JPS6293706A - Robot control method - Google Patents

Abstract

PURPOSE:To execute safely a command control without providing a complicated control box by confirming a fact that there is not wrong recognition, when a voice command is received and a voice is recognized, and thereafter, controlling actually a robot. CONSTITUTION:A voice command is brought to a spectral analysis by a voice analysis module 21 and the voice command is compared with a standard pattern by a voice comparison module 25, and a voice is recognized. When the voice has been recognized, a CPU 19 stores the voice command whose voice has been recognized, in a RAM 17, and when the storage is ended, it is awaited that an operation button 7 becomes on. When the operation button 7 becomes on, 'YES' is selected and the voice command which has been stored in the RAM 17 is read out, and also the voice command which has been read out is outputted to a robot driving control device 11 through an output use interface 23. At the same time, only when the operation button 7 is on, an operation command signal for an actual operation of a robot is outputted to the robot driving control device 11, and the robot is operated a little. In this state, an operator sees its operation and decides whether it coincides with the command contents which have been given by the voice command.

Description

[Detailed description of the invention] [Industrial application field] The present invention relates to a robot control method.

[Prior Art] Conventionally, as a method for commanding and controlling a nursing care robot, for example, a method is known in which a control box having a large number of operation buttons is connected to the robot and the operation buttons are operated as appropriate.

However, this method has the problem that when the number of drive axes of the robot increases and the number of operation buttons increases, the size of the control box itself increases, making it difficult to handle and complicated to operate. In particular, when other nursing tasks must be performed while operating the operation buttons, the user must pay attention to many movements of the robot at once, which is not only more troublesome, but also may cause the button to be pressed incorrectly. There was a problem that it could be stored away.

A known method for commanding and controlling robots that does not cause such problems is to use voice commands.

However, the latter method has the problem of being uneasy and difficult to put into practical use because of noise-induced malfunctions and a drop in recognition rate.

[Purpose of the Invention] The purpose of the present invention is to solve the above-mentioned problems of the conventional technology, and to create a robot that has many axes and performs complicated movements without installing a complicated control box. ,
The object of the present invention is to provide a method for controlling a robot that can be safely commanded and controlled.

[Summary of the Invention] To achieve the above object, the present invention adopts the voice command/control method, and upon receiving a voice command and recognizing the voice, confirms erroneous recognition occurring in the voice recognition, After confirming that there is no erroneous recognition, the robot is actually controlled using the recognized voice command, and if there is any erroneous recognition, the voice command is invalidated.

[Embodiment] An embodiment of the present invention will be described below with reference to FIGS. 1 to 4. FIG. 1 is a perspective view of a robot used to implement this embodiment, FIG. 2 is an explanatory diagram of the control circuit of the robot, FIG. 3 is a flowchart showing the processing of this embodiment, and FIG. 5 is a time chart when executing the embodiment.

Referring to FIG. 1, a robot 1 used to carry out this embodiment is provided with a microphone 3, a voice enable button 5, and an operation button 7 as command input means. The microphone 3 is used to input appropriate voice commands. The audio enable button 5 normally disables the audio input from the microphone 3, and only enables it when pressed. The operation button 7 is used to actually operate the robot 1 based on the voice command after it has been input.

Next, the robots 1 to 1 are controlled based on inputs from the microphone 3, the audio enable button 5, and the operation button 7.
The control circuit will be explained with reference to FIG.

This control circuit consists of a noodle voice command unit 9 and a robot drive control device 11. The voice command unit 9 comprises a voice input interface 13, a digital input 15, a RAM 17, a CPU 19, a voice analysis module 21, an output interface 23, a voice comparison module 25, and a standard pattern module 27. There is.

Input signals from the microphone 3, the audio enable button 11 and the operation button 13 are input to the audio input interface 13 and digital input (DI) 15 of the audio command unit 9, respectively.

The voice command inputted from the voice input interface 13 is subjected to spectrum analysis, etc. in the voice analysis module 21, and then is associated with a standard pattern stored in the standard pattern module 27 in the voice comparison module 25, and is voice recognized. Ru. The RAM 17 temporarily stores the voice commands that have been recognized.

Note that the audio enable button 5 is activated by the CPU 11 in this case.
The input voice command is validated or invalidated through the operation.

Further, the operation button 7 is connected to the RAM via the CPUII.
When the voice command stored in the robot controller 17 is read out, the robot drive and control device 11 is actually operated according to the read voice command.

The CPU II controls each module, etc., and reads and transmits data. For example, when a voice command is read from the RAM 17 based on a command from the operation button 7, the read voice command is used for output. It is output to the robot drive control device 11 via the interface 23.

The robot drive control device 11 actually drives the servo motor M based on the voice command from the microphone 3 and the operation example from the operation command button 7, and controls the robot 1 as appropriate.

Next, an embodiment of the present invention using the robot will be described with reference to FIGS. 3 and 4.

As shown in FIG. 3, when voice command control is started in step 301, first, in step 302, a voice command is input from the microphone 3. For example, Fig. 4(a)
When voice commands like Nohe, B, and C are input, r\E
Select SJ and proceed to step 303.

In step 303, it is determined whether or not the voice enable button 5 is turned on, and the voice command A is executed.

If the voice enable button 5 is off, as in the case where C is input, the voice command is invalidated and the process returns to step 302 to receive the next voice command.

If the voice enable button 5 is turned on (D in FIG. 4(b)) as in the case where voice command B is input, the voice command is validated and the process proceeds to step 304.

In step 304, the voice command is subjected to spectrum analysis by the voice analysis module 21.

Step 305 compares the spectrally analyzed voice command with a standard pattern in the voice comparison module 25;
Voice recognition.

When voice recognition is performed in step 305, step 30
6, the CPU 19 stores the recognized voice command in the RAM.
17.

When the storage is completed, step 307 waits for the operation button 7 to be turned on.

When the operation button 7 is turned on, for example, as shown in E of FIG. 4(C), fYEsJ is selected in step 307, and the process proceeds to step 308.

In step 308, the voice command stored in the RAM 17 is read out, and the read voice command is output to the robot drive control device 11 via the output interface 23. At the same time, in step 308, while the operation button 7 is turned on (E in FIG. 4(d>)
During the period shown in ), the movement command signal for actual robot movement is sent to the robot drive control l! It is output to the device 11. As a result, in step 309, the robot 1 is
> As shown in G, it moves a little.

Therefore, in step 310, the operator observes the robot's motion and determines whether or not the motion matches the contents of the voice command. If it is determined that the previous movement is different from the command content and there is a misrecognition, rYEsJ is selected in step 310, the movement command button 7 is turned off in step 311, and the voice command is canceled. Then, the process returns to step 302 to wait for the next voice command.

On the other hand, if it is determined in step 310 that the motion of the robot shown in G in FIG.

Then, in step 312, as shown at F in FIG. 4(C), press the operation button 7 again and press H in FIG. 4(d).
As shown in FIG. 3, when the robot 1 is fully operated based on the voice command, the voice command control is terminated in step 313.

Therefore, according to this embodiment, since the robot is commanded by voice, there is no need for a complicated control box, etc., and when the robot is operated based on the voice command, the robot is operated once by using the motion command button 7. Since the robot is allowed to fully operate after it is determined that there is no misrecognition, the robot will not malfunction and therefore the robot can be safely commanded and controlled.

Furthermore, since voice input is enabled only when the voice enable switch is turned on, noise and the like are less likely to be input, and there are fewer erroneous recognitions.

Note that in the above embodiment, the robot 1 makes a trial motion based on the recognized voice command, and the operator visually confirms the motion to confirm whether there is any misrecognition, but it is not always necessary to do this. Instead, for example, the recognized voice command can be output from a suitable voice output means, and the operator can confirm the presence or absence of erroneous recognition by listening to it.

[Effects of the Invention] The robot control method of the present invention is configured as described above, so that commands can be given by voice, so there is no need to use a complicated control box. Since actual control is performed, the robot can be commanded and controlled safely without malfunction.

[Brief explanation of drawings]

The drawings relate to embodiments of the present invention, and include Figure 1, Figure 2, and Figure 2. Figure 4 age Figure 3

Claims (3)

[Claims] (1) When a voice command is received and the voice is recognized, the robot is first checked for erroneous recognition in the voice recognition, and after confirming that there is no erroneous recognition, the robot is actually controlled by the voice command. Robot control method. (2) When receiving the voice command, an appropriate time is specified, and only voice commands input within this time are made valid. How to control the robot. (3) To check for erroneous recognition in the voice recognition, make the robot perform a few trial movements based on the voice commands that have been recognized,
2. The method of controlling a robot according to claim 1, wherein the robot control method is performed by visually confirming the motion.