JP2022079273A - Robot control device and robot control method - Google Patents

Abstract

To enable performance in force control to improve more in comparison with a conventional configuration.SOLUTION: A robot control device comprises: a main control part 11 that calculates a command value for controlling force and a command value for torque for each joint that a robot 2 has, on the basis of a command value for force of a robot 2 and a current value for the force, and calculates a control command value for each joint on the basis of the command value for controlling force; a joint control part 15, provided for each joint that the robot 2 has, which calculates a command value for controlling torque in the joint on the basis of the current value of torque in the corresponding joint and the command value for the torque calculated by the main control part 11, and calculates a command value for a motor 21 provided in the joint on the basis of the command value for controlling torque and the command value for controlling calculated by the main control part; and a switching part 119 that switches the calculation of the command value for controlling force performed by the main control part 11 and the calculation of the command value for controlling torque performed by the joint control part 15 valid or invalid.SELECTED DRAWING: Figure 1

Description

The present invention relates to a robot control device and a robot control method capable of simultaneously controlling the position, posture and force of a robot.

In robots (robot arms) such as vertical articulated robots, robot control devices that simultaneously (parallel) control the position and posture and force are used (see, for example, Patent Document 1). The position / posture represents at least one of the position and the posture of the robot. 15 to 17 show an example of the robot control device 1b.

The robot control device 1b shown in FIG. 15 includes a main control unit (upper controller) 11b and a plurality of joint control units (lower controller) 15b. The joint control unit 15b is provided for each joint of the robot 2. The main control unit 11b and each joint control unit 15b are connected by a communication line.

Further, as shown in FIG. 15, the robot 2 has a motor 21 and a sensor 22 (torque sensor 23 and encoder 24) for each joint. The motor 21 and the sensor 22 are each connected to the corresponding joint control unit 15b by a power line or the like. The torque sensor 23 detects the current value of the torque of the corresponding joint. The encoder 24 detects the current value of the corresponding joint angle. Note that FIG. 16 shows only one set of the motor 21, the torque sensor 23, and the encoder 24.

The main control unit 11b controls the entire robot 2 by outputting a command value to each joint control unit 15b. Specifically, the main control unit 11b has a force command value, a position / posture command value, and a command value of the angular velocity for each joint based on the current value of the torque and the current value of the angle of each joint of the robot 2. Is calculated. As shown in FIGS. 16 and 17, the main control unit 11b includes a force calculation unit 111b, a force control unit 112b, a position / attitude calculation unit 113b, a position / attitude control unit 114b, a command value synthesis unit 115b, and a command value conversion unit 116b. I have.

The force calculation unit 111b calculates the current value of the force of the robot 2 based on the current value of the torque of each joint of the robot 2. The torque for each joint of the robot 2 is represented by the joint coordinate system, and in this force calculation unit 111b, the coefficient multiplication unit 1111b multiplies the vector in which the torque for each joint is arranged by the inverse matrix of the inversion of the Jacobi matrix. , Converts the torque for each joint into the force expressed in the Cartesian coordinate system. In FIG. 17, τ represents the current value of torque, J represents the Jacobian determinant, and F represents the current value of force.

The force control unit 112b calculates the command value for force control based on the command value of the force and the current value of the force calculated by the force calculation unit 111b. In the force control unit 112b, the deviation calculator 1121b obtains the deviation between the command value of the force and the current value of the force, and the coefficient multiplication unit 1122b multiplies the deviation of the calculation result by the deviation calculator 1121b by the gain. Then, the command value of force control is obtained. In FIG. 17, Fr represents the command value of the force, and _GF represents the gain.

The position / posture calculation unit 113b calculates the current value of the position / posture of the robot 2 based on the current value of the angle of each joint of the robot 2. The current value of the angle for each joint of the robot 2 is represented by the joint coordinate system, and the position / posture calculation unit 113b converts the current value of the angle for each joint into the current value of the position / posture represented by the orthogonal coordinate system. do. In FIG. 17, θ represents the current value of the angle, and X represents the current value of the position / posture.

The position / posture control unit 114b calculates the position / posture control command value based on the position / posture command value and the position / posture current value calculated by the position / posture calculation unit 113b. In the position / posture control unit 114b, the deviation calculator 1141b obtains the deviation between the command value of the position / posture and the current value of the position / posture, and the coefficient multiplication unit 1142b obtains a gain for the deviation of the calculation result by the deviation calculator 1141b. By multiplying, the command value of position / attitude control is obtained. In FIG. 17, Xr represents a command value of position and attitude, and G _Z represents a gain.

The command value synthesis unit 115b synthesizes the force control command value calculated by the force control unit 112b and the position / attitude control command value calculated by the position / attitude control unit 114b. In the command value synthesizing unit 115b, the adder 1151b adds the command value for force control and the command value for position / attitude control.

The command value conversion unit 116b converts the synthesis result of the command value synthesis unit 115b into a command value of the angular velocity of each joint of the robot 2. In the command value conversion unit 116b, the coefficient multiplication unit 1161b multiplies the synthesis result by the inverse matrix of the Jacobian determinant. That is, the command value conversion unit 116b converts the command value represented by the orthogonal coordinate system into the command value represented by the joint coordinate system. In FIG. 17, θ (dot) r represents the command value of the angular velocity.

The joint control unit 15b controls the motor 21 provided in the corresponding joint in response to a command from the main control unit 11b. As shown in FIGS. 16 and 17, the joint control unit 15b includes a torque acquisition unit 151b and a joint angle control unit 152b.

The torque acquisition unit 151b acquires the current value of the torque at the corresponding joint. The data indicating the current value of the torque acquired by the torque acquisition unit 151b is output to the main control unit 11b (force calculation unit 111b).

The joint angle control unit 152b calculates a command value for the motor 21 provided in the corresponding joint based on the command value of the angular velocity calculated by the main control unit 11b and the current value of the angle for each joint of the robot 2. .. In the joint angle control unit 152b, the speed conversion unit 1521b converts the current value of the angle into the current value of the angular speed, and the subtractor 1523b subtracts the current value of the angular speed obtained by the speed conversion unit 1521b from the command value of the angular speed. , The PI control unit 1524b performs PI control based on the subtraction result by the subtractor 1523b to obtain a command value for the motor 21. The subtractor 1523b and the PI control unit 1524b constitute a speed control unit 1522b.

As described above, in the robot control device 1b shown in FIGS. 15 to 17, it is necessary to operate a plurality of joints in cooperation with each other, and the main control unit 11b has a degree of freedom of multiple joints (for example, 6 degrees of freedom). The result of the control calculation of the position / posture and the force at the same time is synthesized, and the combined result is converted into a signal to the joint control unit 15b of each axis and then output. That is, in the robot control device 1b, the main control unit 11b executes the main calculation of the compliance control. Therefore, this robot control device 1b has an advantage that parameters to be adjusted can be rationally integrated.

Japanese Unexamined Patent Publication No. 2016-168650

In general, in robots such as industrial robots, precision polishing imitation motion and the like are required to be realized by force control, and improvement of dynamics performance such as settling motion or follow-up motion is constantly required.
On the other hand, in the conventional robot control device, the feedback system is configured in the main control unit. That is, in this robot control device, the feedback control calculation is performed by a component having a physical and communication distance from the robot. Therefore, the delay from the detection of torque by the torque sensor to the input of the command value to the motor becomes long. As a result, in this robot control device, there is inevitably more room for wasted time, which is a factor that suppresses high gain that can maintain stability. Moreover, since the waste time itself is not an element that can be canceled by lead compensation or the like, an adverse effect on the response time is unavoidable.
As described above, it is difficult to improve the force control performance (particularly quick response) in the conventional robot control device, and further improvement is required.

The present invention has been made to solve the above-mentioned problems, and an object of the present invention is to provide a robot control device capable of improving the force control performance with respect to a conventional configuration.

The robot control device according to the present invention calculates the command value of force control and the command value of the torque of each joint of the robot based on the command value of the force of the robot and the current value of the force, and the command value of the force control. Based on the main control unit that calculates the control command value for each joint based on the above, the current value of torque at the corresponding joint provided for each joint of the robot, and the command value of torque calculated by the main control unit. The joint control unit calculates the command value of the torque control in the joint, and calculates the command value for the motor provided in the joint based on the command value of the torque control and the control command value calculated by the main control unit. It is characterized by having a switching unit for switching between valid or invalid calculation of a command value for force control by the main control unit and calculation of a command value for torque control by the joint control unit.

According to the present invention, since the configuration is as described above, the force control performance can be improved as compared with the conventional configuration.

It is a figure which shows the configuration example of the robot control apparatus which concerns on Embodiment 1. FIG. It is a figure which shows the configuration example of the robot control apparatus which concerns on Embodiment 1. FIG. It is a flowchart which shows the operation example of the robot control apparatus which concerns on Embodiment 1. It is a figure explaining the operation example of the switching part in Embodiment 1. FIG. It is a flowchart which shows the operation example of the main control part in Embodiment 1. It is a flowchart which shows the operation example of the joint control part in Embodiment 1. It is a figure which shows the structural example of the robot control apparatus which concerns on Embodiment 2. It is a figure which shows the structural example of the robot control apparatus which concerns on Embodiment 2. It is a figure which shows the configuration example of the robot control apparatus which concerns on Embodiment 3. FIG. It is a figure which shows the configuration example of the robot control apparatus which concerns on Embodiment 3. FIG. It is a figure explaining the operation example of the switching part in Embodiment 3. FIG. It is a figure which shows the configuration example of the robot control apparatus which concerns on Embodiment 4. It is a figure which shows the configuration example of the robot control apparatus which concerns on Embodiment 5. It is a figure which shows an example of the learning data used in the learning part in Embodiment 5. It is a figure which shows the configuration example of the robot system including the conventional robot control device. It is a figure which shows the configuration example of the conventional robot control device. It is a figure which shows the configuration example of the conventional robot control device.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
Embodiment 1.
1 and 2 are diagrams showing a configuration example of the robot control device 1 according to the first embodiment. The relationship between the robot control device 1 and the robot 2 is the same as that in FIG. 15, and the description thereof will be omitted.
The robot control device 1 can control the position / posture and the force of the robot 2 at the same time (parallel). As shown in FIGS. 1 and 2, the robot control device 1 includes a main control unit (upper controller) 11 and a plurality of joint control units (lower controller) 15. The joint control unit 15 is provided for each joint of the robot 2. The main control unit 11 and each joint control unit 15 are connected by a communication line.

The main control unit 11 controls the entire robot 2 by outputting a command value to each joint control unit 15. Specifically, the main control unit 11 determines the torque for each joint based on the command value of the force of the robot 2, the command value of the position and posture, the current value of the angle of each joint and the current value of the torque of the robot 2. The command value and control command value of are calculated. In FIGS. 1 and 2, the control command value calculated by the main control unit 11 is the command value of the angular velocity. As shown in FIGS. 1 and 2, the main control unit 11 includes a disturbance torque estimation unit 111, a force calculation unit 112, a torque command value conversion unit 113, a force control unit 114, a position / attitude calculation unit 115, and a position / attitude control unit 116. It includes a command value synthesis unit (main side command value synthesis unit) 117, a command value conversion unit 118, and a switching unit 119. The main control unit 11 is realized by a processing circuit such as a system LSI (Large Scale Integration), a CPU (Central Processing Unit) that executes a program stored in a memory or the like, or the like.

Further, the position / attitude calculation unit 115, the position / attitude control unit 116, and the command value synthesis unit 117 are necessary when the robot control device 1 controls both the position / attitude and the force, and the robot control device 1 controls only the force. It is not necessary if you do.
Further, the disturbance torque estimation unit 111 is unnecessary depending on the configuration of the robot 2.

The disturbance torque estimation unit 111 estimates the disturbance torque (disturbance component) included in the current value of the torque for each joint based on the current value of the angle for each joint of the robot 2. In a general vertical articulated robot, the torque measured at each joint includes the torque due to gravity applied to the robot arm (gravitational torque). Therefore, the most important factor in estimating the disturbance torque in the disturbance torque estimation unit 111 is the estimation of the gravity torque. Further, the disturbance torque estimation unit 111 may estimate and add up disturbance torques other than gravity, such as inertial force or frictional force. The disturbance torque estimation unit 111 is not required for the robot 2 in which the gravitational torque is not applied to each joint.

The force calculation unit 112 converts the current value of the torque of each joint of the robot 2 into the current value of the force of the robot 2. The force calculation unit 112 has a subtractor 1121 and a coefficient multiplication unit 1122.

The subtractor 1121 subtracts the disturbance torque estimated by the disturbance torque estimation unit 111 from the current value of the torque. In FIG. 2, τ represents the current value of torque.
The coefficient multiplying unit 1122 multiplies the subtraction result by the subtractor 1121 by the inverse matrix of the transposition of the Jacobi matrix. In FIG. 2, J represents the Jacobian determinant.

The current value of torque includes gravity torque, but the command value of force is generally given in a form that does not include gravity. Therefore, in order to match this, the force calculation unit 112 subtracts the disturbance torque including the gravitational torque from the current value of the torque. Further, the current value of the torque is a value represented by the joint coordinate system, but since the force control unit 114 needs the current value of the force represented by the orthogonal coordinate system, it is converted by the force calculation unit 112.

The torque command value conversion unit 113 converts the force command value of the robot 2 into the torque command value of each joint of the robot 2. The torque command value conversion unit 113 includes a coefficient multiplying unit 1131 and an adder 1132.

The coefficient multiplying unit 1131 multiplies the command value of the force by the transposed matrix of the Jacobian determinant.
The adder 1132 adds the disturbance torque estimated by the disturbance torque estimation unit 111 to the multiplication result by the coefficient multiplication unit 1131.

The force command value is represented by the orthogonal coordinate system, and the torque command value conversion unit 113 converts the force command value into the torque command value represented by the joint coordinate system. Further, the current value of torque includes disturbance torque such as gravity torque. Therefore, the torque command value conversion unit 113 adds the disturbance torque to the command value of the torque in order to offset this. In FIG. 2, Fr represents a command value of force, and τr represents a command value of torque.

The force control unit 114 calculates a speed command value (force control command value) based on the force command value of the robot 2 and the current value of the force calculated by the force calculation unit 112, as in the conventional configuration. .. The force control unit 114 has a deviation calculator 1141 and a coefficient multiplication unit 1142.

The deviation calculator 1141 calculates the deviation between the command value of the force and the current value of the force. This operation is performed by subtracting the current value of the force from the command value of the force.
The coefficient multiplying unit 1142 obtains a command value of the speed by multiplying the deviation of the calculation result by the deviation calculator 1141 by the gain. In FIG. 2, _GF represents a gain.

The position / posture calculation unit 115 calculates the current value of the position / posture of the robot 2 based on the current value of the angle of each joint of the robot 2. The angle of each joint of the robot 2 is represented by the joint coordinate system, and the position / posture calculation unit 115 converts the angle of each joint into the position / posture represented by the orthogonal coordinate system. The current value of the angle of each joint of the robot 2 is detected by the encoder 24 provided for each joint. In FIG. 2, θ represents the current value of the angle, and X represents the current value of the position and orientation.

The position / posture control unit 116 calculates a speed command value (position / posture control command value) based on the position / posture command value of the robot 2 and the current position / posture value calculated by the position / posture calculation unit 115. The position / attitude control unit 116 has a deviation calculator 1161 and a coefficient multiplication unit 1162.

The deviation calculator 1161 calculates the deviation between the command value of the position / posture and the current value of the position / posture.
The coefficient multiplying unit 1162 obtains a command value of the speed by multiplying the deviation of the calculation result by the deviation calculator 1161 by the gain. In FIG. 2, Xr represents a command value of position and attitude, and G _Z represents a gain.

The command value synthesis unit 117 has a speed command value which is a force control command value calculated by the force control unit 114 and a speed command value which is a position / attitude control command value calculated by the position / attitude control unit 116. Are combined to make one speed command value. The command value synthesizer 117 has an adder 1171.
The adder 1171 adds the command value of the speed calculated by the force control unit 114 and the command value of the speed calculated by the position / attitude control unit 116.

The command value conversion unit 118 converts the synthesis result of the command value synthesis unit 117 into a command value of the angular velocity of each joint of the robot 2. The command value conversion unit 118 has a coefficient multiplication unit 1181.
The coefficient multiplication unit 1181 multiplies the Jacobian determinant inverse matrix by the composition result by the command value composition unit 117. That is, the command value conversion unit 118 converts the command value represented by the orthogonal coordinate system into the command value represented by the joint coordinate system. In FIG. 2, θ (dot) r represents the command value of the angular velocity.

The switching unit 119 switches the force control unit 114 of the main control unit 11 valid or invalid, and also switches the torque control unit 152 of the joint control unit 15 valid or invalid. The switching result by the switching unit 119 is notified to the main control unit 11 and the joint control unit 15, respectively, by communication or the like. As a result, the robot control device 1 can switch between an operation in which the force control is executed by the main control unit 11 as in the conventional configuration, or an operation in which the force control is executed by the joint control unit 15.

Note that FIGS. 1 and 2 show a case where the switching unit 119 is provided in the main control unit 11. However, the present invention is not limited to this, and the switching unit 119 may be provided in another place as long as it can communicate with the main control unit 11 and the joint control unit 15.

The joint control unit 15 controls the motor 21 provided in the corresponding joint in response to a command from the main control unit 11. Specifically, the joint control unit 15 is provided in the joint based on the current value of the torque and the current value of the angle in the corresponding joint, and the command value and the control command value of the torque calculated by the main control unit 11. The command value for the motor 21 is calculated. As shown in FIGS. 1 and 2, the joint control unit 15 includes a torque acquisition unit 151, a torque control unit 152, and a motor control unit 153. The motor control unit 153 has a joint angle control unit 154 and a command value synthesis unit (joint side command value synthesis unit) 155.

The torque acquisition unit 151 acquires the current value of the torque at the corresponding joint. The current value of the torque for each joint of the robot 2 is detected by the torque sensor 23 provided for each joint.

The torque control unit 152 calculates the command value for torque control based on the current value of the torque acquired by the torque acquisition unit 151 and the command value of the torque calculated by the main control unit 11. The torque control unit 152 has a subtractor 1521 and a PID control unit 1522.

The subtractor 1521 subtracts the current value of the torque acquired by the torque acquisition unit 151 from the command value of the torque at the corresponding joint among the command values of the torque calculated by the main control unit 11.
The PID control unit 1522 obtains a command value for torque control by performing PID control based on the subtraction result by the subtractor 1521.

The joint angle control unit 154 calculates the command value for angular velocity control based on the current value of the angle at the corresponding joint and the command value of the angular velocity calculated by the main control unit 11. The joint angle control unit 154 has a speed conversion unit 1541 and a speed control unit 1542. The speed control unit 1542 has a subtractor 1543 and a PI control unit 1544.

The velocity conversion unit 1541 converts the current value of the angle at the corresponding joint into the current value of the angular velocity.

The subtractor 1543 subtracts the current value of the angular velocity obtained by the speed conversion unit 1541 from the command value of the angular velocity at the corresponding joint among the command values of the angular velocity calculated by the main control unit 11.
The PI control unit 1544 obtains a command value for angular velocity control by performing PI control based on the subtraction result by the subtractor 1543.

The command value synthesis unit 155 synthesizes the torque control command value calculated by the torque control unit 152 and the angular velocity control command value calculated by the joint angle control unit 154. In FIG. 1, the command value synthesizing unit 155 has an adder 1551.
The adder 1551 adds the torque control command value calculated by the torque control unit 152 and the angular velocity control command value calculated by the joint angle control unit 154. The command value (current command value), which is the result of synthesis by the command value synthesis unit 155, is output to the motor 21.

Next, an operation example of the robot control device 1 according to the first embodiment shown in FIGS. 1 and 2 will be described with reference to FIG.
In the operation example of the robot control device 1 according to the first embodiment shown in FIGS. 1 and 2, first, as shown in FIG. 3, the switching unit 119 enables or disables the force control unit 114 and sets the torque control unit 152. It is enabled or invalidated (step ST301). The switching result by the switching unit 119 is notified to the main control unit 11 and the joint control unit 15, respectively, by communication or the like.

The switching unit 119 enables or disables the force control unit 114 and enables or disables the torque control unit 152, for example, as shown in FIG. 4, predefines in association with the control mode and executes the process accordingly. For example, when a command for designating the control mode "1" is executed in the program for operating the robot 2, the switching unit 119 enables the force control unit 114 and disables the torque control unit 152 thereafter. Alternatively, a storage device for storing data indicating the control mode of the robot control device 1 is provided inside or outside the robot control device 1, and the switching unit 119 periodically reads data indicating the control mode from the storage device and the data thereof. The force control unit 114 and the torque control unit 152 may be set to be valid or invalid according to the above.

Next, the main control unit 11 determines the command value of the torque for each joint based on the command value of the force of the robot 2, the command value of the position and posture, the current value of the angle of each joint and the current value of the torque of the robot 2. And the command value of the angular velocity is calculated (step ST302). When the torque control unit 152 is invalidated by the switching unit 119, the torque command value is unnecessary, so that the main control unit 11 does not have to calculate the torque command value.

Next, the joint control unit 15 is provided in the corresponding joint based on the current value of the torque and the current value of the angle in the corresponding joint, and the command value of the torque and the command value of the angular velocity calculated by the main control unit 11. The command value for the motor 21 is calculated (step ST303).

Next, an operation example of the main control unit 11 shown in FIGS. 1 and 2 will be described with reference to FIG.
In the operation example of the main control unit 11 shown in FIGS. 1 and 2, first, as shown in FIG. 5, the disturbance torque estimation unit 111 is possessed by the robot 2 based on the current value of the angle of each joint of the robot 2. The disturbance torque applied to the joint is estimated (step ST501). In a vertical articulated robot, the disturbance torque usually includes a component of torque due to gravity (gravitational torque). Therefore, the disturbance torque estimation unit 111 estimates the gravity torque from the current value of the angle and outputs the value. Since the method of estimating the gravitational torque from the current value of the angle is a known technique, the description thereof will be omitted. Further, the disturbance torque estimation unit 111 may estimate disturbance torques caused by other than gravity such as inertial force or frictional force, and add those values to the output value.

Next, the torque command value conversion unit 113 converts the force command value of the robot 2 into the torque command value of each joint of the robot 2 (step ST502). In FIGS. 1 and 1, first, the coefficient multiplying unit 1131 multiplies the commanded value of the force by the transposed matrix of the Jacobian determinant. Since the Jacobian determinant changes depending on the angle of the joint of the robot 2, it is necessary to update it as appropriate. Then, the adder 1132 adds the disturbance torque estimated by the disturbance torque estimation unit 111 to the multiplication result by the coefficient multiplication unit 1131. As a result, the disturbance torque such as the gravitational torque is canceled between the command value of the torque and the current value of the torque, and desired control becomes possible.

Further, the force calculation unit 112 converts the current value of the torque of each joint of the robot 2 into the current value of the force of the robot 2 (step ST503). In FIGS. 1 and 2, the subtractor 1121 subtracts the disturbance torque calculated by the disturbance torque estimation unit 111 from the current value of the torque, and the coefficient multiplying unit 1122 transposes the Jacob matrix with respect to the subtraction result by the subtractor 1121. Multiply the inverse matrix of. In a vertical articulated robot, the current value of the torque acquired by the torque acquisition unit 151 usually includes a torque component due to gravity. Therefore, the force calculation unit 112 may obtain an estimated external force excluding the disturbance torque as the current value of the force by subtracting and removing the disturbance torque including the gravity torque before converting the torque into the force. can.

Next, the force control unit 114 calculates a speed command value (force control command value) based on the force command value of the robot 2 and the current value of the force obtained by the force calculation unit 112 (step ST504). .. In FIGS. 1 and 1, the deviation calculator 1141 obtains the deviation between the command value of the force and the current value of the force by calculation, and the coefficient multiplying unit 1142 gains the deviation of the calculation result by the deviation calculator 1141. By multiplying by, the command value of the speed is obtained.
When the force control unit 114 is invalidated by the switching unit 119, for example, the force control unit 114 omits the above calculation and outputs 0.

Further, the position / posture calculation unit 115 calculates the current value of the position / posture of the robot 2 based on the current value of the angle of each joint of the robot 2 (step ST505).

Next, the position / posture control unit 116 calculates a speed command value (position / posture control command value) based on the position / posture command value of the robot 2 and the current position / posture value calculated by the position / posture calculation unit 115. (Step ST506). In FIGS. 1 and 2, the deviation calculator 1161 calculates the deviation between the command value of the position / posture and the current value of the position / posture, and the coefficient multiplying unit 1162 calculates the deviation of the calculation result by the deviation calculator 1161. By multiplying the gain, the command value of the speed is obtained. The deviation of the position is obtained by subtracting the coordinate value of the current value from the coordinate value of the command value. The attitude deviation can be obtained by obtaining the rotation conversion from the attitude of the current value to the attitude of the command value.

Next, the command value synthesis unit 117 synthesizes the speed command value calculated by the force control unit 114 and the speed command value calculated by the position / attitude control unit 116 to obtain one speed command value (step ST507). ).

Next, the command value conversion unit 118 converts the synthesis result by the command value synthesis unit 117 into a command value of the angular velocity for each joint of the robot 2 (step ST508). In FIGS. 1 and 2, the coefficient multiplication unit 1181 obtains the command value of the angular velocity for each joint by multiplying the synthesis result by the command value synthesis unit 117 by the inverse matrix of the Jacobian determinant.

In the above, as a method of disabling the force control unit 114, a method of setting the output of the force control unit 114 to 0 has been shown, but the present invention is not limited to this as long as the same effect can be obtained. For example, the force control unit 114 sets the gain of the coefficient multiplication unit 1142 to 0, or the command value synthesis unit 117 outputs the command value of the speed calculated by the position / attitude control unit 116 as it is, and the force control unit 114 calculates it. The force control unit 114 can be invalidated by means such as not combining with the command value of the set speed.

Next, an operation example of the joint control unit 15 shown in FIGS. 1 and 2 will be described with reference to FIG.
In the operation example of the joint control unit 15 shown in FIGS. 1 and 2, as shown in FIG. 6, the torque acquisition unit 151 first acquires the current value of the torque at the corresponding joint (step ST601).

Next, the torque control unit 152 calculates the command value for torque control based on the current value of the torque acquired by the torque acquisition unit 151 and the command value of the torque calculated by the main control unit 11 (step ST602). In FIGS. 1 and 2, the subtractor 1521 subtracts the current value of the torque acquired by the torque acquisition unit 151 from the command value of the torque at the corresponding joint among the command values of the torque calculated by the main control unit 11. Then, the PID control unit 1522 performs PID control based on the subtraction result by the subtractor 1521, and obtains a command value for torque control.
When the torque control unit 152 is invalidated by the switching unit 119, for example, the torque control unit 152 omits the above calculation and outputs 0.

Further, the joint angle control unit 154 calculates the command value of the angular velocity control based on the command value of the angular velocity calculated by the main control unit 11 (step ST603). In FIGS. 1 and 2, the speed conversion unit 1541 converts the current value of the angle at the corresponding joint into the current value of the angular velocity, and the subtractor 1543 is among the command values of the angular velocity calculated by the main control unit 11. The current value of the angular velocity obtained by the speed conversion unit 1541 is subtracted from the command value of the angular velocity at the joint, and the PI control unit 1544 performs PI control based on the subtraction result by the subtractor 1543 to control the angular velocity. Get the command value.

Next, the command value synthesizing unit 155 synthesizes the torque control command value calculated by the torque control unit 152 and the angular velocity control command value calculated by the joint angle control unit 154 (step ST604). In FIGS. 1 and 2, the adder 1551 adds the torque control command value calculated by the torque control unit 152 and the angular velocity control command value calculated by the joint angle control unit 154. The command value (current command value), which is the result of synthesis by the command value synthesis unit 155, is output to the motor 21.

In the above, as a method of disabling the torque control unit 152, a method of setting the output of the torque control unit 152 to 0 has been shown, but the method is not limited to this as long as the same effect can be obtained. For example, the torque control unit 152 sets the gain of the PID control unit 1222 to 0, or the command value synthesis unit 155 outputs the command value of the angular velocity control calculated by the joint angle control unit 154 as it is, and the torque control unit 152 outputs the command value as it is. The torque control unit 152 can be invalidated by means such as not combining the calculated torque control command value with the calculated torque control command value.

Further, in the robot control device 1 according to the first embodiment, both the force control unit 114 and the torque control unit 152 can be enabled as shown in the control mode “3” of FIG. However, in this case, it is preferable to prevent the two controls from interfering with each other. For example, the force control unit 114 controls in the low frequency band, the torque control unit 152 controls in the high frequency band, and the like, and the frequency bands to be controlled are divided so as not to affect each other. Is preferable.

Further, the current command value output to the motor 21 may fluctuate greatly depending on the timing at which the switching unit 119 executes switching. In order to avoid this, the robot control device 1 may perform smoothing processing by a low-pass filter or the like at the time of switching by the switching unit 119 to suppress fluctuations in the current command value.

Next, the effect of the robot control device 1 according to the first embodiment will be described.
As described above, in the conventional robot control device 1b, the feedback system is configured by the main control unit 11b. That is, in this robot control device 1b, the feedback control calculation is performed by a component having a physical and communication distance from the robot 2. Therefore, the delay from the detection of torque by the torque sensor 23 to the input of the command value to the motor 21 becomes long. As a result, in this robot control device 1b, there is inevitably more room for wasted time, which is a factor that suppresses high gain that can maintain stability. Moreover, since the waste time itself is not an element that can be canceled by lead compensation or the like, an adverse effect on the response time is unavoidable.

On the other hand, in the robot control device 1 according to the first embodiment, when the switching unit 119 performs force control using the torque control unit 152, the joint control unit 15 performs a calculation for torque control. Therefore, the delay from the detection of the torque by the torque sensor 23 to the input of the command value to the motor 21 is shortened, and the room for wasted time can be reduced. That is, the robot control device 1 according to the first embodiment can also perform adjustment corresponding to high gain of the controller that can maintain stability (gain adjustment of one-variable control for each joint). As described above, the robot control device 1 according to the first embodiment can improve the force control performance (particularly quick response) with respect to the conventional configuration.

Further, in the robot control device 1 according to the first embodiment, when the switching unit 119 performs force control using the force control unit 114, the main control unit 11 calculates the force control as in the conventional configuration. Therefore, the merits of the conventional configuration can be obtained, such as the ability to control the 6 axes of the hand by coordinating all the joints and the rational integration of the parameters to be adjusted. That is, if it is desired to prioritize the merit of the conventional configuration over the improvement of the quick response of the force control, the switching unit 119 can be used in the same manner as the conventional configuration.

As described above, in the robot control device 1, the switching unit 119 enables the torque control unit 152 of the joint control unit 15 when a high-speed response to the force is required. If you want to control precisely, you can select a configuration that matches the performance required for each situation, such as enabling the force control unit 114 of the main control unit 11.

As described above, according to the first embodiment, the robot control device 1 has a command value for force control and a torque for each joint of the robot 2 based on the command value of the force of the robot 2 and the current value of the force. The main control unit 11 that calculates the command value of the above and calculates the control command value for each joint based on the command value of the force control, and the current torque in the corresponding joint provided for each joint of the robot 2. The command value of the torque control at the joint is calculated based on the value and the command value of the torque calculated by the main control unit 11, and is based on the command value of the torque control and the control command value calculated by the main control unit 11. The joint control unit 15 that calculates the command value for the motor 21 provided in the joint, the calculation of the force control command value by the main control unit 11, and the calculation of the torque control command value by the joint control unit 15 are valid or invalid. It is equipped with a switching unit 119 for switching to. As a result, the robot control device 1 according to the first embodiment can improve the force control performance with respect to the conventional configuration. Further, the robot control device 1 can exhibit the required performance even when the conventional configuration has higher performance.

In the first embodiment, the case where the joint angle control unit 154 is provided in the joint control unit 15 is shown. However, the present invention is not limited to this, and the joint angle control unit 154 may be provided in the main control unit 11. For example, if the robot 2 operates at a low speed, it is considered that the influence of the joint angle control unit 154 provided on the main control unit 11 is small.
Further, the joint angle control unit 154 is not an essential configuration and may be removed from the robot control device 1.

Embodiment 2.
In the first embodiment, the joint control unit 15 calculates the command value of the angular velocity control using the command value of the angular velocity calculated by the main control unit 11, and then calculates the command value of the torque control and the command value of the angular velocity control. The case of synthesizing is shown. However, not limited to this, the joint control unit 15 synthesizes the angular velocity command value and the torque control command value at the corresponding joint among the angular velocity command values calculated by the main control unit 11, and then synthesizes the command values. The command value of the angular velocity control may be calculated using the result.
7 and 8 are diagrams showing a configuration example of the robot control device 1 according to the second embodiment. The robot control device 1 according to the second embodiment shown in FIGS. 7 and 8 commands the joint angle control unit 154 and the command value synthesis unit 155 to the robot control device 1 according to the first embodiment shown in FIGS. 1 and 2. The value synthesis unit (joint side command value synthesis unit) 156 and the joint angle control unit 157 have been changed. Other configurations of the robot control device 1 according to the second embodiment are the same as those of the robot control device 1 according to the first embodiment, and the same reference numerals are given and the description thereof will be omitted.

The command value synthesis unit 156 synthesizes the command value of the angular velocity at the corresponding joint among the command values of the angular velocity calculated by the main control unit 11 and the command value of the torque control calculated by the torque control unit 152. In FIG. 8, the command value synthesizing unit 156 has an adder 1561.
The adder 1561 adds the command value of the angular velocity at the corresponding joint among the command values of the angular velocity calculated by the main control unit 11 and the command value of the torque control calculated by the torque control unit 152.

The joint angle control unit 157 calculates the command value for angular velocity control based on the current value of the angle at the corresponding joint and the synthesis result by the command value synthesis unit 156. The joint angle control unit 157 includes a speed conversion unit 1571 and a speed control unit 1572. The speed control unit 1572 has a subtractor 1573 and a PI control unit 1574.

The velocity conversion unit 1571 converts the current value of the angle at the corresponding joint into the current value of the angular velocity.

The subtractor 1573 subtracts the current value of the angular velocity obtained by the speed conversion unit 1571 from the synthesis result by the command value synthesis unit 156.
The PI control unit 1574 obtains a command value for angular velocity control by performing PI control based on the subtraction result by the subtractor 1573.
The command value (current command value) of the angular velocity control calculated by the joint angle control unit 157 is output to the motor 21 provided in the corresponding joint.

Similar to the switching unit 119 in the first embodiment, the switching unit 119 in the second embodiment switches the force control unit 114 of the main control unit 11 to valid or invalid, and enables or invalidates the torque control unit 152 of the joint control unit 15. Switch to invalid. Similar to the first embodiment, the force control unit 114 and the torque control unit 152 in the second embodiment are disabled by setting the output to 0 when disabled by the switching section 119. To torque.

As described above, in the robot control device 1 according to the second embodiment, the joint control unit 15 controls the angular velocity command value and the torque control at the corresponding joint among the angular velocity command values calculated by the main control unit 11. The command values are combined, and the angular velocity control is performed based on the combined result. The same effect as that of the robot control device 1 according to the first embodiment can be obtained for the robot control device 1 according to the second embodiment. Further, the robot control device 1 according to the second embodiment has a correspondence relationship close to that of the conventional compliance control.

Embodiment 3.
In the first and second embodiments, the switching unit 119 shows the case where the force control unit 114 and the torque control unit 152 are switched between valid and invalid. However, not limited to this, the joint control unit 15 is provided with a function of controlling the position and posture, and the switching unit 119 is added to the above, and the position and attitude control unit 116 and the joint control unit 15 of the main control unit 11 are added. The position / attitude control may be enabled or disabled.
9 and 10 are diagrams showing a configuration example of the robot control device 1 according to the third embodiment. The robot control device 1 according to the third embodiment shown in FIGS. 9 and 10 has an inverse kinematics calculation unit 120 added to the main control unit 11 with respect to the robot control device 1 according to the second embodiment shown in FIGS. 7 and 8. However, the angle control unit 158 is added to the joint control unit 15. Further, the command value synthesis unit 156 is changed to the command value synthesis unit (joint side command value synthesis unit) 159, and the command value synthesis unit 159 is also connected to the angle control unit 158. Further, as described above, the switching unit 119 has an additional function of switching the position / posture control to valid or invalid with respect to the switching unit 119 in the second embodiment. Other configurations of the robot control device 1 according to the third embodiment are the same as those of the robot control device 1 according to the second embodiment, and the same reference numerals are given and the description thereof will be omitted.

The inverse kinematics calculation unit 120 calculates the command value of the angle for each joint so as to realize the command value of the position / posture based on the command value of the position / posture of the robot 2. Since this operation is known as inverse kinematics, its description will be omitted. If the time series of the command values of the position and the posture is obtained in advance, the inverse kinematics calculation unit 120 may perform the above calculation offline in advance. Further, the inverse kinematics calculation unit 120 may perform the above calculation each time a command value of the position / posture is given.

The angle control unit 158 controls the angle of the joint to control the angle of the joint based on the current value of the angle at the corresponding joint and the command value of the angle calculated by the inverse kinematics calculation unit 120. Obtain the command value by calculation. The angle control unit 158 has a subtractor 1581 and a P control unit 1582.

The subtractor 1581 subtracts the current value of the angle from the command value of the angle at the corresponding joint among the command values of the angle calculated by the inverse kinematics calculation unit 120.
The P control unit 1582 obtains a command value for angle control by performing P control (proportional control) based on the subtraction result by the subtractor 1581.

The command value synthesis unit 159 includes the command value of the angular velocity at the corresponding joint among the command values of the angular velocity calculated by the main control unit 11, the command value of the torque control calculated by the torque control unit 152, and the angle control. The command value of the angle control calculated by the unit 158 is synthesized. In FIG. 10, the command value synthesizing unit 159 has an adder 1591.
The adder 1591 includes an angular velocity command value at the corresponding joint among the angular velocity command values calculated by the main control unit 11, a torque control command value calculated by the torque control unit 152, and an angle control unit 158. Adds the command value of the angle control calculated by.

Then, the speed control unit 1572 controls the angular velocity of the corresponding joint based on the synthesis result (command value of the angular velocity) by the command value synthesis unit 159.

The switching unit 119 switches the position / attitude control unit 116 and the angle control unit 158 valid or invalid in addition to the force control unit 114 and the torque control unit 152. FIG. 11 shows an example of the correspondence between the control mode and the validity or invalidity of each control unit (force control unit 114, torque control unit 152, position / attitude control unit 116, and angle control unit 158). As shown in FIG. 11, it is preferable to enable the position / attitude control unit 116 when the force control unit 114 is enabled, and to enable the angle control unit 158 when the force control unit 114 is disabled. However, other combinations are also possible.

Next, when the position / attitude control unit 116 is disabled by the switching unit 119 and when the angle control unit 158 is disabled by the switching unit 119, the respective invalidation methods will be described.

When disabling the position / attitude control unit 116, for example, the position / attitude control unit 116 may always output 0 regardless of the calculation result of the position / attitude control unit 116. Alternatively, the same effect can be obtained even if the position / attitude control unit 116 sets the gain of the coefficient multiplication unit 1162 to 0. In addition, the operation of the command value synthesis unit 117 is changed, the calculation result of the force control unit 114 is output to the command value conversion unit 118 as it is, and the calculation result of the position / attitude control unit 116 is calculated by the main control unit 11. It may not be reflected in the command value of the angular velocity. When both the force control unit 114 and the position / attitude control unit 116 are invalidated, the command value synthesis unit 117 can always output 0 to invalidate both control operations. Further, the position / attitude control unit 116 may stop its own operation.

When disabling the angle control unit 158, for example, the angle control unit 158 may always output 0 regardless of the calculation result of the angle control unit 158. Alternatively, the same effect can be obtained even if the angle control unit 158 sets the gain of the P control unit 1582 to 0. In addition, the operation of the command value synthesis unit 159 is changed, and among the command values of the angular speed calculated by the main control unit 11, the command value of the angular speed at the corresponding joint and the torque control calculated by the torque control unit 152 are used. Only the command value may be combined so that the calculation result of the angle control unit 158 is not reflected in the control of the speed control unit 1572. Further, the angle control unit 158 may stop its own operation.

Further, the switching unit 119 not only switches the above four control units valid or invalid, but may also switch control settings such as the gain of each control unit at the same time.

In the above, the torque control command value calculated by the torque control unit 152 is the angular velocity command value at the corresponding joint among the angular velocity command values calculated by the main control unit 11, and the angle control unit 158. The case where it is combined with the command value of the angle control calculated by is shown. However, the present invention is not limited to this, and the torque control command value may be combined with the angular velocity control command value calculated by the speed control unit 1542, as in the robot control device 1 according to the first embodiment. .. In this case, the joint control unit 15 has the command value of the angular velocity at the corresponding joint among the command values of the angular velocity calculated by the main control unit 11 and the command value of the angle control calculated by the angle control unit 158. Is synthesized, and the speed control unit 1542 controls the angular velocity at the joint based on the synthesis result.

In the robot control device 1 according to the third embodiment, in addition to the force control, the position and attitude control can be switched between the main control unit 11 and the joint control unit 15. As a result, the robot control device 1 according to the third embodiment can be expected to have the effect of improving the control performance and facilitating the adjustment as compared with the robot control device 1 according to the first and second embodiments.

Embodiment 4.
In the first to third embodiments, the operation of the switching unit 119 is directly specified from the program for operating the robot 2, or manually written in a storage device provided inside or outside the robot control device 1. It was supposed to be specified. However, the operation is not limited to this, and the operation of the switching unit 119 can be automatically specified.

FIG. 12 is a diagram showing a configuration example of the robot control device 1 according to the fourth embodiment. The robot control device 1 according to the fourth embodiment shown in FIG. 12 has an analysis unit 121 and a determination unit 122 added to the robot control device 1 according to the first to third embodiments shown in FIGS. 1, 7 and 9. There is. Other configurations of the robot control device 1 according to the fourth embodiment are the same as those of the robot control device 1 according to the first to third embodiments, and the same reference numerals are given and the description thereof will be omitted. In FIG. 12, among the configurations of the main control unit 11, the configurations other than the switching unit 119, the analysis unit 121, and the determination unit 122 are omitted.

The analysis unit 121 analyzes one or more of the work contents performed by the robot 2, the program for operating the robot 2, and the control settings of the robot 2. The analysis unit 121 has one or more of the work analysis unit 1211, the program analysis unit 1212, and the control setting analysis unit 1213. FIG. 12 shows a case where the analysis unit 121 has a work analysis unit 1211, a program analysis unit 1212, and a control setting analysis unit 1213.

The work analysis unit 1211 analyzes the work that the robot 2 intends to perform. That is, the work analysis unit 1211 acquires which force work (for example, polishing, cutting, painting, or fitting) the robot 2 will perform next by communication or the like.

The program analysis unit 1212 analyzes the command of the program for operating the robot 2 or its argument. That is, the program analysis unit 1212 acquires a command or an argument thereof to be executed next by the robot 2 by communication or the like.

The control setting analysis unit 1213 analyzes the control setting such as the gain of the force control. That is, the control setting analysis unit 1213 acquires setting information such as gain or compliance currently set in the robot control device 1 by communication or the like.

The determination unit 122 determines the switching by the switching unit 119 based on the analysis result by the analysis unit 121.

Here, when the work to be executed by the robot 2 is analyzed by the work analysis unit 1211, the determination unit 122 determines which control mode is appropriate based on the analysis result and a predetermined rule. .. For example, when the work analysis unit 1211 analyzes that the next work of the robot 2 is polishing, the determination unit 122 enables the force control unit 114 and disables the torque control unit 152.

Further, when the command of the program for operating the robot 2 or its argument is analyzed by the program analysis unit 1212, the determination unit 122 determines which control mode is appropriate based on the analysis result and a predetermined rule. judge. For example, the determination unit 122 enables the force control unit 114 and disables the torque control unit 152 when the program analysis unit 1212 acquires a command for the robot 2 to execute the next pressing. Further, for example, when the program analysis unit 1212 obtains a command for the robot 2 to execute face-to-face contact, the determination unit 122 invalidates the force control unit 114 and enables the torque control unit 152. If the appropriate switching differs depending on the argument even with the same command, the determination unit 122 determines the switching of the switching unit 119 including the argument.

Further, when the control setting analysis unit 1213 analyzes the control setting such as the gain of the force control, the determination unit 122 determines which control mode is appropriate based on the analysis result and the predetermined rule. judge. For example, when the control setting analysis unit 1213 analyzes that the force control gain ( _GF ) is set in only one specific direction in the Cartesian coordinate system, the determination unit 122 cooperates with each other on the six axes of the Cartesian coordinate system. Since it is considered that precise force control is required, the force control unit 114 is enabled and the torque control unit 152 is disabled. On the other hand, when the determination unit 122 does not require the precise force control as described above and it is sufficient to increase the compliance in all joints, the force control unit 114 is disabled and the torque control unit 152 is effective. To.

In the above, the case where the determination unit 122 determines the operation of the switching unit 119 based on one analysis result is shown. However, not limited to this, the determination unit 122 comprehensively determines the operation of the switching unit 119 from the plurality of analysis results, or determines the operation of the switching unit 119 by taking a majority vote of the plurality of analysis results. The operation may be determined by a method.

The determination result by the determination unit 122 is transmitted to the switching unit 119 by communication. Then, the switching unit 119 switches the operations of the main control unit 11 and the joint control unit 15 based on the determination result. As a result, which control unit performs force control (and position / attitude control) is switched, and the behavior of the robot control device 1 changes.

In the robot control device 1 according to the fourth embodiment, the user does not have to manually specify which control unit is enabled or disabled for the robot control device 1 according to the first to third embodiments. Since appropriate switching is automatically performed, there is an effect that it is easy for the user to use.

Embodiment 5.
In the robot control device 1 according to the fourth embodiment, the case where the operation of the switching unit 119 is automated is shown. However, in this robot control device 1, it is necessary to determine in advance the operation of the determination unit 122 that determines the operation of the switching unit 119. On the other hand, it is also possible to automatically determine the operation of the determination unit 122 by learning.

FIG. 13 is a diagram showing a configuration example of the robot control device 1 according to the fifth embodiment. The robot control device 1 according to the fifth embodiment shown in FIG. 13 has a learning unit 123 and a robot simulator 124 added to the robot control device 1 according to the fourth embodiment shown in FIG. Other configurations of the robot control device 1 according to the fifth embodiment are the same as those of the robot control device 1 according to the fourth embodiment, and the same reference numerals are given and the description thereof will be omitted. In FIG. 13, among the configurations of the main control unit 11, the configurations other than the switching unit 119, the analysis unit 121, the determination unit 122, the learning unit 123, and the robot simulator 124 are omitted.

The learning unit 123 learns the relationship between one or more of the work contents performed by the robot 2, the program for operating the robot 2, and the control settings of the robot 2 and the performance value in the operation of the robot 2 by switching the switching unit 119. do.

At this time, the learning unit 123 first prepares a large number of combinations of one or more of the work contents performed by the robot 2, the program for operating the robot 2, and the control settings of the robot 2, and the combination thereof is used as a robot simulator. The robot 2 of 124 is operated in a simulated manner, and the command value of the position and orientation of the robot 2, the current value of the position and orientation, the command value of the force, the current value of the force, and the like are acquired. The learning unit 123 performs this process for all combinations that can be executed by the switching unit 119. Then, the learning unit 123 evaluates the performance using the deviation between the command value of the position / posture and the current value of the position / posture, the deviation between the command value of the force and the current value of the force, and calculates the performance value. ..

More specifically, the learning unit 123 determines the work content performed by the robot 2, the trajectory pattern of the position command based on the work start position, the time series pattern of the force command value from the start of the work, or the tool coordinate system. The relationship between the compliance setting of each axis as a reference and the performance value for each control mode of the switching unit 119 is learned by supervised learning. For example, as shown in FIG. 14, the learning unit 123 obtains and collects performance values for each control mode for each work content, position command trajectory, force command time series, and compliance setting. The learning unit 123 performs the above processing for a large number of cases. Although the robot simulator 124 is not an indispensable configuration, it is better to perform the learning by the robot simulator 124 instead of the actual machine because the learning by the learning unit 123 requires a very large amount of data.

When the learning by the learning unit 123 is successful, based on the learning result, the work content of the robot 2 to be executed, the program for operating the robot 2, or the control setting of the robot 2 is selected for each control mode of the switching unit 119. It is possible to predict the performance value of. Therefore, the determination unit 122 determines the switching by the switching unit 119 based on the learning result (performance value) by the learning unit 123. That is, the determination unit 122 determines that the control mode predicted to produce the highest performance is the best, and notifies the switching unit 119 of the result. Then, the switching unit 119 switches the control unit in the robot control device 1 valid or invalid based on the notification.

If it is difficult to make an appropriate determination because there is no similar case in the learned data, the determination unit 122 is not determined based on learning, but is determined in advance by a person as in the fourth embodiment. The judgment may be made according to the above rules.

In the robot control device 1 according to the fifth embodiment, even if the operation of the determination unit 122 is not determined in advance by a person, appropriate switching is automatically determined based on the learned result. Therefore, the robot control device 1 according to the fifth embodiment can easily determine a switching rule with respect to the robot control device 1 according to the fourth embodiment, and can be expected to have an effect that the control performance is further improved.

Further, in the present invention, within the scope of the invention, it is possible to freely combine each embodiment, modify any component of each embodiment, or omit any component in each embodiment. be. For example, in the first to fifth embodiments, the joint angle control unit controls the speed, but it is also possible to control the position and posture through the control of other physical quantities such as acceleration or current. ..

1 Robot control device 2 Robot 11 Main control unit 15 Joint control unit 21 Motor 22 Sensor 23 Torque sensor 24 Encoder 111 Disturbance torque estimation unit 112 Force calculation unit 113 Torque command value conversion unit 114 Force control unit 115 Position / posture calculation unit 116 Position / posture Control unit 117 Command value synthesis unit (Main side command value composition unit)
118 Command value conversion unit 119 Switching unit 120 Inverse kinematics calculation unit 121 Analysis unit 122 Judgment unit 123 Learning unit 124 Robot simulator 151 Torque acquisition unit 152 Torque control unit 153 Motor control unit 154 Joint angle control unit 155 Command value synthesis unit (joint) Side command value synthesizer)
156 Command value synthesis unit (joint side command value synthesis unit)
157 Joint angle control unit 158 Angle control unit 159 Command value synthesis unit (joint side command value synthesis unit)
1121 subtractor 1122 coefficient multiplying unit 1131 coefficient multiplying unit 1132 adder 1141 deviation calculator 1142 coefficient multiplying unit 1161 deviation adder 1162 coefficient multiplying unit 1171 adder 1181 coefficient multiplying unit 1211 work analysis unit 1212 program analysis unit 1213 control setting analysis unit 1521 Adder 1522 PID Control 1541 Speed Converter 1542 Speed Control 1543 Adder 1544 PI Control 1551 Adder 1561 Adder 1571 Speed Converter 157 2 Speed Control 1573 Subtractor 1574 PI Control 1581 P Control 1582 P Control 1591 adder

Claims (9)

The command value of force control and the command value of torque for each joint of the robot are calculated based on the command value of the force of the robot and the current value of the force, and the control command for each joint is calculated based on the command value of the force control. The main control unit that calculates the value and
A command value for torque control at the joint is calculated based on the current value of the torque at the corresponding joint and the command value of the torque calculated by the main control unit, which is provided for each joint of the robot, and the torque is calculated. A joint control unit that calculates a command value for a motor provided in the joint based on a control command value and a control command value calculated by the main control unit, and a joint control unit.
A robot control device including a switching unit for switching between valid or invalid calculation of a command value for force control by the main control unit and calculation of a command value for torque control by the joint control unit. The main control unit
A disturbance torque estimation unit that estimates a disturbance torque including gravity torque based on the current value of the angle of each joint of the robot, and a disturbance torque estimation unit.
A force calculation unit that calculates the current value of the robot's force based on the current value of the torque of each joint of the robot and the disturbance torque estimated by the disturbance torque estimation unit.
It is characterized by having a torque command value conversion unit that calculates a command value of torque for each joint of the robot based on a command value of the force of the robot and a disturbance torque estimated by the disturbance torque estimation unit. The robot control device according to claim 1. The joint control unit
It is provided with a joint-side command value synthesizer that obtains a command value for the motor by synthesizing the control command value at the corresponding joint among the command value of torque control and the control command value calculated by the main control unit. The robot control device according to claim 1 or 2, wherein the robot control device is characterized by the above. The main control unit calculates the command value of the angular velocity as the control command value, and then
The joint control unit
A joint angle control unit that calculates an angular velocity control command value at a corresponding joint based on an angular velocity command value calculated by the main control unit.
The claim is characterized by comprising a joint-side command value synthesizing unit that obtains a command value for the motor by synthesizing a torque control command value and an angular velocity control command value calculated by the joint angle control unit. 1 or the robot control device according to claim 2. The main control unit calculates the command value of the angular velocity as the control command value, and then
The joint control unit
A joint-side command value synthesizer that synthesizes the command value of the angular velocity at the corresponding joint among the command value of the torque control and the command value of the angular velocity calculated by the main control unit.
A claim characterized by being provided with a joint angle control unit that obtains a command value for the motor by calculating a command value for angular speed control at a corresponding joint based on a synthesis result by the joint side command value synthesis unit. Item 1 or the robot control device according to claim 2. The main control unit
A position / posture calculation unit that calculates the current value of the position / posture of the robot based on the current value of the angle of each joint of the robot.
An inverse kinematics calculation unit that calculates the command value of the angle of each joint of the robot based on the command value of the position and posture.
A position / posture control unit that calculates a position / posture control command value based on a position / posture command value and a position / posture current value calculated by the position / posture calculation unit.
It is equipped with a main-side command value synthesis unit that obtains control command values for each joint possessed by the robot by synthesizing the command values for force control and the command values for position / attitude control calculated by the position / attitude control unit.
The joint control unit
An angle control unit that calculates the command value for angle control at the corresponding joint based on the current value of the angle at the corresponding joint and the command value of the angle calculated by the inverse kinematics calculation unit.
Joint side command value synthesis that synthesizes the control command value at the corresponding joint among the command value of torque control, the command value of angle control calculated by the angle control unit, and the control command value calculated by the main control unit. With a part,
The switching unit calculates a command value for force control and a command value for position / attitude control by the main control unit, and also calculates a command value for torque control and a command value for angle control by the joint control unit. The robot control device according to any one of claims 1 to 5, wherein the robot control device is switched between valid and invalid. An analysis unit that analyzes one or more of the work contents performed by the robot, the program for operating the robot, and the control settings of the robot.
The robot control device according to any one of claims 1 to 6, further comprising a determination unit for determining switching by the switching unit based on an analysis result by the analysis unit. A learning unit that learns the relationship between one or more of the work contents performed by the robot, the program for operating the robot, and the control settings of the robot, and the performance value in the operation of the robot by switching the switching unit. When,
The robot control device according to claim 7, wherein the determination unit determines switching by the switching unit based on the performance value learned by the learning unit. It is a robot control method using a robot control device including a main control unit and a joint control unit provided for each joint of the robot.
The main control unit calculates a force control command value and a torque command value for each joint of the robot based on the force command value of the robot and the current force value, and is based on the force control command value. The control command value for each joint is calculated.
The joint control unit calculates a torque control command value at the joint based on the current value of the torque at the corresponding joint and the torque command value calculated by the main control unit, and the torque control command value. And, based on the control command value calculated by the main control unit, the command value for the motor provided in the joint is calculated.
A robot control method characterized in that a switching unit switches between valid or invalid calculation of a command value for force control by the main control unit and calculation of a command value for torque control by the joint control unit.

Images