JP2762789B2 - Acceleration / deceleration control method for articulated robot - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an acceleration / deceleration control method effective for shortening the execution time of a teaching program in a teaching / playback type articulated robot.

[0002]

2. Description of the Related Art In recent years, when an operation program taught to an articulated robot (hereinafter referred to as a robot) is executed by playback, it is increasingly important that the operation speed be further reduced and the robot do more work. It has become.

Conventionally, when a robot is operated at a speed according to a taught operation program, for example, at the time of acceleration when starting operation from a stopped state or at the time of deceleration when stopped from an operating state,
If the angular acceleration of each joint of the robot is Ru excessive name there is,
In this case, an excessive load is applied to the speed reducer or the like constituting the robot mechanical unit, or an excessive force is applied, so that the robot arm vibrates . To prevent these ,
The robot arm vibrates for each joint of the robot
In general, a non-specific angular acceleration, that is, a maximum angular acceleration is set, and control is performed so as not to exceed the maximum angular acceleration. Hereinafter, a conventional control method will be described with reference to the drawings.

FIG. 2 shows a flowchart of the conventional acceleration / deceleration control. First, the arithmetic unit of the robot control device reads a teaching program prior to an actual operation. Next, in the case of the CP operation, an interpolation operation is performed. In the case of the PTP operation, an operation amount equivalent division corresponding to the interpolation operation is performed, and an interpolation operation (step 6) for calculating the operation speed and the angular acceleration is performed. Next, the calculated angular acceleration is compared with a preset maximum angular acceleration. If the calculated angular acceleration is larger than the preset maximum angular acceleration, the operation speed is controlled so as to limit each angular acceleration to the preset maximum angular acceleration. Is performed (step 7). Here, the position (feedback) control (step 8) is performed according to the changed operation speed to operate the robot.

[0005]

[SUMMARY OF THE INVENTION However, as in the conventional control method, when performing calculation to the feedback control of the speed during deceleration by setting maximum angular acceleration that is set in advance, the deviation delay by the feedback control is < br /> occur because the angular acceleration that is the calculated set maximum angular acceleration
Even when larger than degrees, when acceleration and deceleration time is shorter than the feedback time constant, the robot will operate without angular acceleration in actual operation reaches the maximum angular acceleration setting
Become That is, when the acceleration / deceleration time is short, the calculated angle
There is a difference between acceleration and angular acceleration in actual operation.
Flip actual angular acceleration would operate at setting maximum angular acceleration smaller angular acceleration, as a result, smaller angular acceleration
Therefore, it takes a long time to operate, and it is not possible to operate in the shortest operation pattern allowed by the robot.

An object of the present invention is to provide a robot control method capable of executing a taught operation program in the shortest time without causing vibration of a robot arm. And

[0007]

In order to achieve the above-mentioned object, the present invention provides a method in which a robot first performs a playback operation.
If acceleration / deceleration control is performed at the set maximum angular acceleration and the acceleration / deceleration time is shorter than the feedback time constant of position control,
The calculation is performed so that the angular acceleration is larger than the set maximum angular acceleration, the angular acceleration is changed, and the acceleration control is performed again.

[0008]

According to the above-described method of the present invention , a conventional teaching program for accelerating and decelerating in a short time is used.
The acceleration / deceleration time can be shortened because the angular acceleration can be increased as compared with the above, and as a result, the operation time of the entire program can be shortened.

[0009]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the robot acceleration / deceleration control according to the present invention will be described below with reference to the drawings. FIG. 1 shows a flowchart of an embodiment of the acceleration / deceleration control of the robot according to the present invention.

In FIG. 1, prior to actual operation, the robot reads a taught program, performs an interpolation operation if the operation is a CP operation, and corresponds to an interpolation operation of the CP operation if the operation is a PTP operation. The motion amount is equally divided (step 1). This allows the interpolation time (typically 10 ms
An angle (interpolation data) at which each axis of the robot is operated is calculated at every 50 ms. The value obtained by dividing the operation angle by the interpolation time is the operation angular velocity of each axis, and the time derivative is the operation angular acceleration.

Next, the calculated angular acceleration is set to a preset maximum angular acceleration for each joint, which is allowed by a speed reducer or the like constituting the robot mechanical section, or the robot arm is prevented from vibrating during acceleration / deceleration. If the calculated angular acceleration is larger than the preset maximum angular acceleration for each joint, the temporarily set maximum angular acceleration for each joint (hereinafter referred to as the set maximum angular acceleration) is used. To perform temporary acceleration / deceleration control (step 2). The acceleration / deceleration time (generally 10 ms to 500 ms) calculated in the temporary acceleration / deceleration control (step 2) is divided by the feedback control time constant (typically 50 ms to 100 ms) of the position (feedback) control (step 5). Time coefficient calculation (step 3) is performed. When the time coefficient calculated in step 3 is 1 or more, the temporary acceleration / deceleration control (step 2) is formally adopted to perform acceleration / deceleration control, and when the time coefficient calculated in step 3 is less than 1 Calculates the maximum angular acceleration in inverse proportion to the calculated time coefficient (
If K, set maximum acceleration to 1 / K times inversely proportional
Calculation, eg acceleration / deceleration time is 10ms, feedback control
When the time constant is 50 ms (K = 10/50 = 0.2), 1 /
K = 1 / 0.2, that is, change the set maximum acceleration to 5 times
Then, reacceleration / deceleration control (step 4) is performed. That is,
When the acceleration / deceleration time is smaller than the feedback control time constant, for example, when the speed is low or when it is necessary to decelerate immediately after acceleration, change the acceleration / deceleration by increasing the maximum angular acceleration as the acceleration / deceleration time becomes shorter. Recalculate the angular velocity at the time. The position (feedback) control (step 5) is performed according to the angular velocity recalculated here, and the robot is operated by performing the speed control and the current control.

According to such control, the calculated angular acceleration
Degree only if and acceleration and deceleration time is greater when than the set maximum acceleration is less than the feedback control time constant, since the larger the maximum angular acceleration so as to be inversely proportional to the time factor, the actual angular acceleration specified maximum acceleration Crossing
As a result, the overall operation time can be reduced without the robot arm vibrating.

[0013]

As is apparent from the above description, according to the acceleration / deceleration control method of the present invention, the acceleration / deceleration time is reduced without causing the robot arm to vibrate, and as a result, the overall operation time of the robot is reduced. Is performed, it is possible to provide an articulated robot that can execute the teaching program in the shortest time, and it is possible to improve the operating efficiency of the articulated robot.

[Brief description of the drawings]

FIG. 1 is a flowchart showing an embodiment of a method for controlling acceleration / deceleration of an articulated robot according to the present invention.

FIG. 2 is a flowchart showing a conventional articulated robot acceleration / deceleration control method;

[Explanation of symbols]

 1 Step 1 (interpolation control) 2 Step 2 (temporary acceleration / deceleration control) 3 Step 3 (time coefficient calculation) 4 Step 4 (re-acceleration / deceleration control)

──────────────────────────────────────────────────の Continued on front page (51) Int.Cl. ⁶ Identification code FI G05B 19/42 V

Claims (1)

(57) [Claims] In a teaching / playback type articulated robot, an interpolation operation between teaching points is performed based on a position of a teaching point taught and a speed between the teaching points, and for each joint calculated as a result of the calculation. The angular acceleration is compared with a preset maximum angular acceleration, and if the calculated angular acceleration is larger, a temporary acceleration / deceleration control is performed using the preset maximum angular acceleration to calculate an acceleration / deceleration time. ,
The acceleration / deceleration time is divided by the feedback control time constant to calculate a time coefficient. When the calculated time coefficient is 1 or more, the temporary acceleration / deceleration control is regarded as formal acceleration / deceleration control, and when the time coefficient is less than 1 Is an acceleration / deceleration control method for an articulated robot that performs re-acceleration / deceleration control by calculating and changing the maximum angular acceleration so as to be inversely proportional to the calculated time coefficient.