JPS63295394A - Speech recognizer in crane controller - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to a crane control device that controls a crane using a voice recognition device, and particularly relates to the voice recognition device.

(Prior art) Generally, in a crane control device that controls a crane using a voice recognition device, the voice commands have various functions, and the voice commands are executed only when they are reliably recognized depending on the type of voice command. It includes things that are necessary and instructions that must be executed immediately even if there is some doubt as to whether recognition has been performed correctly. For example, in the case of the voice command word "stop" that instructs a crane to stop while it is moving, the difference score value between the standard pattern and the human voice pattern is large, so there is a possibility that the human voice is not "stop". Even if something happens, it is necessary to consider it as a "stop" and immediately execute the crane stop operation. This is because if the stopping operation is not performed when it is necessary to stop, the crane may collide with other objects and cause damage, the crane itself may break, or people may be injured. On the other hand, in the case of the voice command word ``Rifmag release'', which instructs the work of releasing the iron load from the electromagnet and lowering it when moving it by attracting it with the electromagnet of the crane, the score value is small, and therefore the input voice Only when it is certain that this is a ``Rifmag release'', it is necessary to treat it as a ``Rifmag release'' and perform the work to deenergize the crane's electromagnet and release the load. This is because if the electromagnet is opened by mistake when the load is in a position where it should not be unloaded, there is a risk that a serious accident such as the load falling may occur.

In the past, human speech patterns were compared point by point with a group of standard patterns, and when the difference between the two was the smallest and the difference was below a certain threshold, the standard pattern was used as the recognition command word. There is a problem in that some types of voice commands require more gradual or more reliable recognition, but this cannot be achieved.

(Problems to be Solved by the Invention) The present invention aims to solve the problems of the above-mentioned conventional techniques.

That is, an object of the present invention is to change the recognition standard according to the type of command in a voice recognition device for a crane device that issues crane control commands by voice, thereby enabling safer operation. It is.

(Means for Solving the Problems) In order to achieve the above object, the present invention provides a voice recognition device for a crane control device that identifies a voice command and drives a crane device in response to the recognition result. , a voice command recognition means that compares an input voice command pattern and a standard pattern, and outputs a command word corresponding to the smallest score value as a recognized command word as a code, and also outputs the score value; recognition command word classification means for classifying recognition command words outputted by the voice command recognition means; and score values of the recognition command words classified by the recognition command word classification means are set for each corresponding command word. The invention is characterized by comprising a means for comparing the command with a threshold value and determining whether or not to execute the action of the corresponding command.

(Function) The present invention sets a small threshold for command identification (classification) for commands of the type that will cause serious consequences if executed while being misrecognized, such as "Rifmag release". It performs more accurate recognition, and sets a large threshold to prevent commands from being overlooked for commands such as "stop" that would have serious consequences if they were overlooked and neglected. Make sure there is no such thing. For this purpose, a threshold value is set in advance as a reference for determining whether or not to execute a command depending on the type of command. The type of recognition command word is then identified, and the corresponding score value is compared with a set threshold value specific to that word.

Since the command is executed flexibly depending on the type of command, safe operation of the lane device is possible.

(Embodiment) FIG. 1 is a block diagram showing an outline of a crane control system in which the voice recognition device of the present invention is used, and includes a recognition device control panel 1, a crane control panel 2, a crane 3, and a wireless device mobile station. 4, headset microphone 5, and wireless modem 6
and console 7. The recognition device control panel 1 includes a wireless device fixed station 11, a voice recognition device 12, a voice response device 13, an interface relay board 14, and a wireless modem 15.

The wireless device fixed station 11 is a device that receives a signal sent from the wireless device mobile station 4, demodulates it into a voice command signal, or modulates a voice signal from the voice response device 13, and transmits the signal.

The voice recognition device 12 works together with the voice response device 13 to interpret analog voice information from the wireless device fixed station 11 and compare it with pre-registered voice command information (standard pattern) to determine the degree of agreement. identifies the content of the voice command,
It outputs an operation command word for controlling the corresponding crane.

The voice response device 13 generates a voice signal in response to the voice recognition result and sends this voice signal to the wireless device fixed station 11 and the wireless device in order to confirm whether the voice recognition has been performed correctly. This is communicated to the driver via the mobile station 4.

The interface relay board 14 includes the voice recognition device 12
It converts the output from the relay to drive the relay.

The wireless modem lδ performs transmission and reception with the wireless modem 6 connected to the console 7 on the ground, demodulates the received command signal from the console 70, and inputs the command signal to the voice recognition device 12.
It is used to transmit the signal to all the decoding units or to modulate the signal from the voice recognition device 12 to transmit it wirelessly to the console 7.

The crane control panel 2 is an interface relay panel 14
relay output and limit switch 33 of crane 3
It has a contactor that is controlled by detection signals etc. from the crane, drives the motor 31, brake 32, etc. for the crane 3, and also serves as a power source for alarms, illumination lights, etc. (not shown).

The wireless device mobile station 4 is carried by the driver and modulates a voice command from a headset microphone 5 and wirelessly transmits it to the recognition device control panel 1. This is a wireless device that receives the audio output from the response device 13 and transmits it to the driver through the earphone of the headset.

The console 7 and wireless modem 6.15 are installed on the ground in order to enable console input in addition to voice input from the headset microphone 5 and wireless device mobile station 4, and are the first part of the entire system. Used for startup, final shutdown, and other operations.

Next, the operation of the crane system configured as described above will be explained.

The voice command received by the wireless device fixed station 11 is recognized by the voice recognition device 12, and in response to the recognized command, various control signals necessary for controlling the operation of the crane are output. In this embodiment, this function is realized by the operation of a crane control program executed by a microcomputer for the voice recognition device.

FIG. 2 shows an outline of the flow of the operation.

After starting the program, as initial value settings for the program, the inching constant CO and steady rest constant Do are manually set from the console 7 so that the crane device to be used responds appropriately to voice commands.

In addition, the threshold value N1 to N +'% for each command that determines whether or not to execute the recognized command, which is a feature of the present invention.
1 and the maximum allowable threshold value N0 are manually determined.

Next, it starts the voice recognition program and waits for voice commands from the driver.

First, when the operator gives a voice command to turn on the power to the crane, the voice recognition device 12 recognizes the command and sends a corresponding command signal to the interface relay board 14 to operate the power relay and turn on the crane. The control panel 2 and the crane 3 are powered on, the crane is put into an operable state, and it waits for the next command from the operator.

When the driver issues a command by voice, it is compared point by point with each standard pattern stored in the memory of the voice recognition device, the difference or score value is calculated, and the corresponding standard pattern with the smallest score value is selected. It is determined that the recognition command word
Compare with o.

When the score value SCO is larger than the threshold value N0, the recognition command word is rejected and the input of the next voice command is awaited. When the score value SCO is smaller than the threshold value N0,
Recognition command word, response buffer RE, 5BUF (
(not shown).

Next, it is determined what the command (recognition command word) stored in the response buffer RESBUF is. That is, the recognized command word and the standard command word are sequentially compared and a match is found. Next, in accordance with the determination result, the process moves to a step of determining whether or not the score value SCO, which is a feature of the present invention, is smaller than the threshold value N8. Recognized voice commands include those such as "power on", "power off", "alarm", "lighting person", "lighting off", etc. that can immediately and unconditionally execute the operation of the crane equipment. After the score value is determined, the operation subroutine is immediately executed. That is, when the content of the response buffer RESBUF is "power supply person", it is as described above, and when the content is "power off", it is SC.
When O<N2 is determined and the determination is VH2,
A subroutine is executed that turns all outputs of the voice recognition device to zero, turns off the power to the crane device, and returns to the beginning.

When "Alarm", "Lighting person", and "Lighting off" are set,
SCO<Na, SCO<Na, SCO<Na is determined, and when the determination is '/E9, the alarm, subroutine, lighting person subroutine, and lighting off subroutine are executed, respectively, and alarms, power supply issues, etc. It outputs a control signal and returns to the state of waiting for the next voice command.

On the other hand, "East direction", "West direction", "South direction", "North direction", "Upward winding up", "Downward direction (downward winding)"
Start, stop, etc. commands that instruct the direction of movement, such as start, stop, etc.
In the case of a command that instructs crane operation in combination with commands such as inching and steady rest, by going through a common SCO<Nn determination step, when the score value SCO is smaller than the threshold value Nn, The command indicating the direction of movement is stored in the -H movement direction instruction buffer BUF, and the state returns to the state of waiting for the next voice command.

Similarly, in the case of execution instruction commands such as "start", "stop", "jog", and "steady rest" used in combination with the motion direction instruction command, each SC
After passing through the O<N determination step, the start subroutine, crane stop subroutine, and inching operation subroutine are processed, respectively. The start subroutine is a routine that generates an output for causing the crane to start an operation according to the contents of the operation direction instruction buffer yBUF, and returns to a state of waiting for the next voice command.

The crane stop subroutine uses operation direction instruction buffer B.
This stops the crane operating according to the contents of UP and clears the contents of the work direction instruction buffer BUF. The inching operation subroutine is
It performs inching or steady rest motion.

For example, if the recognition command word is "stop", its score value is compared with threshold value N8.

As a result of the comparison, if the value is smaller than the threshold value, the process immediately moves to a subroutine that executes that operation, ie, "stops".

When the score value SCO is greater than the threshold value N, the command is not executed and the next voice command is awaited.

Further, for example, the recognition command word and the standard command word are sequentially compared, and if they match at "r riff mag release", the score value is compared with the corresponding threshold value Ni. As a result of the comparison, if it is smaller than the threshold value, the process moves to a subroutine that executes that operation, ie, ``riffmag release''. When the score value SCO is larger than the threshold value N1, the command is not executed and the input of the next voice command is awaited.

In this way, regarding the recognition command word recognized by the recognition unit,
First, the type is identified (determined), and compared with a threshold value set corresponding to the identified command word to determine whether or not to execute the action of the command. For commands that are misrecognized, such as "Release Rifmag," and which would have serious consequences if executed, the threshold value is set to a small value for more accurate recognition, and the command, such as "Stop," is For commands that would have serious consequences if overlooked and neglected, the threshold is set large to prevent the command from being overlooked, making it possible to operate the lane device safely. Become.

Note that in this embodiment, the score value and the threshold value are compared and determined for all command words, but the comparison and determination may be performed only for important specific words.

(Effects of the Invention) As described in detail above, the present invention is capable of controlling a crane using voice recognition, and is effective against commands of the type that would lead to serious consequences if executed incorrectly, such as "Rifmag Release". The threshold value is set to a small value for more accurate recognition, and for 0 types of commands such as "stop" which will lead to serious consequences if the command is overlooked and neglected, it will be used. Since the threshold value is set large to prevent commands from being missed, the lane device can be operated safely.

[Brief explanation of the drawing]

FIG. 1 is a block diagram schematically showing a device for controlling a crane using a voice recognition device according to an embodiment of the present invention. (a) to (j) of FIG. 2 are operational flowcharts for explaining the operation of the apparatus of the embodiment of FIG. 1. 1... Recognition device control panel, 2... Crane control panel,
3... Crane, 4... Wireless device mobile station, 5...
Headset microphone, 6... Wireless modem, 7... Console, 11... Wireless device fixed station, 12... Voice recognition device, 13... Voice response device, 14... Interface relay board , 15... Wireless modem. Patent Applicant Nippon Hoist Machinery Co., Ltd. Nippon Steel Corporation Figure 1 Figure 2 Figure 2

Claims (1)

[Claims] In a voice recognition device for a crane control device that identifies a voice command and drives a crane device in response to the recognition result, the voice recognition device compares an input voice command pattern with a standard pattern, and determines the difference between the input voice command pattern and a standard pattern. A voice command recognition means for outputting a code as a recognition command word corresponding to the command word with the smallest score value, and a voice command recognition means for outputting the score value; and a recognition command for classifying the recognition command word outputted by the voice command recognition means. A word classification means, and a score value of the recognition command word classified by the recognition command word classification means is compared with a threshold value set for each corresponding command word, and the operation of the corresponding command is executed. 1. A voice recognition device for a crane control device, comprising: means for determining whether or not.