JP7151694B2 - combine - Google Patents

Description

The present invention relates to a combine harvester for cutting and harvesting rice and the like.

Conventionally, voice recognition technology is used in operating a combine harvester. For example, it is used to instruct the operation of an auger that conveys grains to a truck outside the machine (see, for example, Patent Document 1).

Patent No. 2647063

Speech recognition of such a combine requires speech recognition in the presence of various noises. For example, engine sound, threshing sound, running sound, and the like.

Therefore, speech recognition is difficult, and more accuracy than ever before is required.

On the other hand, there is also the problem that such noise is deteriorating the worker's environment.

SUMMARY OF THE INVENTION An object of the present invention is to provide a combine harvester capable of more accurately recognizing speech in consideration of such problems of the conventional combine harvester.

Another object of the present invention is to provide a combine harvester that can improve the environment for workers.

A first aspect of the present invention is
a sound input unit installed at a predetermined location inside the cabin;
A detection unit that extracts voice from the sound input from the sound input unit and recognizes voice using a voice recognition database in an external cloud unit;
a command database unit that stores pre-registered command words and their corresponding motion command sequences;
The command word voice-recognized by the detection unit is collated with the command database unit, and an action command sequence corresponding to the command word is executed.
a control unit;
a display unit that displays the instruction words;
a recording unit for recording engine noise frequency and volume data in the electrical controller;
an articulator that adjusts the frequency and volume of the canceling sound according to the recorded frequency and volume values;
and a sound generator that generates the canceling sound,
The sound generator is installed at four corners near the ceiling of the cabin,
The sound input unit is a combine harvester installed at a position where a plurality of sound waves having opposite phases output from the sound generating units installed at the four corners do not attenuate due to interference.
A second aspect of the present invention is
The frequency and volume data of the engine noise in the recorder is a first combine that can be updated during the maintenance mode of the combine.
A first invention related to the present invention is
a sound input unit installed at a predetermined location inside the cabin;
A detection unit that extracts voice from the sound input from the sound input unit and recognizes voice using a voice recognition database in an external cloud unit;
a command database unit that stores pre-registered command words and their corresponding motion command sequences;
a control unit that collates the command word voice-recognized by the detection unit with the command database, and executes an operation command sequence corresponding to the command word;
and a display section for displaying the instruction word.

A second invention related to the present invention is
The predetermined place where the sound input unit is installed is a front upper position inside the cabin, a front operation unit inside the cabin, or an operator's head . It is a combine of the related 1st invention .

A third invention related to the present invention is
The detection unit also has a function of extracting only engine noise in addition to the voice from the input sound,
The combine harvester according to the second aspect of the present invention is provided with a sound wave output section that outputs a sound wave having a phase opposite to that of the engine noise from a predetermined position inside the cabin.

A fourth invention related to the present invention is
The sound wave output unit is installed at a plurality of locations,
According to a third aspect of the present invention related to the present invention, the input unit is installed at a position where a plurality of opposite-phase sound waves output from the sound wave output units installed at the plurality of locations are not attenuated by interference. It's a combine.

A fifth invention related to the present invention is
In a combine with a driver's seat in the cabin,
a recording unit in the electrical controller that records the frequency and volume of engine noise;
an articulator that adjusts the frequency and volume of the canceling sound according to the recorded frequency and volume values;
A combine harvester comprising: a sound generator for generating the canceling sound.

A sixth invention related to the present invention is
The sound generating unit is installed above the driver's seat of the cabin, and generates the canceling sound from there, or transmits a signal of the canceling sound to the headphones on the head of the operator, and from there the canceling sound. It is a combine of a fifth invention related to the present invention, which generates

A seventh invention related to the present invention is
The combine according to a fifth aspect of the present invention is related to the present invention, wherein the engine noise frequency and volume data in the recording unit can be updated during a maintenance mode of the combine.

An eighth invention related to the present invention is
In a combine with a driver's seat in the cabin,
A sound collecting unit installed in a predetermined place that collects the sound inside the cabin,
a sound articulator that adjusts the frequency and volume of the canceling sound according to the frequency and volume of the engine noise from the sound signal transmitted from the sound collector;
A combine harvester comprising: a sound generator for generating the canceling sound.

A ninth invention related to the present invention is
The predetermined place where the sound collecting part is installed is the combine harvester according to the eighth invention related to the present invention, which is installed inside the cabin or inside the engine room.

A tenth invention related to the present invention is
The sound generating unit is installed above the driver's seat of the cabin and generates the canceling sound therefrom, or transmits a canceling sound signal to an operator's headphones to generate the canceling sound therefrom. It is the combine of 8th invention relevant to this invention .

An eleventh invention related to the present invention is
An eighth aspect of the combine harvester related to the present invention , wherein the sound collector or the articulator detects an abnormality in the frequency of collected engine noise and displays the abnormality on a monitor in the driver's seat of the cabin.

A twelfth invention related to the present invention is
According to the present invention, the sound collecting unit records the transition of the frequency of the collected sound, extracts regular behavior during the key-on cycle by deep learning, and judges behavior deviating from the regular behavior as abnormal. It is the combine of the 11th invention which carries out.

According to the first aspect of the present invention, it is possible to give instructions by voice recognition, and it is possible to prevent the machine from being dirty or the precision machine from being soiled due to machine operation in a field with a lot of dirt such as mud. In addition, the sound environment for workers can be improved. Furthermore, there is no worry that the canceling sound interferes and attenuates at the sound input section.
According to the second aspect of the present invention, it is possible to cope with aging deterioration of the engine.
According to the first invention related to the present invention, it is possible to give instructions by voice recognition, and it is possible to prevent the machine from being dirty or the precision machine from being soiled due to machine operation in a field with a lot of dirt such as mud.

Furthermore, according to the first invention, when performing the speech recognition, the speech recognition database of the external cloud part can be used, and there is no need to prepare one combine harvester, one by one, the cost can be reduced and the accuracy is improved. .

According to the second invention related to the present invention, it is easy to pick up the utterance of the operator.

According to the third invention related to the present invention, engine noise is not included in the utterance of the operator. In other words, there are various sounds inside the cabin, such as engine noise and threshing noise, and the threshing sound must be heard in order to quickly identify any abnormalities in the threshing section. On the other hand, the engine noise is loud and annoying even in normal operation, and is a sound that is continuously generated and mixed with the operator's utterance to impair speech recognition. This engine noise can be reduced to improve the accuracy of speech recognition of the worker's utterance.

According to the fourth invention related to the present invention, there is no concern that the canceling sound interferes and is attenuated in the sound input section.

According to the fifth invention related to the present invention, the sound environment for the operator can be improved.

According to the sixth invention related to the present invention , the canceling sound can be generated better.

According to the seventh invention related to the present invention, it is possible to cope with aged deterioration of the engine.

According to the eighth invention related to the present invention, canceling sound can be generated from collected sound.

According to the ninth invention related to the present invention, sound can be collected at a place where sound collection is easy.

According to the tenth invention related to the present invention , the canceling sound can be generated from a better place.

According to the eleventh invention related to the present invention, it is possible to detect when there is an abnormality in the engine.

According to the twelfth invention related to the present invention , engine abnormality can be detected better.

BRIEF DESCRIPTION OF THE DRAWINGS The side view of the combine concerning embodiment in this invention Schematic perspective view of the same combine Table showing the contents of the command database of the combine The perspective view of another Example of the same combine The perspective view of another Example of the same combine The perspective view of another Example of the same combine The perspective view of another Example of the same combine The perspective view of another Example of the same combine (A), (B) Schematic plan view and side view for explaining the sound cancellation state of the same combine Schematic perspective view of a combine according to another embodiment of the present invention Block diagram of data processing of the combine The perspective view of another Example of the same combine Block diagram of data processing of the combine The perspective view of another Example of the same combine The perspective view of another Example of the same combine (A), (B), (C) Engine abnormality detection graph of the same combine Block diagram showing how the combine harvester detects anomalies The perspective view of another Example of the same combine The perspective view of another Example of the same combine

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. FIG. 1 is a left side view of a combine according to an embodiment of the present invention.

A combine 1 of the present embodiment will be described. This combine 1 has a control section 2 , a reaping section 3 , a threshing section 4 , a sorting section 5 , a storage section 6 , a waste straw processing section 7 , a driving section 8 and a running section 9 . The control unit 2 includes a cabin 20, a driver's seat 21, and the like.

The harvesting unit 3 cuts the culms, the threshing unit 4 threshes the culms, the sorting unit 5 sorts the grains, and the grains are stored in the storage unit 6 . Moreover, the waste straw processing part 7 processes the waste straw after threshing.

The reaping unit 3 has a divider, a lifting device and a cutting device. The divider guides the culms of the field to the raising device, and the raising device is the device that serves to raise the culms guided by the divider. The cutting device cuts the culms raised by the raising device, and conveys the cut culms to the threshing section 4 by a conveying device (not shown).

The threshing section 4 is a device having a feed chain and a threshing cylinder. The threshing cylinders thresh the culms being conveyed by the feed chain.

The sorting unit 5 includes a rocking sorting unit, an air sorting unit, a grain conveying unit, and a straw waste discharging unit.

The reservoir 6 includes a grain tank and a discharge auger. The grain tank stores grains conveyed from the sorting unit 5 . A discharge auger is a device that can discharge grains stored in a grain tank to an arbitrary location.

The waste straw processing unit 7 is provided behind the threshing unit 4 and includes a waste straw conveying device and a waste straw cutting device.

The traveling unit 9 includes a pair of left and right crawler type traveling devices provided below the main frame, and a transmission.

The drive unit 8 is a portion that supplies power to the traveling unit 2, the reaping unit 3, the threshing unit 4, the sorting unit 5, the storage unit 6, and the waste straw processing unit 7 described above.

FIG. 2 is a schematic perspective view of a combine according to an embodiment of the invention. Here, 10 is an operator, 11 is a sound input unit, 12 is a detection unit, 13 is an external cloud unit, 14 is CNA communication, 15 is a control unit, 16 is a command database unit, and 17 is a display unit.

When the worker 10 utters, for example, "go forward", the sound input unit 11 such as a microphone receives the sound wave. A sound signal input to the sound input unit 11 is extracted as an audio signal by the detection unit 12 . Further, the extracted speech signal is subjected to speech recognition using a speech recognition database provided in the external cloud unit 13 . For example, it is voice-recognized as instruction words such as "forward", "backward", and "stop".

The detection unit 12 sends the voice recognition result to the control unit 15 in the cabin 20 via CAN communication 14 . At the same time, the result is also displayed on the display unit 17 in front of the worker 10 so that the worker 10 can confirm it.

The control unit 15 incorporates a command database unit 16 . As shown in FIG. 3, this command database unit 16 stores instruction words such as "forward" and "backward", and "HST opening XX value transmission" and "HST opening XX" corresponding to these instruction words. Operation command sequences such as "value transmission" are stored in a database.

Therefore, while referring to the database unit 16, the control unit 15 collates the instruction words that have been sent to identify the corresponding operation command sequence. ◯ value transmission”, and transmits a control signal to the HST (not shown) so as to set the opening to the ◯◯ value. As a result, the opening of the HST becomes ◯◯, and the combine 1 moves forward.

Further, in the embodiment of FIG. 2, the sound input section 11 is installed in the front upper part inside the cabin 20 . It is set at a position where it is easy to pick up the voice of the worker 10. - 特許庁

FIG. 4 shows another embodiment, in which the sound input section 11 and the detection section 12 are provided in the operation section 18 in front of the cabin 20 . Other structures are the same as in the example of FIG. Since it is provided in the front operation part 18, the operator 10 can speak easily.

FIG. 5 shows still another embodiment, in which the sound input unit 11 is provided in the microphone of the headset of the worker 10 . Other structures are the same as in the example of FIG. Since it is the microphone of the headset of the worker 10, it is easy to pick up the voice of the worker 10. - 特許庁

FIG. 6 shows still another embodiment, in which the sound input unit 11 and the detection unit 12 are installed above the front inside the cabin 20 . It is set at a position where it is easy to pick up the voice of the worker 10. - 特許庁The detection unit 12 also has a function of extracting only the engine noise from the sound input from the sound input unit 11 in addition to the voice signal. A sound wave output unit 19 is further provided at a predetermined position inside the cabin 20 to output a sound wave having a phase opposite to that of the engine noise. Its function will be described later.

FIG. 7 shows still another embodiment, which differs from the embodiment in FIG. 6 only in the positions of the sound input section 11 and the detection section 12. In FIG. That is, the sound input section 11 and the detection section 12 are provided in the operation section 18 in front of the cabin 20 . Since it is provided in the front operation part 18, the operator 10 can speak easily. This detection unit 12 also has a function of extracting only engine noise in addition to the voice signal from the sound input from the sound input unit 11 . A sound wave output unit 19 is provided at a predetermined position inside the cabin 20 to output a sound wave having a phase opposite to that of the engine noise. Its function will be described later.

FIG. 8 shows still another embodiment, which differs from the embodiment in FIG. 6 only in the position of the sound input section 11. In FIG. That is, the sound input unit 11 is provided in the microphone of the headset of the worker 10 . Since it is the microphone of the headset of the worker 10, it is easy to pick up the voice of the worker 10. - 特許庁This detection unit 12 also has a function of extracting only engine noise from the sound input from the sound input unit 11 in addition to the voice signal. A sound wave output unit 19 is provided at a predetermined position inside the cabin 20 to output a sound wave having a phase opposite to that of the engine noise.

Next, the engine noise extracting function of the detection unit 12 and the operation of the sound wave output unit 19 will be described. 6, 7, and 8, the detector 12 can extract only the engine noise in addition to the voice signal, and can create a sound wave signal of the opposite phase. Then, the opposite-phase sound wave signals are transmitted to the sound wave output units 19 at the four corners, and the opposite-phase sound wave signals are output therefrom. The opposite phase sound wave signal may be generated by the sound wave output unit 19 instead of the detection unit 12 .

9A is a schematic plan view of the cabin 20, and FIG. 9B is a schematic front view of the cabin 20. FIG. In this case, the sound input unit 11 is arranged in the center of the lower front part of the cabin 20 .

As shown in FIGS. 9A and 9B, in this embodiment, sound wave output units 19 are installed at four corners near the ceiling of the cabin 20 . Although peaks and troughs of the sound waves are generated depending on the transmission state of the sound waves, the anti-phase sound waves are attenuated at the position where the peaks and troughs overlap and interfere.

Therefore, it is necessary to install the sound input unit 11 at a position where anti-phase sound waves can reliably reach so that the engine noise can be effectively canceled. For example, it is possible to generate sounds of opposite phases from the four corners inside the cabin 20, and to install the sound input unit 11 as shown in FIG. desirable.

Next, an embodiment of the present invention for the purpose of improving the sound environment for workers will be described. FIG. 10 shows an embodiment thereof, wherein 30 is an electrical controller, 31 is a recording section built therein, 32 is an articulation section, and 33 is a sounding section. 11 is a block diagram for explaining the operation of the device of FIG. 10. FIG.

10 and 11, the engine frequency and sound volume data obtained in advance are transmitted from the personal computer 34 to the recording unit 31 via the serial bus. The recording unit 31 is an E ² ROM or the like. The data is transmitted from the recording unit 31 to the articulation unit 32 via a serial bus or CAN. The articulator 32 is also an E ² ROM or the like. The sound adjuster 32 adjusts the frequency and volume of the sound signal for canceling the engine sound in accordance with the obtained engine frequency and volume values. The sound generator 33 obtains the canceling sound data and generates it in the cabin 20 . Note that the sound generator 33 is installed above the driver's seat 21 in this embodiment. As a result, the worker can work in an environment in which the engine noise is cancelled.

FIG. 12 shows another embodiment, differing in that the sound generator 33 is headphones 33a set on the head of the operator 10 instead of the driver's seat 21. FIG. It is transmitted wirelessly from the articulator 32 . FIG. 13 is a block diagram showing its operation.

10 to 13, the frequency and volume data of the engine noise in the recording unit 31 and the sound adjusting unit 32 can be updated from the machine controller 35, for example, during the maintenance mode of the combine harvester. As a result, it is possible to cope with the change in the engine sound due to the secular change of the engine. Furthermore, fine adjustment is also possible according to the operator's sense.

Next, an embodiment will be described in which noise is reduced by collecting sound rather than recording the frequency and volume of the engine sound in advance.

FIG. 14 is a diagram showing an embodiment thereof, wherein 40 is a sound collecting section installed above the interior of the cabin 20. As shown in FIG. The sound is collected by the sound collector 40 and the signal is transmitted to the articulator 32 . Then, the articulator 32 extracts only the engine noise from the collected sound, creates canceling sound data corresponding thereto, and sends it to the sound generator 33 . The sound generator 33 is installed above the driver's seat 21 as shown in the figure. Alternatively, the sound collector 40 may extract engine noise and send engine noise data to the articulator 32 .

FIG. 15 shows another embodiment, differing in that the sound generator 33 is installed on the headphone 33a on the head of the operator.

16 and 17 are diagrams showing an example of extracting engine noise in the sound collector 40, detecting an abnormality and giving a warning. FIG. 16(A) records changes in normal frequency and sound volume from engine key-on to engine stop. On the other hand, FIG. 16B is a graph when an abnormality occurs. From this phenomenon, as shown in FIG. 16C, when an abnormality is detected, an abnormality signal is sent to the monitor to issue a warning.

FIG. 17 shows an example of determining such an anomaly. Waveforms of various frequencies and volumes in the normal state are collected and stored in the storage unit 41a. A normal region is calculated from among them by a method such as deep learning in the calculation unit 41b. By sending the normal region data to the sound collecting unit 40, the sound collecting unit 40 can determine the abnormality of the engine noise in real time.
Abnormalities can also be detected in the same way when engine noise is extracted by the articulator 32 .

FIG. 18 shows another embodiment, in which the sound collector 40 is installed in an engine room 42 in which an engine is installed. Further, the sound generator 33 is provided in the driver's seat 21 . By providing the sound collector 40 in the engine room 42, the engine noise can be obtained more accurately without being mixed with other noise.

FIG. 19 is a different embodiment from FIG. 18, differing in that the sound generator 33 is placed on the headphone 33a on the head of the operator.

It should be noted that the sound collecting section and the articulation section of the present invention may be realized by means of hardware dedicated to their functions, or may be partly realized by software using a general-purpose computer.

INDUSTRIAL APPLICABILITY The present invention can realize a combine harvester capable of more accurately recognizing voices, or can realize a combine harvester capable of improving the worker's environment, and is therefore useful for harvesting rice.

1 combine harvester 2 control unit 20 cabin 21 driver's seat 3 reaping unit 4 threshing unit 5 sorting unit 6 storage unit 7 waste straw processing unit 8 drive unit 9 traveling unit 10 operator 11 sound input unit 12 detection unit 13 external clout unit 15 control unit 16 command database unit 17 display unit 18 operation unit 19 sound wave output unit 30 electrical controller 31 recording unit 32 articulation unit 33 sound generation unit 40 sound collection unit 42 engine room

Claims (2)

a sound input unit installed at a predetermined location inside the cabin;
A detection unit that extracts voice from the sound input from the sound input unit and recognizes voice using a voice recognition database in an external cloud unit;
a command database unit that stores pre-registered command words and their corresponding motion command sequences;
a control unit that collates the command word voice-recognized by the detection unit with the command database unit , and executes an operation command sequence corresponding to the command word;
a display unit that displays the instruction words;
a recording unit for recording engine noise frequency and volume data in the electrical controller;
an articulator that adjusts the frequency and volume of the canceling sound according to the recorded frequency and volume values;
and a sound generator that generates the canceling sound,
The sound generator is installed at four corners near the ceiling of the cabin,
In the combine, the sound input unit is installed at a position where a plurality of sound waves having opposite phases output from the sound generating units installed at the four corners do not attenuate due to interference . 2. The combine of claim 1 , wherein the engine noise frequency and volume data in the recorder are updateable during a service mode of the combine.

Images