) is performed by the output torque from the servo driver 7 according to the result of the PID control. In the case of the support shaft 35, position control (or speed control of the servomotor) is performed in accordance with the output value obtained by adding the synchronous control compensation value to the output torque from the servo driver 15 (P5).

Synchronous control compensation can be calculated by first calculating the positional deviation between the first and second support shafts 33 and 35 on both sides (P6), then multiplying the positional deviation by the experimental compensation gain and adding the PID control results. (P7), the calculated value is converted into analog speed voltage through the D / A converter (P8) and input to the servo driver 15 (P9), and the servo driver 15 according to the input speed command. Position control of the load shaft 37 on both the support shaft (P10).

By the way, in the conventional position control method and control apparatus for the chip mounter orthogonal robot, the load of the header portion 39 moving along the transverse direction of the load shaft 37 is not considered in the position control for component mounting. Therefore, in the case where the header portion 39 having a relatively large size and a large load is biased toward either one of the first support shaft 33 or the second support shaft 35 on both sides, between these shafts 33 and 35. Twisting or the like due to the load of the header portion 39 is caused. Thus, it may damage the honor of the chip mounter orthogonal robot that requires high precision and accurate position control, and there are also problems that mechanical damage and noise may occur accordingly.

Accordingly, an object of the present invention is to consider a conventional problem, and a predetermined compensation value proportional to the load load of both the first support shaft and the second support shaft, which are varied by the load of the header portion moving in the longitudinal direction of the load shaft. The present invention provides a robot position control method and a control device capable of performing optimal high precision control by reducing the positional deviation of the load shafts on the first support shaft and the second support shaft.

1 is a schematic configuration diagram of a position control system of an orthogonal robot for a chip mounter of the present invention;

2 is a feed forward gain distribution diagram of a first support shaft and a second support shaft for calculating a feed forward gain;

3 is a control flow diagram of FIG.

Figure 4 is a comparison graph between the positional deviation between the support shaft of both sides and the positional deviation thereof according to the present invention,

5 is a schematic configuration diagram of an orthogonal robot for a chip mounter,

6 is a configuration diagram of a position control system of a conventional orthogonal robot for a chip mounter;

7 is a control flowchart of FIG. 6.

<Description of the code | symbol about the principal part of drawing>

3: main controller 5, 13: position controller

7, 15: servo driver 9, 17: servo motor

21, 23: feed forward gain regulator 27, 29: adder

33, 35: support shaft 37: load shaft

39: header part

According to the present invention, there is provided a load shaft having both side free ends supported by a pair of first and second support shafts opposed to each other at a predetermined separation distance and moving in a longitudinal direction thereof, and a length direction of the load shaft. In the position control method of the robot having a header for reciprocating and the control unit for PID control the load shaft and the header, an experimental experiment on the bias feed forward gain between the pair of support shaft Assuming information and storing it in a memory; PID control by obtaining a position error between the target position and the actual position of the load shaft; Detecting a current position value of the header portion; Calculating feed forward gains of the respective support shafts based on the detected current position values of the header portion and the assumed bias feed forward gain information; It is achieved by the position control method of the robot comprising the step of applying each of the feed forward gains to the pair of support shafts as position compensation.

Here, the information on the bias feed forward gain is preferably determined experimentally as a value when the positional deviation between the pair of support shafts with respect to the feed forward gain is minimized in the no load state of the load shaft.

The calculating of the feed forward gain may include: obtaining a speed command value obtained by differentiating a current position value of the header part; Calculating a preliminary feedforward gain based on the speed command value; And applying the preliminary feedforward gain to detect the actual feedforward gain, wherein the preliminary feedforward gain is calculated by the following expression based on the speed command value. Can be calculated,

[Equation 1]

,S1 =

,

S2 =

(S1 is the feed forward gain of the first support shaft, S2 is the feed forward gain of the second support shaft, A is the stroke of the header part, B is the experimentally obtained feed when the load is applied to the support shafts on both sides. Forward gain, and l is the distance of the header to the first support shaft.)

In addition, the actual feedforward gain may be calculated by the following equation.

P1 = S1 + Bfg1,

P2 = S2 + Bfg2

(Where P1 is the actual feedforward gain of the first support shaft, P2 is the actual feedforward gain of the second support shaft, Bfg1 is the bias feedforward gain of the first support shaft, and Bfg2 is the bias feedforward gain of the second support shaft) .

On the other hand, the position compensation is reflected to the sum of the deviation compensation result multiplied by the position deviation and the compensation gain between the two support shafts, the feed forward gain, and the PID control results, the control of the first support shaft or the second support shaft When applied to the support shaft on the other side, it is possible to achieve a very accurate position control of the robot.

On the other hand, according to another field of the present invention, the object is a load shaft which is supported on both sides of the pair of first and second support shafts opposed to each other at a predetermined separation distance and moving along the longitudinal direction, In the position control apparatus of the robot having a drive unit for driving the load shaft, respectively, and a header portion for reciprocating the longitudinal direction of the load shaft, an experiment on the bias feed forward gain between the pair of support shaft A memory for storing information; A detection section for detecting a current position of the header section; An arithmetic processing unit for calculating a feedforward gain of each of the support shafts based on the detected current position value of the header unit and the assumed bias feedforward gain information; And controlling the position of each supporting shaft by PID control of the load shaft and the header unit, and by applying the feedforward gain to the driving unit by an output signal from the operation unit processing unit. By the device.

Here, it is preferable that the arithmetic processing unit is simply configured with a feedforward gain adjuster, and the position compensation of each support shaft by the feedforward gain is also possible in the PID controller in the control unit.

Hereinafter, the present invention will be described in detail with reference to the accompanying drawings.

1 is a configuration diagram of a position control system of an orthogonal robot for a chip mounter of the present invention. The position control system 1 of the orthogonal robot for the chip mounter, as described with reference to FIG. 5 of the related art, has a main control section 3 and a position command value? From the main control section 3, respectively. The pair of position controllers 5 and 13 receiving the branch, and the motor systems 11 and 19 that are operated and controlled by receiving the output signals from the position controllers 5 and 13, respectively, the servo driver 7, 15) and servo motors 9 and 17. On each upstream side of the pair of position controllers 5 and 13, a feedback control system 25 through which an output signal for synchronous control compensation is transmitted is provided. In the position control system, separate feedforward gain adjusters 21 and 23 are also provided.

The main controller 3 reads the target position data from the upper reference input element (not shown) and outputs it as the target position command value θ. In the position controllers 5 and 13, the current position values fb1 and fb2 of the motor systems 11 and 19 fed back from the servo drivers 7 and 15 to the position command value θ from the main controller 3, respectively. The position errors e1 and e2 added with) are input. Then, each position controller 5 and 13 performs proportional, integral, and derivative control according to the input position errors e1 and e2. The resultant position values, that is, output torques τ1 and τ2 that are output according to the above, control the position of the servomotors 9 and 17 through the servo drivers 7 and 15. Here, the position control of the servomotors 9 and 17 can be precisely referred to as position control by drive speed control according to the speed command values v1 and v2 of the servo drivers 7 and 15.

In the feedback control system 25, when the set position of the position-controlled load shaft 37 does not coincide with each other on the basis of the support shaft on either side of the first support shaft 33 or the second support shaft 35. A compensation control signal proportional to the error is input to the support shaft on the other side through the return path. That is, when the positions of the two motor systems 11, 19 set by the PID control result do not coincide with each other, for example, the second support shaft 35 with respect to the first support shaft 33, which is the reference axis being controlled. If the position of the load shaft 37 set in the () does not match, this difference value is detected through an error detector or the like. Then, the main controller 3 makes the synchronous control compensation proportional to this difference value to the position controller 13 which controls the position of the load shaft 37 on the second support shaft. The synchronous control compensation may be determined by multiplying the positional deviation φ represented by the misalignment between the first support shaft 33 and the second support shaft 35 and the experimental compensation gain 30.

)에 기초하여 피드포워드게인을 산출하고, 이 산출된 피드포워드게인을 목표위치지령값(θ)을 미분한 속도지령값과 곱하여 서보드라이버(7, 15)에 입력시킨다. 그러면, 서보드라이버(7, 9)의 출력토오크에 의하여 구동되는 서보모터(9, 17)는, 소정의 속도보상이 되게 된다.On the other hand, each feed forward gain adjuster (21, 23), in order to compensate for the positional deviation between the first support shaft 33 and the second support shaft 35, which is varied by the position movement of the header portion 39, Position value of the header portion 39 (

The feed forward gain is calculated based on the &quot;), and the calculated feed forward gain is input to the servo drivers 7 and 15 by multiplying the target position command value [theta] by the differential speed command value. Then, the servomotors 9 and 17 driven by the output torques of the servo drivers 7 and 9 become a predetermined speed compensation.

Hereinafter, a method of calculating the feedforward gain will be briefly described.

3 is a distribution diagram of feed forward gains of both first and second support shafts for calculating feed forward gains. Bias feedforward experimentally drives the load shaft 37 in a no-load state, that is, the header portion 39 is removed, and at this time, both the first support shaft 33 and the second support are supported. It is the feedforward gain when the positional deviation between the axes 35 is minimized. In the figure, it is shown as a straight line parallel to the X axis, and this value is a constant constant value.

On the other hand, the calculation of the feed forward gain calculates the speed command value by differentiating the current position value of the fed-back header unit 39, calculates the preliminary feed forward gain based on the speed command value, and then calculates the actual feed forward gain. You can get it. The preliminary feedforward gain is calculated according to Equation 1 below. .

[Equation 1]

,S1 =

,

S2 =

(S1 is a preliminary feedforward gain of the first support shaft, S2 is a preliminary feedforward gain of the second support shaft, A is a stroke of the header portion, and B is experimentally applied when a load is applied to the support shafts on both sides. The feedforward gain obtained and l is the distance of the header portion with respect to the first support shaft.)

The preliminary feedforward lanes of each of the first support shaft 33 and the second support shaft 35 are shown above the bias feedforward gain in the drawing, respectively. The actual feedforward gain may be calculated through Equation 2 below.

[Equation 2]

P1 = S1 + Bfg1,

P2 = S2 + Bfg2

(Where P1 is the actual feedforward gain of the first support shaft, P2 is the actual feedforward gain of the second support shaft, Bfg1 is the bias feedforward gain of the first support shaft, and Bfg2 is the bias feedforward gain of the second support shaft) .

When calculating the actual feedforward gain with reference to the drawings, for example, assuming that the header portion 39 is at the α point, the first feed shaft 33, that is, the actual feedforward gain of the S1 axis, is determined by the first support shaft ( 33 can be obtained as the preliminary feedforward gain (β value on the gain distribution line of S1 axis) + bias feedforward gain (γ on the bias feedforward distribution line), and the actual feed of the second support shaft 35, that is, the S2 axis. The forward gain can be obtained by preliminary feedforward gain of the first support shaft (? Value on the gain distribution line of the S2 axis) + bias feedforward gain (γ on the bias feedforward gain distribution line).

In the control method of the present control system 1 having such a configuration, as can be seen through the control flow diagram of FIG. 3, first, the main control unit 3 reads target position data from an upper reference input element (not shown). (S1), and outputs the target position command value [theta] corresponding thereto (S2). Then, the adders 27 and 29 compare the target position command value θ with the current positions of the fed-back motor systems 11 and 19 to provide the output signals to the position controllers 5 and 13. Accordingly, the position controllers 5 and 13 perform PID control according to the positional error between the target position command value θ and the current position values fb1 and fb2 (S4).

At this time, the main controller 3 separately controls the first support shaft 33, which is the reference axis to be controlled, and the second support shaft 35 corresponding thereto, respectively (S5). In the case of the first support shaft 33 which is an axis, the actual feedforward gain calculated through the process as described with reference to FIG. 2 is provided from the feedforward gain adjuster 21 (S9), and this actual feedforward gain is obtained. And the result of PID control are added together (S10), and an output signal proportional thereto is sent to the D / A converter (S11). The D / A converter converts this input signal into an analog speed voltage and inputs it to the servo driver 7 (S12), whereby the position control of the servomotor 9 corresponding to the output torque of the servo driver 7 is performed. It is lost (S13).

On the other hand, in the case of the second support shaft 35, after the positional deviation between the first support shaft 33 and the second support shaft 35 is calculated (S6), the actual feed forward from the feed forward gain adjuster 30 Gain is provided (S7), and this actual feedforward gain, the PID control result, and the deviation compensation value multiplied by the positional deviation and the experimental compensation gain are summed up (S18), respectively. Then, the output signal proportional to the calculated sum value is converted into the analog speed voltage through the D / A converter, and then input to the servo drive 15, and the converted analog speed voltage is supplied to the submotor 17. Acting as an output torque, the load shaft on one side set on the second support shaft is thus set to the same position as the load shaft on the other side on the first support shaft.

4 is a comparative graph experimentally showing the positional deviation between the support shafts of the two sides and the conventional positional deviation thereof in order to determine the effect of the orthogonal robot for the chip mounter to which the position control method and the control device as described above are applied. In order to compare this effect, in the conventional system 41 and the system 1 of the present invention, the header portion 39 on the load shaft is positioned so as to be adjacent to the first support shafts S1 and 33, X The axis is time and the Y axis is position deviation. As can be seen from this comparison graph, it can be seen that the positional deviation of the present system 1 is reduced by almost half compared to that of the conventional one 41. As a result, mechanical twist between the first support shaft 33 and the second support shaft 35 does not occur, whereby the component can be mounted very accurately.

As described above, according to the position control method and control apparatus of the robot of the present invention, the loads of the first and second support shafts on both sides that are varied by the load of the header portion moving in the longitudinal direction of the load shaft are proportional thereto. It is possible to apply a predetermined compensation value, thereby reducing the positional deviation of the load shaft on each of the first support shaft and the second support shaft. Thus, mechanical damage and noise can be reduced, and an excellent effect of performing optimum high precision control and high speed operation is provided.

Claims (10)

A load shaft which is supported by a pair of first and second support shafts opposed to each other at a predetermined separation distance and moved in a longitudinal direction thereof, a header portion which reciprocates in the longitudinal direction of the load shaft; In the position control method of the robot having a load shaft and a control unit for PID control the header portion, Assuming experimental information regarding bias feedforward gain between the pair of support shafts and storing the experimental information in a memory; PID control by obtaining a position error between the target position and the actual position of the load shaft; Detecting a current position value of the header portion; Calculating feed forward gains of the respective support shafts based on the detected current position values of the header portion and the assumed bias feed forward gain information; And applying each feed forward gain to the pair of support shafts as position compensation. The method of claim 1, The bias feed forward gain information may be determined experimentally as a value when the positional deviation between the pair of support shafts with respect to the feed forward gain is minimized in the no load state of the load shaft. Control method. The method according to claim 1 or 2, The calculating of the feedforward gain may include: Obtaining a speed command value differentiating the current position value of the header unit; Calculating a preliminary feedforward gain based on the speed command value; Detecting the actual feedforward gain by applying the preliminary feedforward gain. The method of claim 3, The calculation of the preliminary feedforward gain is calculated by a formula expressed as follows based on the speed command value: [Equation 1] S1 =

, S2 =

(S1 is the feed forward gain of the first support shaft, S2 is the feed forward gain of the second support shaft, A is the stroke of the header part, B is the experimentally obtained feed when the load is applied to the support shafts on both sides. Forward gain, and l is the distance of the header to the first support shaft.) (correction) The method of claim 3, The actual feedforward gain is calculated by the following formula: [Equation 2] P1 = S1 + Bfg1, P2 = S2 + Bfg2 (Where P1 is the actual feedforward gain of the first support shaft, P2 is the actual feedforward gain of the second support shaft, Bfg1 is the bias feedforward gain of the first support shaft, and Bfg2 is the bias feedforward gain of the second support shaft) The method of claim 1, The position compensation is a position control method of the robot, characterized in that the sum of the deviation compensation result multiplied by the position deviation and the compensation gain between the two support shafts, the feed forward gain, and the PID control results. The method of claim 6, The position compensation, the position control method of the robot, characterized in that applied to the support shaft of the other side of the first support shaft or the second support shaft controlled. A pair of free end portions supported by a pair of first and second support shafts opposed to each other at a predetermined distance from each other and moving along the longitudinal direction thereof, a driving unit driving the load shafts respectively, and a longitudinal direction of the load shafts In the position control apparatus of the robot having a header portion for reciprocating A memory for storing experimental information about bias feed forward gain between the pair of support shafts; A detection section for detecting a current position of the header section; An arithmetic processing unit for calculating a feedforward gain of each of the support shafts based on the detected current position value of the header unit and the assumed bias feedforward gain information; And controlling the position of each supporting shaft by PID control of the load shaft and the header unit, and by applying the feedforward gain to the driving unit by an output signal from the operation unit processing unit. Device. The method of claim 8, The operation processing unit is a position control apparatus for a robot, characterized in that the feed forward gain adjuster. The method according to claim 8 or 9, Position compensation of each support shaft by said feed forward gain is possible also in the PID controller in the said control part.

Images