JP2023157271A - Industrial vehicle control device, industrial vehicle control system and program for industrial vehicle control device - Google Patents

Abstract

To provide an industrial vehicle control device, an industrial vehicle control system and a program for the industrial vehicle control device capable of accurately acquiring an estimation index for estimating a work state of an industrial vehicle with a few man-hours.SOLUTION: A work state estimation section 30 performs machine learning with input of operation information related to an operation state of a forklift and input of correct answer data of a work state. The work state estimation section 30 is capable of performing the machine learning on the basis of the input of the operation information on actual work conducted in the past and the correct answer data showing a work state as a result of the operation information. Thus, a control device is capable of acquiring an estimation index for estimating the work state of the forklift through a process to perform the machine learning using the operation information in the past and the correct answer data. Since the estimation index is based on actual data in the past, the control device is capable of accurately estimating the work state.SELECTED DRAWING: Figure 4

Description

The present invention relates to an industrial vehicle control device, an industrial vehicle control system, and a program for the industrial vehicle control device.

2. Description of the Related Art As a control device for an industrial vehicle, for example, one described in Patent Document 1 is known. The control device for an industrial vehicle described in Patent Document 1 estimates the working state of the industrial vehicle based on operation information for the industrial vehicle, and controls the industrial vehicle based on the estimation result of the working state.

JP 2019-189435 specification

In order for the control device to estimate the work state, a work state estimation model is constructed in which estimation indicators such as transition conditions and transition thresholds for state transition to each work state are set. However, designing an estimation index for estimating a working state has the problem that the design work is difficult, it is difficult to improve accuracy, and the number of man-hours for design is enormous.

An object of the present invention is to provide an industrial vehicle control device, an industrial vehicle control system, and a program for an industrial vehicle control device that can accurately acquire estimation indicators for estimating the working state of an industrial vehicle with a small number of man-hours. It is to provide.

A control device for an industrial vehicle according to one aspect of the present invention is a control device for estimating a working state of an industrial vehicle, and includes a working state estimation unit that estimates a working state of the industrial vehicle using machine learning. The working state estimating unit performs machine learning by inputting operation information regarding the operating state of the industrial vehicle and inputting correct answer data regarding the working state.

A control device for an industrial vehicle includes a working state estimation unit that estimates a working state of the industrial vehicle using machine learning. Therefore, the control device can design an estimation index for estimating the working state of the industrial vehicle through machine learning. Here, the working state estimating unit performs machine learning by inputting operation information regarding the operating state of the industrial vehicle and inputting correct answer data regarding the working state. In this way, the work state estimation unit can perform machine learning based on the input of operation information that was actually performed in the past and the correct data that indicates what kind of work state the operation information resulted in. can. Thereby, the control device can acquire an estimation index for estimating the working state of the industrial vehicle through the process of simply performing machine learning using past operation information and correct answer data. Further, since the estimation index is based on past actual data, it is possible to estimate the working state with high accuracy. As described above, an estimation index for estimating the working state of an industrial vehicle can be obtained with high precision with a small number of man-hours.

The work state estimation unit may receive current operation information and output the current work state based on the learning result. Thereby, the work state estimation section can easily estimate the current work state simply by inputting the current operation information.

The control device for the industrial vehicle includes a load state estimating section that outputs a load state related to the state of the load handled by the industrial vehicle based on the work state output from the work state estimating section, a work state outputted by the work state estimating section, and The apparatus may further include a working state correction section that corrects the working state based on the cargo state outputted by the cargo state estimating section. In this case, the control device can improve the accuracy of estimating the working state by performing correction in consideration of the cargo state of the industrial vehicle.

The operation information may include at least one of an accelerator operation amount, a tire angle, a lift operation amount, a reach operation amount, and a tilt operation amount. These parameters are parameters that reflect the intentions of workers in industrial vehicles. The work state estimation unit can perform appropriate machine learning by using such parameters as operation information.

The work state estimation unit may perform machine learning using LSTM as a machine learning model and using operation information acquired for each time series as learning data. When performing machine learning using operation information acquired for each time series, LSTM can perform learning that takes into account not only the most recent data but also data from the past. Therefore, the work state estimation unit can obtain a more accurate estimation index.

An industrial vehicle control system according to one aspect of the present invention includes the above-described industrial vehicle control device.

A program for an industrial vehicle control device according to one aspect of the present invention is used in the above-mentioned industrial vehicle control device.

These industrial vehicle control systems and industrial vehicle control device programs can obtain the same functions and effects as the industrial vehicle control device described above.

According to the present invention, there is provided an industrial vehicle control device, an industrial vehicle control system, and an industrial vehicle control device capable of accurately acquiring a working state estimation model for estimating the working state of an industrial vehicle with low man-hours. programs can be provided.

1 is a block diagram showing a driving support system according to an embodiment of the present invention. (a) is a perspective view showing how the photographing section is attached to the forklift, and (b) is a schematic diagram for explaining the angle of the photographing section. It is a figure for explaining the working state of a forklift. It is a block diagram of a work state acquisition part. FIG. 3 is a diagram illustrating how the work state estimating unit performs machine learning. It is a flowchart which shows the processing content of a driving support system. It is a graph explaining the effect of a luggage state estimation part and a work state correction part. It is a graph showing experimental results of the control device.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the drawings. In addition, in the drawings, the same or equivalent elements are given the same reference numerals, and overlapping explanations will be omitted.

FIG. 1 is a block diagram showing a driving support system 100 (industrial vehicle control system) including a control device 20 according to an embodiment of the present invention. The driving support system 100 is a system for remotely controlling an industrial vehicle. As shown in FIG. 1, the driving support system 100 includes a forklift 1 (industrial vehicle) and a remote control device 2.

The forklift 1 includes an operation control section 11 and a plurality of photographing sections 12. The operation control unit 11 receives a command signal from the remote control device 2, and performs operation control and steering control based on the command signal. A plurality of photographing units 12 are provided at each part of the forklift 1 and photograph the surrounding environment of the forklift 1. The photographing unit 12 acquires a photographed video as support information used for work support, and transmits it to a display control unit 22, which will be described later. An example of the mounting positions of the plurality of imaging units 12 is shown in FIG. 2(a). The photographing section 12 is provided at the front end of the vehicle body of the forklift 1, at the end in the width direction, at the ceiling, or the like. Note that when the XYZ coordinates are set as shown in FIG. 2(b), the imaging unit 12 at each location may be installed in a tilted state around each of the X, Y, and Z axes. In FIG. 2A, the imaging units 12 are provided at eight locations, and each captures images from different locations.

As shown in FIG. 1, the remote control device 2 includes a storage section 15, an operation section 16, a display section 17, an operation information acquisition section 18, and a control device 20. The storage unit 15 is a device that stores various information. The operation unit 16 is a user interface through which an operator inputs operations to remotely control the forklift 1. The display unit 17 is a user interface for outputting video. The display unit 17 has a first area D1, a second area D2, and a third area D3 that can output different images.

The operation information acquisition unit 18 acquires operation information when an operator is operating an operation target (here, the operation unit 16). The operation information acquisition section 18 is configured by, for example, a sensor provided on the operation lever of the operation section 16 or a means for detecting the operation content based on a signal indicating the operation content of the operation section 16.

The control device 20 is a control unit that controls the entire remote control device 2 . The control device 20 includes an ECU (Electronic Control Unit) that collectively manages the remote control device 2 . The ECU is an electronic control unit that includes a CPU [Central Processing Unit], a ROM [Read Only Memory], a RAM [Random Access Memory], a CAN [Controller Area Network], a communication circuit, and the like. The ECU realizes various functions by, for example, loading a program stored in a ROM into a RAM and executing the program loaded into the RAM by a CPU. The control device 20 includes a driving command section 21, a display control section 22, a work state acquisition section 23, and a viewpoint information acquisition section 26.

The driving command section 21 is a unit that generates a command signal based on the operation input through the operation section 16 and transmits it to the driving control section 11 .

The display control section 22 is a unit that controls the display contents of the display section 17. The display control unit 22 causes the display unit 17 to display information for supporting the operator's remote control work. The display control unit 22 controls the display contents of the first area D1, second area D2, and third area D3 of the display unit 17 using information in the database of the storage unit 15.

The display control unit 22 selects a video to be displayed on the display unit 17 based on the viewpoint information acquired by the viewpoint information acquisition unit 26. The first area D1 and the second area D2 are large screen areas that display a specific image in a large size, and the image of the imaging unit 12 selected by the display control unit 22 from among the images of the plurality of imaging units 12 is displayed. Ru. The third area D3 corresponds to a small screen section that displays a plurality of images in a small size, and displays images from all the photographing units 12 on the forklift 1 as environmental information for checking the surroundings.

For example, FIG. 3(a) is a table defining the working status of the forklift 8. Here, examples of work states include "Stop", "Move fwd", "Approach to shelf", "Adjust heading", "Load/Unload", Regarding the seven states related to the movement of the forklift 3 and the removal/unloading work of the forklift 3, such as "Retreat" and "Move reverse", the state with cargo (Load) and the state without cargo (Load) are determined. (NoLoad), a total of 14 work states are defined. The driving of the forklift 3 corresponds to "Stop", "Move fwd", "Approach to the shelf", "Prepare to travel", and "Move reverse". The work of taking out/unloading cargo corresponds to "Adjust heading" and "Load/Unload". The definition and number of each work state are not limited to the above, and can be changed as appropriate. Further, the working state may be defined only by the movement of the forklift 3 (industrial vehicle). FIG. 3(b) is a schematic diagram showing each working state when the forklift 1 takes out and unloads cargo from the shelf SF. In the database of the storage unit 15, each work state is associated with the photographing unit 12 that can image the location to be confirmed in the work state. Therefore, the display control unit 22 can display images of areas that should be checked on a large screen in the first area D1 and the second area D2 in accordance with the working status of the forklift 1 being operated by the operator. Note that the video displayed on the display unit 17 under the control of the display control unit 22 for driving support may be referred to as a “driving support video”.

There are individual differences among operators regarding the appropriate switching timing of driving support images. Further, when the state of the work performed by the forklift 1 changes, the image to be displayed on the display unit 17 changes. In view of the circumstances, the work state acquisition unit 23 and the viewpoint information acquisition unit 26 adapt the driving support video to the individual operator and perform processing corresponding to the state transition of the work state.

The work state acquisition unit 23 acquires the work state being performed by the forklift 1. The work status acquisition unit 23 acquires the work status based on the operation information acquired by the operation information acquisition unit 18. The work state acquisition unit 23 acquires work states using a work state estimation model in which estimation indicators such as transition conditions and transition thresholds for state transition to each work state are set. The work state acquisition unit 23 may acquire the work state estimation model stored in the storage unit 15. Details of the work status acquisition unit 23 will be described later. The work status acquisition unit 23 transmits the acquired work status to the viewpoint information acquisition unit 26.

The viewpoint information acquisition unit 26 acquires the operator's viewpoint based on the work status acquired by the work status acquisition unit 23. Here, the storage unit 15 has a database in which the working status of the forklift 1 and line-of-sight information based on the line-of-sight of the operator are linked. Therefore, the viewpoint information acquisition unit 26 acquires line-of-sight information corresponding to the working state by comparing the working state of the forklift 1 with the database. The viewpoint information acquisition unit 26 transmits the acquired viewpoint information to the display control unit 22. Thereby, the display control section 22 can control the image displayed on the display section 17 based on the viewpoint information, and can display the optimum image for the working condition of the forklift 1 and for the individual operator.

Next, with reference to FIG. 4, the detailed configuration of the work status acquisition section 23 will be described. FIG. 4 is a block diagram showing the configuration of the work status acquisition section 23. As shown in FIG. As shown in FIG. 4, the work state acquisition section 23 includes a work state estimation section 30, a cargo state estimation section 31, and a work state correction section 32.

The working state estimation unit 30 estimates the working state of the forklift 1 using machine learning. The work state estimation unit 30 receives the current operation information u(t) and outputs the current work state x(t) based on the learning result. The work state estimation unit 30 has a machine learning model M1.

As the work status, classifications "1" to "14" shown in FIG. 3(a) are used. The operation information includes at least one of an accelerator operation amount, a tire angle, a lift operation amount, a reach operation amount, and a tilt operation amount. The accelerator operation amount is a parameter indicating the operation amount of the accelerator of the forklift 1, and the larger the accelerator operation amount is, the faster the vehicle speed of the forklift 1 is. The tire angle is a parameter indicating the turning angle of the tires of the forklift 1, and the larger the tire angle, the more the forklift 1 turns. The lift operation amount is a parameter indicating the amount of operation for lifting the fork of the forklift 1, and the larger the lift operation amount is, the more the fork is lifted. The reach operation amount is a parameter indicating the amount of operation for extending the fork of the forklift 1 forward, and the larger the reach operation amount, the more the fork is moved forward. The tilt operation amount is a parameter indicating the amount of operation for tilting the fork of the forklift 1, and the larger the tilt operation amount, the greater the tilting of the fork.

The work state estimation unit 30 compares the input current operation information u(t) with a work state estimation model in which estimation indicators such as transition conditions and transition thresholds for state transition to each work state are set. Then, the current working state x(t) is estimated and output. In the work state estimation model, it is set which parameter among the above-mentioned operation information should become large (small) to cause a transition from one work state to another work state. For example, as shown in FIG. 3(b), in "Adjust heading" of "Working state 4", the forklift 1 is making a large turn, and in "Load" of "Working state 5" In this case, the forklift 1 is moving at a slow speed. Therefore, the work state estimating unit 30 determines that when the accelerator operation amount is large and the tire angle is large, the state changes from "work state 4" to "work state 5" when each parameter becomes less than a predetermined threshold value. It can be assumed that In such a work state estimation model, the optimum estimation index for the operator is set by machine learning based on the operator's past driving performance. Note that an individual work state estimation model may be created for each operator, or a work state estimation model that is not limited to individual operators may be created. The working state estimation model is stored in the storage unit 15, and the working state estimation unit 30 acquires an appropriate working state estimation model corresponding to the operator from the storage unit 15 at a necessary timing.

FIG. 5 is a block diagram showing the state of the work state estimating section 30 during learning. As shown in FIG. 5, the working state estimating unit 30 performs machine learning by inputting operation information regarding the operating state of the forklift 1 and inputting correct answer data regarding the working state. The machine learning process shown in FIG. 5 may be performed in advance before the actual remote control of the forklift 1 is performed. Data prepared in advance is used as the operation information and correct answer data. For example, the forklift 1 is actually operated in a laboratory or the like, and data is acquired by linking operation information at a certain time with the actual working state at that time. The actual working state at this time becomes the correct answer data.

The work state estimation unit 30 may use LSTM (Long Short Term Memory) as the machine learning model M1. At this time, the work state estimation unit 30 may perform machine learning using operation information acquired for each time series as learning data. The operation information acquired for each time series is, for example, a data group of “operation information u(t) and correct work state y(t)” acquired at predetermined intervals from “time 0” to “time T”. configured. The work state estimating unit 30 may use, as the machine learning model M1, an RNN (Recurrent Neural Network) that can respond to operation information acquired on a time-series basis.

Returning to FIG. 4, the cargo state estimation unit 31 outputs the cargo state g(t) regarding the state of the cargo handled by the forklift 1 based on the work state x(t) output by the work state estimation unit 30. The luggage state estimation unit 31 has a luggage state estimation model (state transition model) M2 created in advance. The baggage condition estimating section 31 acquires the baggage condition estimation model M2 from the storage section 15. The baggage condition estimating unit 31 estimates the baggage condition g(t) by comparing the estimated working condition x(t) with the baggage condition estimation model M2. The luggage status g(t) indicates either "NL: No Load (no luggage)" or "L: Load (with luggage)". The baggage condition estimation section 31 outputs the estimated baggage condition g(t) to the work condition correction section 32.

The baggage state estimation model M2 prohibits the baggage state from changing outside of the specified working state. For example, in the example of FIG. 3, in "Working State 5", the forklift 1 approaches the shelf SF without any cargo, and when it receives the cargo, it transitions to the "Working State 6", which is a preparation for traveling. In this case, the luggage state estimation model M2 allows a transition from "NL: No luggage" to "L: With luggage" when the state transitions from "Working state 5" to "Working state 6"; Transitions other than those specified are prohibited. Further, in "Working state 12", the forklift 1 approaches the shelf SF with the cargo there, and after unloading the cargo onto the shelf SF, the forklift 1 transitions to the "Working state 13", which is a preparation for traveling. In this case, the luggage state estimation model M2 allows a transition from "L: with luggage" to "NL: without luggage" when the state transitions from "work state 12" to "work state 13"; Transitions other than those specified are prohibited.

The work state correction unit 32 corrects the work state based on the work state x(t) outputted by the work state estimation unit 30 and the cargo state g(t) outputted by the cargo state estimation unit 31. The subsequent working state s(t) is output. The work state correction unit 32 determines whether the work state x(t) output by the work state estimation unit 30 is a work state with baggage or a work state without baggage. Next, the work state correction section 32 compares the presence or absence of the cargo in the work state x(t) with the cargo state g(t) output by the cargo state estimation section 31. If the presence or absence of luggage is the same in both cases, the work state correction unit 32 sets the same content as the work state x(t) as the corrected work state s(t). On the other hand, if there is a discrepancy in the presence or absence of luggage, the work state correction unit 32 corrects the work state x(t) to match the luggage state g(t). For example, while the work state estimating unit 30 estimates that there has been a transition from the work state (t-1) in which there is luggage to the work state (t) in which there is no luggage, the work state estimation unit 31 estimates that the work state has transitioned from the work state (t-1) with luggage to the work state (t) in which there is luggage. If it is estimated that there is no transition between (t-1) and the baggage state g(t) with baggage, the work state correction unit 32 corrects the work state x(t) with no baggage, Let the working state be s(t). In addition, in the table of FIG. 3(a), erroneous estimation is likely to occur within the same motion category. For example, in an operation called "Adjust heading", it is easy to mistake "Working state 4" and "Working state 11". Therefore, the work state correction unit 32 corrects the presence or absence of luggage within the same action category. Therefore, if "Working state 4" is incorrect, the working state correction unit 32 corrects it to "Working state 1" which is in the same operation category as "Working state 4". If "Working state 11" is incorrect, the working state correction unit 32 corrects the working state to "Working state 4" which is in the same operation category as "Working state 11".

Next, with reference to FIG. 6, an example of processing contents showing a driving support method by the control device 20 will be described. The process shown in FIG. 6 is performed when the driving support video is displayed on the display unit 17 and the operator is operating the forklift 1 by remote control. As shown in FIG. 6, the operation information acquisition unit 18 acquires operation information by the operator (step S10). Next, the work state estimation unit 30 estimates the work state based on the operation information acquired in step S (step S20). Next, the baggage condition estimating unit 31 obtains the baggage condition based on the working condition estimated in step S20 (step S30). Next, the work state correction unit 32 corrects the work state based on the work state estimated in step S20 and the cargo state estimated in step S30 (step S40). Next, the viewpoint information acquisition unit 50 acquires viewpoint information based on the work state corrected in step S50 (step S50). Next, the display control unit 22 selects a driving support video to be displayed on the display unit 17 based on the viewpoint information acquired in step S50 (step S60).

Next, the functions and effects of the control device 20 according to this embodiment will be explained.

The control device 20 includes a working state estimation unit 30 that estimates the working state of the forklift 1 using machine learning. Therefore, the control device 20 can design an estimation index for estimating the working state of the forklift 1 by machine learning. Here, the working state estimating unit 30 performs machine learning by inputting operation information regarding the operating state of the forklift 1 and inputting correct answer data regarding the working state. In this way, the work state estimating unit 30 performs machine learning based on the input of operation information that was actually performed in the past and the correct data indicating what kind of work state the operation information resulted in. I can do it. Thereby, the control device 20 can acquire an estimation index for estimating the working state of the forklift 1 by simply performing machine learning using past operation information and correct answer data. Further, since the estimation index is based on past actual data, it is possible to estimate the working state with high accuracy. As described above, the estimation index for estimating the working state of the forklift 1 can be obtained with high accuracy with a small number of man-hours.

The work state estimating unit 30 may receive current operation information and output the current work state based on the learning result. Thereby, the work state estimating unit 30 can easily estimate the current work state simply by inputting the current operation information.

The control device 20 includes a load state estimation section 31 that outputs a load state related to the state of the load handled by the forklift 1 based on the work state outputted by the work state estimation section 30; and a work state correction section 32 that corrects the work state based on the cargo state output from the cargo state estimation section 31. In this case, the control device 20 can improve the accuracy of estimating the working state by performing the correction in consideration of the cargo state of the forklift 1.

With reference to FIG. 7, the effects of the cargo condition estimating section 31 and the working condition correcting section 32 will be explained. FIG. 7A is a graph showing the estimation result of the working state estimated by the working state estimating unit 30 when the operation information at a certain time is shown in time series on the horizontal axis. FIG. 7B is a graph showing the estimation result when the work state correction unit 32 corrects the work state estimated by the work state estimation unit 30. The broken line graphs in FIGS. 7(a) and 7(b) are graphs indicating the presence or absence of luggage. When the dashed line graph exists in the upper row, it indicates a state of "with luggage", and when it exists in the lower row, it indicates a state of "without luggage".

For example, in FIG. 7(a), it is estimated that the state has transitioned from "Working state 4", in which the fork is adjusted in a state where there is no cargo, to "Working state 11", in which the fork is in position in a state in which there is a cargo, It is estimated that the state then transitioned to "work state 5", which is a cargo handling state with cargo present. In this way, as shown by "FD" in FIG. 7(a), in places where the estimation results rapidly fluctuate locally, the work state estimating unit 30 once estimates an incorrect work state, and then estimates the correct work state. Indicates that the working status is estimated.

On the other hand, the baggage state estimating unit 31 prohibits the state transition from "work state 4" to "work state 11" and the transition from "NL: no baggage" to "L: baggage present". Therefore, the baggage state estimating unit 31 outputs the baggage state “NL: No baggage” to the work state correction unit 32 in response to the estimation result of the work state estimating unit 30 which is “Working state 11” where there is baggage. Thereby, the work state correction unit 32 corrects the “work state 11” in which there is baggage to the “work state 4” in which there is no baggage in accordance with the estimation result of the baggage state estimation unit 31. As a result, as shown in FIG. 7(b), "Working state 4" does not go through "Working state 11" at the correct timing, that is, the actual timing of transition from "With baggage" to "No baggage". It can be estimated that the state has transitioned to "Working state 5". Thereby, in FIG. 7(b), it is possible to suppress the occurrence of an erroneous working state like "FD" in FIG. 7(a).

The operation information may include at least one of an accelerator operation amount, a tire angle, a lift operation amount, a reach operation amount, and a tilt operation amount. These parameters are parameters that reflect the intentions of the worker in the forklift 1. The work state estimation unit 30 can perform appropriate machine learning by using such parameters as operation information.

The work state estimating unit 30 may perform machine learning using LSTM as the machine learning model M1 and using operation information acquired for each time series as learning data. When performing machine learning using operation information acquired for each time series, LSTM can perform learning that takes into account not only the most recent data but also data from the past. Therefore, the work state estimating unit 30 can obtain a more accurate estimation index.

The driving support system 100 (industrial vehicle control system) according to this embodiment includes the above-described industrial vehicle control device 1.

The program for the industrial vehicle control device according to this embodiment is used in the industrial vehicle control device 20 described above.

Specifically, the program of the industrial vehicle control device causes the computer system to execute a working state estimation step of estimating the working state of the industrial vehicle using machine learning. In the working state estimating unit step, machine learning is performed by inputting operation information regarding the operating state of the industrial vehicle and inputting correct answer data for the working state.

These industrial vehicle control systems and the programs of the industrial vehicle control device 20 can obtain the same functions and effects as the industrial vehicle control device 20 described above.

Referring to FIG. 8, experimental results will be described when a working state is estimated using the control device 20 according to this embodiment. Here, for a specific work pattern, the percentage of agreement between the actual work state and the estimated result (accuracy rate) was calculated. In FIG. 8, a solid line graph G1 shows the estimation result, and a broken line graph G2 shows the actual working state. Among the graphs shown in Fig. 8, the estimated results match the actual working conditions in the locations where graphs G1 and G2 overlap, and the estimated results match the actual working conditions in locations where the graphs G1 and G2 do not overlap. Indicates that the A plurality of operators were prepared, work data was acquired 44 times, and the average value of the correct answer rate for all 44 times was calculated and evaluated. The accuracy rate of the estimation results of the control device 20 according to the embodiment improved by about 10% compared to the estimation results using a work state model designed without using machine learning.

Although some preferred embodiments of the present invention have been described above, the present invention is not limited to the above embodiments.

In the embodiments described above, the driving support system supports the operator's work during remote operation, but may also support when the industrial vehicle is operated by a man. Further, support may be provided when an operator performs a simulation operation of an industrial vehicle.

The industrial vehicle is not limited to a forklift, and a towing tractor, skid steer loader, etc. may also be employed.

[Form 1]
A control device for estimating the working state of an industrial vehicle, the control device comprising:
a working state estimation unit that estimates the working state of the industrial vehicle using machine learning;
The working state estimating unit is an industrial vehicle control device that performs machine learning by inputting operation information regarding the operating state of the industrial vehicle and inputting correct answer data for the working state.

[Form 2]
The industrial vehicle control device according to embodiment 1, wherein the work state estimating unit receives the current operation information and outputs the current work state based on a learning result.

[Form 3]
a cargo state estimating section that outputs a cargo state related to the state of cargo handled by the industrial vehicle based on the working state outputted by the working state estimating section;
According to form 1 or 2, the method further includes a working state correction unit that corrects the working state based on the working state outputted by the working state estimating unit and the cargo state outputted by the cargo state estimating unit. control equipment for industrial vehicles.

[Form 4]
The control device for an industrial vehicle according to any one of modes 1 to 3, wherein the operation information includes at least one of an accelerator operation amount, a tire angle, a lift operation amount, a reach operation amount, and a tilt operation amount.

[Form 5]
The industrial vehicle according to any one of modes 1 to 4, wherein the work state estimating unit uses LSTM as a machine learning model and performs machine learning using the operation information acquired for each time series as learning data. Control device.

[Form 6]
An industrial vehicle control system comprising the industrial vehicle control device according to any one of Embodiments 1 to 5.

[Form 7]
A program for an industrial vehicle control device used in the industrial vehicle control device according to any one of claims 1 to 5.

DESCRIPTION OF SYMBOLS 1... Forklift, 30... Work state estimation part, 31... Load state estimation part, 32... Work state correction part, 100... Driving support system (control system of industrial vehicle).

Claims (7)

A control device for estimating the working state of an industrial vehicle, the control device comprising:
a working state estimation unit that estimates the working state of the industrial vehicle using machine learning;
The working state estimating unit is an industrial vehicle control device that performs machine learning by inputting operation information regarding the operating state of the industrial vehicle and inputting correct answer data for the working state. The control device for an industrial vehicle according to claim 1, wherein the work state estimator receives the current operation information and outputs the current work state based on a learning result. a cargo state estimating section that outputs a cargo state related to the state of cargo handled by the industrial vehicle based on the working state outputted by the working state estimating section;
The work state correction unit according to claim 1, further comprising: a work state correction unit that corrects the work state based on the work state outputted by the work state estimation unit and the cargo state outputted by the cargo state estimation unit. Control equipment for industrial vehicles. The control device for an industrial vehicle according to claim 1, wherein the operation information includes at least one of an accelerator operation amount, a tire angle, a lift operation amount, a reach operation amount, and a tilt operation amount. The control device for an industrial vehicle according to claim 1, wherein the work state estimating unit uses LSTM as a machine learning model and performs machine learning using the operation information acquired for each time series as learning data. An industrial vehicle control system comprising the industrial vehicle control device according to any one of claims 1 to 5. A program for an industrial vehicle control device used in the industrial vehicle control device according to any one of claims 1 to 5.

Images