JPH01205985A - Bilateral control method - Google Patents

Abstract

PURPOSE:To reduce the burden of an operator by making a torque produced on a master arm in proportion to servo gain processing data obtained from a deviation between the position of the master arm and the position of a slave arm. CONSTITUTION:Slave torque inputs S, T are multiplied by a reaction gain in a reaction gain processing section 1. Then, thus computed value is multiplied by servo gain processing data determined by a positional deviation between a master arm and a slave arm and obtained by a servo gain processing section 3 after being subjected to an absolute value process, and is then used as an instruction value for a reaction torque. With this arrangement, even though a heavy load such as a torque wrench or the like is loaded on the slave arm, no reaction torque is fed back to the master arm so that no burden is applied to the operator. Further, when the slave arm strikes upon an article and so forth so as to be stuck, a reaction torque is fed back to the master arm so as to allow the operate to sense this strike.

Description

[Detailed Description of the Invention] Human and Industrial Application Fields The present invention relates to a bilateral control method for a master/slave type servo manipulator, which is improved so that the reaction torque generated in the master arm does not burden the operator. It is.

B. Summary of the Invention In the bilateral control method according to the present invention, the reaction torque generated in the mask arm by bilateral control is proportional to the servo gain processing data found from the deviation between the master arm position and the slave arm position. If the deviation is small, even if the load on the slave arm is large, the reaction torque will be small and will not put a burden on the operator.On the other hand, if the slave arm collides with something and the position deviation is large, a large reaction torque will be generated and the operation will be difficult. If the operator moves the master arm too quickly, the positional deviation becomes large and a large reaction torque is generated, and the master arm is braked to prevent the operator from moving the master arm too quickly.

C. Conventional technology In the bilateral control system of a master/slave type servo manipulator, when a load is applied to the slave side, a feedback torque is generated on the mask side as a bilateral reaction force.

Conventionally, this reaction torque is obtained by multiplying the slave torque inputs S and T from the torque sensor by a reaction force gain in a reaction force gain processing section 1, and then limiting the reaction force IJ as appropriate in a reaction force processing section 2, as shown in Fig. 2. This is given by driving the master motors M, M.

Note that FIG. 2 shows the configuration for a specific axis, and the position deviation between the mask position inputs M, P by the position sensor and the slave position inputs S, P is multiplied by the servo gain in the servo gain processing section 3, and the obtained result is The slave speed input S, S by the speed sensor is subtracted as a feedback value from the obtained servo gain processing data, and the PI compensation processing unit 4
P [compensation is performed, and the result is supplied to the slave motors S and M as a motor drive current value.

In addition, in order to prevent operability from decreasing due to the weight and friction of the mask, the mask torque input M using a torque sensor,
The friction gain processing section 6 calculates the deviation between T and the self-weight compensation value obtained by calculating the master position inputs M and P using the self-weight compensation processing section 5.
Multiply the friction gain by , add the mask speed input M, S from the speed sensor and the self-weight compensation value to the obtained value, subtract the reaction torque from the slave from this added value, and then supply the motor to the master motors M, M. It is taken as the drive current value.

In addition, in order to prevent reaction torque from occurring due to the slave's own weight, the self-weight compensation processing unit 7 calculates the self-weight compensation value from the slave position inputs S and P, and subtracts the self-weight compensation value from the slave torque human power S and T to generate the reaction force. Multiplying by gain.

D Problems to be Solved by the Invention As is clear from Figure 2, in bilateral control, when a load is applied to the slave arm, the drive current of the slave motors S and M increases in order to maintain the current position. , As a result, the slave torque human forces S and T increase.

Conventionally, when the slave torque human forces S and T increase, the reaction torque directly increases, even if it is within the reaction force limit, so the mask operator feels a large load, which affects operability. In view of the above-mentioned prior art, it is an object of the present invention to provide a bilateral control method for applying a reaction torque so as to reduce the burden on the operator.

E, 1llllff! A bilateral control method according to the present invention is characterized in that the reaction torque generated in the master arm is made proportional to servo gain processing data determined from the deviation between the position of the master arm and the position of the slave arm. .

F. Effect In the above configuration, if the positional deviation between the master arm and the slave arm is small, the reaction torque will be small even if the load on the slave arm is large. Therefore, there is no burden on the operator.

On the other hand, if the slave arm collides with something and the positional deviation is large, a large reaction torque will be generated, making the operator aware of the collision. Furthermore, if the operator moves the master arm too quickly, the positional deviation will increase, so a large reaction torque will be generated and the master arm will have to be braked to prevent it from moving too fast.

G. Example An embodiment of the present invention will be described in detail with reference to FIG.

The embodiment shown in FIG. 1 is an application of the present invention to the conventional example shown in FIG. 2, and the present invention is realized by a portion 8 surrounded by a broken line.

That is, the reaction force gain processing section 1 is applied to the slave torque human power S, T.
The reaction force gain is multiplied by the reaction force gain, and the result is converted into an absolute value by the servo gain processing data determined by the positional deviation between the master arm and slave arm obtained by the servo gain processing unit 3, and multiplied by the command value of the reaction torque. It is said that

9 is an absolute value processing section, and 1o is a multiplication section.

In addition, as in the past, in order to prevent reaction torque from occurring due to the slave's own weight, the self-weight compensation processing unit 7 calculates the self-weight compensation value from the slave position human forces S and P, and the slave torque human force S
, the dead weight compensation value is subtracted from T. Further, the reaction force torque is appropriately limited by the reaction force limit processing section 2 to drive the master motors M, M.

On the other hand, in order to prevent operability from decreasing due to the mask's own weight and friction, the mask torque input value M is determined by the torque sensor.
, T and the self-weight compensation value obtained by calculating the mask position inputs M and P in the self-weight compensation processing section 5, the friction gain is multiplied by the friction gain in the friction gain processing section 6, and the speed sensor is added to the obtained value. The master speed inputs M and S and the self-weight compensation value are summed, and the reaction torque by the slave is subtracted from this added value to obtain the motor drive current value supplied to the master motor M0M.

In addition, for follow-up control of the mask arm and slave arm, the servo gain processing section 3 multiplies the positional deviation between the mask position inputs M, P and the slave position inputs S, P by the servo gain, and from the obtained servo gain processing data, The slave speed human power S, S is subtracted as a feedback value, the obtained difference is subjected to PI compensation in the PI compensation processing section 4, and the result is supplied to the slave motors S, M as a motor drive current value.

From the above, even if the slave torque inputs S, T become large, for example, even if the deviation between the slave torque inputs S, T and the self-weight compensation value becomes large, as long as the difference in position between the master arm and the slave arm is small, As a result, the reaction torque generated in the master arm becomes smaller. Therefore, even when the slave arm carries a heavy load such as a torque wrench, the drive current of the slave motors S and M increases in order to maintain the current position, and as a result, the slave torque input S and T increase. However, it is not a burden to the operator of the master arm.

On the other hand, if the slave arm collides with an object, the positional deviation between the master arm and the slave arm increases and the reaction torque increases, so the operator is not able to recognize the collision because the reaction force of the collision is correctly returned to the operator. , improved operability can be expected.

Further, it is dangerous if the operator suddenly moves the master arm, but since the positional deviation increases in such a case as well, the reaction torque is fed back and braking is applied to the sudden operation of the master arm.

Note that in the above embodiment, the multiplication section 10 is replaced by the reaction force gain processing section 1.
Although it is placed in the latter stage, it may also be placed in the earlier stage. The multiplication section 10 may be omitted by making the reaction force gain processing section 1 variable in gain and changing the reaction force gain in proportion to the servo gain processing data.

(Effects of the Invention According to the present invention, even when a heavy load such as a torque wrench is applied to the slave arm of a master/slave type mask arm, the reaction torque does not return to the mask arm and the operator It's not a burden.

On the other hand, if the slave arm cannot move due to collision with an object, etc., a reaction torque is returned to the mask arm, allowing the operator to recognize the collision. Further, even when the mask arm is moved rapidly, a reaction torque is returned to the mask arm, and the sudden operation of the master arm can be braked.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing one embodiment of the present invention, and FIG.
The figure is a block diagram of a conventional example. In the drawing, M and P are mask position inputs, S and P are slave position inputs, M and M are master motors, S and T are slave torque inputs, 1 is a reaction force gain processing section, 3 is a servo gain processing section, 8 9 is an improved part, 9 is an absolute value processing part, and 1o is a multiplication part. Figure 2 Example of travel

Claims (1)

[Claims] A bilateral control method characterized in that the reaction torque generated in the master arm by bilateral control is proportional to servo gain processing data determined from the deviation between the position of the master arm and the position of the slave arm.