JPH0679246B2 - High-speed positioning method - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to a numerical control device (hereinafter referred to as CNC) with a built-in computer.
, Etc., such as robots, etc., and digitally controlled using a computer, and relates to a positioning control method for positioning a machine.

Conventional technology In devices such as CNCs and robots that digitally control a machine using a computer to perform machine positioning, pulse distribution calculation is performed according to the command speed of the block at a specified sampling cycle, and this calculation is performed. The pulse distribution amount, that is, the movement amount, is output to the position control unit in the next sampling cycle, the servo motor is driven, and the position control unit detects whether or not the position commanded by the block is reached. It was determined whether the deviation amount was within the in-position width set by the parameter, and when it was within the in-position width, the processing of the next block was started.

Problems to be Solved by the Invention FIG. 4 is an explanatory diagram of a conventional positioning control method, FIG. 4 (a) shows a sampling period T, and FIG. 4 (b) shows a pulse distribution calculation in each block. By this pulse distribution calculation, the movement amount P output to the position control unit at each sampling cycle at the command speed f in the block is calculated.
(n), P (n + 1) ... Is calculated, and this movement amount P (n), P (n + 1) ... is transferred to the position control section with a delay of one cycle as shown in FIG. 4 (c). Is output. Then, upon receiving the movement amounts P (n), P (n + 1) ..., The servo motor rotates as shown by A in FIG. 4 (c). On the other hand, the position deviation amount ε (m) is detected every predetermined period from the value of the error register in the position control unit, and this position deviation amount ε (m) is detected.
Is within the in-position width set by the parameter, and if it is within the in-position, pulse distribution calculation for the next block is started.

In FIG. 4, 3 from t (n) for the first block.
The pulse distribution calculation is performed over the sampling cycle, and the movement amount P (n), P (n + 1), P (n + 2) calculated by the pulse distribution with a delay of one sampling cycle in the position control unit is the command speed f. It is input to the position control unit. Then, the motor rotates and the time t (n
It describes the case where the detected position deviation amount ε (m) in the cycle beginning with +5) is within the in-position width. The pulse distribution calculation for the block is performed, and the calculated movement amount is output to the position control unit in the cycle starting at the next sampling cycle t (n + 6). This means that even if the operation of the next block may be started immediately after it is detected that the in-position width has been entered,
The operation of the next block is started with a delay of one sampling period T, and there is a disadvantage that the processing time becomes long.

Therefore, an object of the present invention is to provide a high-speed positioning method capable of allowing the operation of the next block (rotation of the motor) to be performed without delaying one sampling cycle after the position deviation amount enters the in-position width. To do.

Means for Solving the Problems According to the present invention, pulse distribution calculation is performed for each predetermined sampling cycle in response to a movement command of one block read by a program, and the movement amount calculated by pulse distribution is calculated in the next sampling cycle. Output to the control unit, move the machine, detect the position deviation amount from the position control unit at the start of the sampling period after the sampling period after the pulse distribution calculation is completed, and detect the position deviation amount and the position deviation. By predicting the position deviation amount detected in the next cycle according to the command speed by the pulse distribution calculation in the sampling cycle one before that which detected the amount, and the predicted position deviation amount falls within the set in-position width. The above problem was solved by arranging to start processing of the next block upon completion of positioning.

The transfer function of the servo system controlled by the working position control unit is uniformly determined by the hardware of the machine being specified.

Therefore, if the position deviation amount is detected for each sampling period after the next sampling period after the pulse distribution calculation is completed, this position deviation amount and the command speed of the sampling period before the sampling period in which the position deviation amount is detected are used. ,
The position deviation amount detected in the next sampling cycle according to the transfer function is predicted, and when the predicted position deviation amount is within the in-position width, the pulse distribution calculation of the next block is immediately started, and the next timing cycle is calculated. Pulse distribution is input to the position control unit and drives the servo motor, so the machine immediately starts the operation of the next block when it reaches the position commanded by one block. Compared with the positioning method, the processing operation of the next block is started one sampling period earlier.

Embodiment FIG. 1 and FIG. 2 are explanatory views of an embodiment of the present invention. In FIG. 1, a sampling cycle is T, a certain block is read, and time t (n), t (n + 1) , t (n + 2) in each cycle, as a result of calculation of pulse distribution in each cycle, that is, pulse distribution amount in each cycle, that is, movement amount P (n), P (n + 1), P (n + 2) Is obtained, and the obtained movement amounts P (n) to P (n + 2) are output at a speed f commanded by the program to the position control unit with a delay of one sampling period. , The transfer function of the servo system controlled by the speed control unit shows an example of first-order lag, and the servo motor changes the command movement amount P (n) to P (n + 2) at the command speed f of each cycle. In contrast, Fig. 1, Fig. 2, F (t), F '(t)
It will be driven as shown by the speed curve shown by.

That is, assuming that the time constant of the first-order delay of the transfer function of the servo system is Tc and the time point t (n + 1) when the movement amount P (n) is input to the speed command section is the origin, the servo for the movement command at the command speed f is set. The response speed curve F (t) of the motor is expressed by the following first equation.

F (t) = f (1-e- ^{(t / Tc)} ) (1) This response speed curve shows time t (n + 2) and time t (n +) of the next cycle.
Between 3), it is equal to the sum of the curves of g1 (t) and g2 (t) in FIG. The curve g1 (t) indicates that there is a movement command of speed f in one cycle between time t (n + 1) and t (n + 2), and the movement command disappears at time t (n + 2). The speed decay curve of the servo motor after the time t (n + 2) is shown, and g2 (t) is the time t (n +
It shows the speed curve of the servo motor when there is a movement command of speed f in 2), and the time between the time t (n + 2) and t (n + 3) (T
+ T), the velocity curve F (t) of the first equation is
It becomes like a formula.

F (t) = f (1-e ^{-((T + t) / Tc)} ) (2) The curve g1 (t) is given by the following third equation.

g1 (t) = f (1-e- ^{(T / Tc)} ). e- ^{(t / Tc)} (3) The curve g2 (t) is given by the following fourth equation.

g2 (t) = f (1-e- ^{(t / Tc)} ) (4) As a result, g1 (t) + g2 (t) = f (1-e- ^{(T / Tc)} ) ・ e- ^{(t / Tc)} + f (1-e- ^{(t / Tc)} ) = f (1-e ^{-((T + t) / Tc)} ) (5), g1 (t) + g2t is equal to the right side of the second equation Becomes

Similarly, between times t (n + 3) and t (n + 4), the value obtained by adding the similarly obtained curves g3 (t) and g4 (t) is F (t).

Similarly, the speed (damping) curve F '(t) of the servo motor after the time t (n + 4) is obtained by adding the curves g3 (t) and g5 (t).

That is, the speed F of the servo motor at time t (n + 3), t (n + 4)
From the first equation, (2T) and F (3T) are F (2T) = f (1-e- ^{(2T / Tc)} ) (6) F (3T) = f (1-e- ^{(3T / Tc) )} ) (7), and if the elapsed time from time t (n + 4) is t, then F '
From the 7th formula, F ′ (t) ＝ f (1-e ^{− (3T / Tc)} ) ・ e ^{− (t / Tc)} … (8), g3 (t) from the 6th formula, g3 (t t) = f (1-e- ^{(2T / Tc)} e) -e ^{-((T + t) / Tc)} (9) g5 (t) is g5 (t) = f (1-e- ^{(T / Tc)} ) ・ e- ^{(t / Tc)} … (10) g3 (t) ＋ g5 (t) is calculated from the 9th and 10th equations. G3 (t) ＋ g5 (t) ＝ f (1-e ^{− (2T / Tc) )} -e ^{- ((T + t / Tc)} + f (1-e- ^{(T / Tc)} ) ^{-e- (t / Tc)} = f (e ^{-((T + t / Tc)} -e- ^{((3T + t / Tc)} + e- ^{(t / Tc)} -e ^{-((T + t) / Tc)} ) = f (1-e- ^{(3T / Tc)} ) ・ e- ^{(T / Tc)} … (11 ), Which is equal to the right-hand side of Eq. 8, and F '(t) = g3 (t) + g5 (t) (12).

On the other hand, since the movement amount is obtained by integrating the velocity, at the time when the movement amounts P (n) to P (n + 1) have been input to the position control unit, that is, at time t (n + 4). The position deviation amount is a value obtained by integrating the speed curve F ′ (t) from time t (n + 4) until the speed becomes “0”, that is, the area indicated by a in FIG. Then, from Equation 12, this area is the velocity curve g5 (t), g3
It is a value obtained by integrating (t) from time t (n + 4) to the time when the velocity becomes "0", that is, the sum of the areas shown in FIGS. 2b and 2c.

Considering the area b, since the servo motor follows the movement command and moves by the movement command amount, the speed is increased during the sampling cycle T between the time t (n + 3) and t (n + 4). The command of f, that is, the movement command amount of f · T is issued,
On the other hand, since the servo motor is rotated as shown by the speed curve g4 (t), the amount moved during the sampling period T from time t (n + 3) to t (n + 4) is the speed. Curve g4 (t)
Is a value obtained by integrating t (n + 3) to t (n + 4), and the remaining amount is equal to the area b. That is, the areas d and b shown in FIG. 2 have the same value. As a result, the value of area b is Therefore, the area b can be obtained.

Further, considering the area c, when the time t (n + 3) is the origin and the speed at this time is f ₀ , the area c is the speed curve g3.
It is a value obtained by integrating (t) from T to infinity. That is, it is represented by the following fifteenth expression.

On the other hand, the value obtained by integrating the speed curve g3 (t) from 0 to infinity means the position deviation amount ε (m) corresponding to the speed curve g3 (t) at the time t (n + 3). Therefore, when the position deviation amount ε (m) is detected at the time t (n + 3), the area c is calculated by the equations 15 and 16 as c = ε (m) e ^{− (T / Tc)} (17) To be

As a result, the predicted position deviation amount ε (m + 1) at the time t (n + 4) of the next cycle is the sum of the areas c and b.
From Equation 4 and Equation 17, ε (m + 1) = ε (m) e− ^{(T / Tc)} + fTc (1-e− ^{(T / Tc)} ) (18)

That is, when the pulse distribution calculation ends and the next cycle starts (or the end of the cycle when the last pulse distribution calculation is performed), the first
In the example of the figure, if the position deviation amount ε (m) is read from the error register of the position control unit at time t (n + 3), the predicted position deviation amount ε (m + 1) detected in the next cycle is It can be predicted from Equation 18. In the 18th equation, T is the sampling period, Tc is the time constant of the servo system, and f is the position deviation amount ε (m) detected 1
Since it is the command speed obtained by the pulse distribution calculation in the immediately preceding cycle, the predicted position deviation amount ε in the next cycle is
(m + 1) can be predicted by the detected position deviation amount ε (m) in the period. In the calculation of the predicted deviation amount ε (m + 1) started in the next cycle after the pulse distribution calculation is finished, the command speed f is the speed commanded by the program at the first time, but the second time and thereafter. Becomes "0".

FIG. 3 is a flow chart for executing the positioning method in the present embodiment, and the processing is performed at each sampling cycle.

First, it is determined whether or not there is a distribution pulse for pulse distribution calculation (step S1), and if there is a distribution pulse, distribution calculation processing is performed (step S10), that is, in FIG. 1, the distribution pulse amount P (n ), P (n + 1), P (n + 2) are calculated and output to the position controller in the next cycle. Then, if there is no pulse distribution to be calculated, it is judged whether or not the flag H is "1" (step S2). Since the flag H is not initially "1" as will be described later, the process proceeds to step S3. The command speed f in the cycle is stored in the register R, then the position deviation amount ε (m) is detected from the error register of the position control unit (step S5), and the detected position deviation amount and the register R are stored. From the commanded speed f, the predicted position deviation amount ε (m + 1) in the next cycle is calculated by the 18th equation (step S6). Then, it is determined whether or not the calculated prediction deviation amount ε (m + 1) is within the in-position width set by the parameter (step S7). If it is not within the in-position width, the flag H is set to "1". Is set (step S8), and the processing of this pulse distribution calculation cycle ends. Then, in the next cycle, since the flag H is set to "1", the process proceeds from step S2 to step S4, "0" is stored in the register R, and the position deviation amount ε (m ) Is detected (step S5), and then the predicted position deviation amount ε (m + 1) is calculated by the equation (18). This time, the command speed f stored in the register R is calculated.
Since it is “0”, in Eq. 18, only ε (m + 1) = ε (m) q ^{− (T / Tc)} is calculated to calculate the predicted position deviation amount ε (m + 1). Then, it is determined whether it is within the in-position range.

The following process is performed for each cycle, and the predicted position deviation amount ε (m + 1)
Is within the in-position width, and when it is within the in-position width, the process proceeds from step S7 to step S9, the flag H is reset, and the distribution process of the next block is started (step S10).

In this way, the position deviation amount ε (m + 1) in the next cycle is predicted, and if the predicted position deviation amount ε (m + 1) is within the in-position width, processing of the block next to the cycle is performed. From the start of processing, the processing speeds up by one sampling period when the processing shifts from one block to the next block.

For example, in FIG. 1, the position deviation amount ε (m + 1 at the time t (n + 5), which is the start of the next cycle, is calculated from the position deviation amount ε (m) detected at the time t (n + 4). ) And the predicted position deviation amount ε
If (m + 1) is within the in-position width, the processing of the next block is started from the cycle from time t (n + 4). Therefore, the conventional method starts from time t (n + 5). Compared to the case where the processing of the next block is started, it can be advanced by one sampling period T.

EFFECTS OF THE INVENTION As described above, the present invention is a positioning method that predicts the amount of positional deviation after one sampling period, and immediately starts processing the next block when the predicted amount of positional deviation is within the in-position width. Can be positioned at high speed, and
Enables high-speed processing.

[Brief description of drawings]

FIG. 1 is an explanatory view of a positioning method according to an embodiment of the present invention,
FIG. 2 is an explanatory view of the operation principle in the same embodiment, FIG. 3 is an operation processing flowchart of the same embodiment, and FIG. 4 is an explanatory view of a conventional positioning method. T ... timing cycle, P (n), P (n + 1) ... movement amount by pulse distribution calculation, f ... command speed, F (t), F '(t) ... motor speed.

Claims (2)

[Claims] 1. A movement command for one block read from a program is subjected to pulse distribution calculation at every predetermined sampling cycle, and the movement amount calculated in pulse distribution calculation is output to a position control section in the next sampling cycle. Is moved to detect the position deviation amount from the position control unit at the start of the sampling period after the sampling period after the period when the pulse distribution calculation is finished, and the detected position deviation amount and the position deviation amount immediately before the position deviation amount are detected. The position deviation amount detected in the next cycle is predicted by the command speed by the pulse distribution calculation in the sampling cycle of, and when the predicted position deviation amount falls within the set in-position width, the next block is determined as the positioning completion. A high-speed positioning method that starts processing of. 2. A delay of a servo system controlled by the position control unit is defined as a primary delay, and a position deviation amount detected in the next cycle is predicted from the detected position deviation amount and command speed. High-speed positioning method described in the range 1st term.