JP2023008644A - Information processing method, generating method of learning model, program, information processing equipment and excavation system - Google Patents

Abstract

To provide an information processing method or the like that supports an operator manipulating an excavator.SOLUTION: An information processing method includes obtaining observation data observed in excavation work with a jacking method, inputting obtained observation data to a model that outputs a prediction about the excavation work when inputting the observation data, and executing a processing to indicate information based on the prediction output from the model by a computer. The information processing method includes indicating an observation data obtained via a network and observed in excavation work with a jacking method, receiving directions for an excavator 50, and sending the directions to the excavator 50 via the network.SELECTED DRAWING: Figure 1

Description

The present invention relates to an information processing method, a learning model generation method, a program, an information processing device, and an excavation system.

Most lifelines such as electric power, gas, water and sewage, and communications are supplied to consumers through pipes buried underground.

A jacking method is used in which an excavator is installed in a starting hole and a driving pipe is laid from the starting hole to form a pipeline (Patent Document 1). Since it is a non-open-cut construction method that does not excavate the route in the middle, the construction occupied area and noise are small, and the impact on the lives of local residents can be suppressed.

JP 2015-166531 A

However, during excavation, obstacles such as hard bedrock or other buried pipelines may be found. In the jacking method, the operator cannot visually confirm obstacles, and must deal with obstacles based on little information such as sound and vibration. Therefore, an operator who operates an excavator must have sufficient knowledge and experience in order to properly lay the propulsion pipe according to the conditions of the construction site. However, skilled operators are in short supply.

An object of one aspect is to provide an information processing method or the like for supporting an operator who operates an excavator.

The information processing method acquires observation data observed during excavation work by the jacking method, inputs the acquired observation data into a model that outputs predictions related to excavation work when the observation data is input, and outputs the data from the model. The computer executes processing for displaying information based on the output prediction.

In one aspect, it is possible to provide an information processing method or the like for assisting an operator who operates an excavator.

It is an explanatory view explaining composition of an excavation system. It is an explanatory view explaining composition of an excavation system. It is an explanatory view explaining the composition of an excavator. It is an explanatory view explaining a record layout of excavation data DB. 4 is a flowchart for explaining the flow of processing of a program; FIG. 10 is a flowchart for explaining the flow of processing of a subroutine for abnormality determination; FIG. It is an example of a screen displayed on the display unit. It is an example of a screen displayed on the display unit. FIG. 4 is an explanatory diagram for explaining the configuration of a learning model; 4 is a flow chart illustrating the flow of program processing in the machine learning stage; FIG. 10 is a flowchart for explaining the flow of program processing in the stage of using a learning model; FIG. It is an example of a screen displayed on the display unit of the second embodiment. 11 is a flowchart for explaining the flow of processing of a program according to Embodiment 3; FIG. 11 is an explanatory diagram illustrating the configuration of an excavation system according to Embodiment 4; FIG. 11 is a functional block diagram of an excavation system according to Embodiment 5;

[Embodiment 1]
FIG. 1 is an explanatory diagram illustrating the configuration of an excavation system 10. As shown in FIG. The excavation system 10 of the present embodiment is a system for a jacking method that forms a pipeline by burying a propulsion pipe 52 while excavating horizontally or obliquely from a previously excavated starting hole 19 .

The excavation system 10 includes an excavation control device 40 , an excavator 50 , an external sensor 471 , a surveying device 472 , an information processing device 30 and a control panel 37 . The excavator 50, the external sensor 471 and the surveying device 472 are connected to the excavation control device 40 via wires or wirelessly. The excavation control device 40 and the information processing device 30 are connected via a network.

The excavator 50 has an excavating head 51 and is installed inside the launch hole 19 . The surveying device 472 is used for surveying the traveling direction of the drilling head 51 . Details of the information processing device 30, the excavation control device 40, the excavator 50, and the surveying device 472 will be described later.

The control panel 37 is connected to the information processing device 30 via wire or wireless. The control panel 37 has a display section 371, a joystick 372, control switches 373, and the like. The control panel 37 is used when an operator operates the excavator 50 . The control panel 37 has an appearance similar to that used by an operator to operate the excavator 50 at a conventional excavation site. By using the control panel 37, the operator can operate the excavator 50 in an operating procedure similar to that of a conventional excavation site operation.

The external sensor 471 is a sensor that collects data about the excavation situation, such as a microphone, vibration sensor, temperature sensor, and ground displacement sensor. External sensors 471 that collect various data are appropriately arranged in the starting hole 19, around the starting hole 19, on the ground near the excavation route, and the like. The external sensor 471 may be a so-called IoT (Internet of Things) device that is directly connected to the network without going through the excavation control device 40 .

FIG. 2 is an explanatory diagram for explaining the configuration of the excavation system 10. As shown in FIG. FIG. 2 shows details of the configuration of the information processing device 30 and the excavation control device 40 .

The information processing device 30 includes a control unit 31, a main storage device 32, an auxiliary storage device 33, a communication unit 34, an output unit 35, an input unit 36, a control panel I/F (Interface) 379, and a bus. The control unit 31 is an arithmetic control device that executes the program of this embodiment. One or a plurality of CPUs (Central Processing Units), GPUs (Graphics Processing Units), multi-core CPUs, or the like is used for the control unit 31 . The control unit 31 is connected to each hardware unit forming the information processing device 30 via a bus.

The main storage device 32 is a storage device such as SRAM (Static Random Access Memory), DRAM (Dynamic Random Access Memory), flash memory, or the like. The main storage device 32 temporarily stores information necessary during the process performed by the control unit 31 and the program being executed by the control unit 31 .

The auxiliary storage device 33 is a storage device such as SRAM, flash memory, hard disk, or magnetic tape. The auxiliary storage device 33 stores an excavation data DB (Database) 61, programs to be executed by the control unit 31, and various data necessary for executing the programs. The communication unit 34 is an interface that performs communication between the information processing device 30 and a network or other devices.

Output unit 35 includes display unit 351 and speaker 352 . The display unit 351 is, for example, a liquid crystal display panel or an organic EL (electro-luminescence) panel. The display unit 351 may be a separate display device from the information processing device 30 . The display unit 351 may be a wearable device such as an HMD (Head Mount display). The speaker 352 may be headphones, earphones, or the like that are separate from the information processing device 30 .

The input unit 36 is an input device such as a keyboard and mouse. The input unit 36 and the display unit 351 may be laminated to form a touch panel. The control panel I/F 379 is an interface that connects the information processing device 30 and the control panel 37 . A control panel I/F 379 is a general-purpose interface such as a USB (Universal Serial Bus).

The information processing device 30 is a general-purpose personal computer, tablet, smartphone, or the like used by an operator. The information processing device 30 may be configured by a combination of a large computer, a virtual machine operating on the large computer, a plurality of personal computers performing distributed processing, or a cloud computing system and a terminal device.

The excavation control device 40 includes a control section 41, a main memory device 42, an auxiliary memory device 43, a communication section 44, an excavator I/F 46, a measurement I/F 47, and a bus. One or a plurality of CPUs, GPUs, multi-core CPUs, or the like is used for the control unit 41 . The control unit 41 is connected to each hardware unit constituting the excavation control device 40 via a bus.

The main storage device 42 is a storage device such as SRAM, DRAM, flash memory, or the like. The main storage device 42 temporarily stores information necessary during the processing performed by the control unit 41 and the program being executed by the control unit 41 .

The auxiliary storage device 43 is a storage device such as SRAM, flash memory, hard disk, or magnetic tape. The auxiliary storage device 43 stores programs to be executed by the control unit 41 and various data necessary for executing the programs. The communication unit 44 is an interface that performs communication between the excavation control device 40 and the network.

Excavator I/F 46 is an interface that connects excavator control device 40 and excavator 50 . The measurement I/F 47 is an interface that connects the surveying device 472 and the external sensor 471 with the excavation control device 40 . Excavator I/F46 and measurement I/F47 are general-purpose interfaces, such as USB, for example.

The excavation control device 40 is a general-purpose personal computer, tablet, smartphone, or the like. The excavation control device 40 may be a control computer built into the excavator 50 .

FIG. 3 is an explanatory diagram for explaining the configuration of the excavator 50. As shown in FIG. The excavator 50 has a rotary section 55 , a retractable section 56 and an internal sensor 57 in addition to the aforementioned excavating head 51 . The rotating portion 55 and the advancing/retreating portion 56 are, for example, hydraulic drive mechanisms.

A drilling head 51 is connected to a leading propulsion tube 52 . Each time the drilling head 51 advances a distance corresponding to one propulsion tube 52 , a new propulsion tube 52 is connected behind the rearmost propulsion tube 52 . Since the step of connecting the propulsion pipe 52 is conventionally carried out by the jacking method, the explanation is omitted. After the drilling head 51 reaches the destination pit, the drilling head 51 and the leading propulsion pipe 52 are separated. A pipeline is formed by the above.

FIGS. 1 and 3 schematically show a press-fit excavating head 51 suitable for excavating soft ground and relatively soft ordinary ground. The tip of the press-fit excavation head 51 is a plane inclined with respect to the excavation direction, and receives pressure from the ground in the direction indicated by the thick arrow in FIG. The operator can adjust the direction of excavation by operating the rotary section 55 to rotate the excavation head 51 via the propulsion tube 52 .

The internal sensor 57 detects, for example, the thrust when the advancing/retreating section 56 advances and retreats the excavating head 51, the rotational torque when the rotating section 55 rotates the excavating head 51, the distance by which the excavating head 51 advances and retreats, and the angle by which the excavating head 51 rotates. , the power consumption of the excavator 50, the vibration state of the excavator head 51, and other conditions of the excavator 50 and the buried propelling pipe 52.

A target body 473 is arranged at the rear end of the drilling head 51 . The targets 473 are, for example, LEDs (Light Emitting Diodes) arranged at the center and periphery of the drilling head 51, respectively. Surveying instrument 472 is, for example, an electronic theodolite. The electronic theodolite is used for surveying with its optical axis facing the direction of the excavation target.

The electronic theodolite surveys the advancing direction of the drilling head 51 based on the positional relationship between the optical axis and the target object 473 . The direction of travel can be quantified, for example, by the distance a target 473 located in the center of the drilling head 51 deviates from the optical axis in the viewing optical field of view of an electronic theodolite. The direction of travel may be quantified by the angle of the direction of travel of the drilling head 51 with respect to the optical axis.

The surveying device 472 may be an electronic theodolite of a type that irradiates the rear end of the excavation head 51 with a laser and surveys the traveling direction of the excavation head 51 based on the reflected light instead of using the target 473 .

The surveying device 472 transmits survey results to the excavation control device 40 . The survey result may include a photographed image of the target 473 .

Returning to FIG. 2, the description continues. The survey result, the data detected by the internal sensor 57, and the data detected by the surveying device 472 are transmitted to the information processing device 30 in real time via the network. The control unit 31 records the received data in the excavation data DB 61 .

The control unit 31 displays on the display unit 371 the relationship between the position of the target object 473 photographed by the surveying device 472 and the optical axis. For example, the control unit 31 displays on the display unit 371 an image of the target object 473 photographed by the surveying device 472 and characters indicating the orientation of the excavation head 51 . The control unit 31 may reproduce the captured image of the target object 473 based on the surveyed orientation of the excavation head 51 . The operator can intuitively grasp the orientation of the drilling head 51, that is, the change in the orientation of the hole being drilled.

The control unit 31 displays other various data on the display unit 351 or the like. Examples of screens displayed on the display unit 351 will be described later.

The control panel 37 outputs audio data from the speaker 352 . Note that when a plurality of microphones are arranged, it is desirable that the operator can select which microphone's voice is to be output from the speaker 352 . The control unit 31 may synthesize voices from a plurality of microphones and output them from the speaker 352 .

The operator checks the survey data displayed on the display unit 371 , the sound output from the speaker 352 , the display on the display unit 351 , and the like, and operates the control switch 373 . Operations performed by the operator are transmitted to the excavation control device 40 via the information processing device 30 and the network. The excavation control device 40 controls the excavator 50 based on the operations performed by the operator.

When an abnormality occurs in the data, the control unit 31 notifies the operator of the occurrence of the abnormality, or automatically stops excavation. For example, when the drilling head 51 collides with another buried pipeline, hard bedrock, concrete structure, or the like, the direction of the drilling head 51 changes suddenly, and loud noise and vibration are generated.

The operator checks the data and, if necessary, contacts the site supervisor or the like to ask for instructions. For example, if the foreman directs the drilling head 51 to continue advancing to destroy the obstacle, the operator continues the drilling operation. Similarly, when the foreman instructs to bypass the obstacle, the operator pulls back the excavating head 51 once, adjusts the excavating direction, and resumes excavating.

Since the data on the excavation situation is digitized, the operator can send the data to information devices such as smartphones owned by the site supervisor as needed. Therefore, the operator and the site supervisor can share information sufficiently. Therefore, on-site supervisors can make accurate decisions based on accurate information.

As described above, the operator can operate the excavator 50 from his/her home or office away from the excavation site. By using the excavation system 10 of the present embodiment, the time required to travel to a remote excavation site can be saved, so that the actual work hours of the operator can be extended.

Note that the operator may operate a plurality of excavators 50 in parallel. For example, when drilling a plurality of holes in parallel, there is a low possibility that troubles such as obstacles will occur in the second and subsequent drilling. Thus, the operator can operate excavators 50 at different excavation sites in parallel. Therefore, it is possible to provide the excavation system 10 in which one operator can be in charge of a plurality of sites at the same time.

FIG. 4 is an explanatory diagram for explaining the record layout of the excavation data DB 61. As shown in FIG. The excavation data DB 61 is a DB that records various data measured by the external sensor 471, the surveying device 472 and the internal sensor 57 during excavation work, and the operation of the excavator 50 by the operator.

The excavation data DB 61 has an excavation ID (Identifier) field, an operator ID field, a start date/time field, an excavation time field, a result field, an excavation head field, and a data field during construction. The excavation data DB 61 has one record for one excavation. FIG. 4 schematically shows one record.

An excavation ID uniquely assigned to a hole to be excavated using the excavation system 10 is recorded in the excavation ID field. An operator ID uniquely given to an operator who is in charge of operating the excavation system 10 is recorded in the operator ID field. The date and time when excavation was started is recorded in the start date and time field. The excavation time field records the time required for excavation.

The excavation result is displayed in the result field. "Normal end" indicates that the excavation ended normally as planned. For example, if excavation is stopped because it intersects with another pipeline that has already been buried during excavation, the excavated distance and the operator or on-site supervisor will be displayed as follows: The result field records the reasons why it decided to stop drilling.

The type of excavation head 51 is recorded in the excavation head field. As described above, when the drilling head 51 for the press-fitting method is used, "for press-fitting method" is recorded in the drilling head field. For example, when a hydraulic balance type drilling head 51 suitable for water-retaining sand layers and hard ground is used, "for hydraulic balance type" is recorded in the drilling head field.

Similarly, when the impact crushing drilling head 51 suitable for gravel, cobblestone and rock layers is used, "for impact crushing method" is recorded in the drilling head field. When a drilling head 51 for a single pipe drilling method or a double pipe drilling method suitable for cohesive soil, sandy soil and gravel soil is used, "for single pipe drilling method" is used in the drilling head field. Or "For double pipe drilling method" is recorded.

Various chronological data such as thrust data, rotational torque data, survey data, acoustic data, vibration data, operation data and video data recorded during excavation work are recorded in the construction data field. The data recorded in the data field during construction is an example of observation data of the present embodiment observed during excavation work.

FIG. 4 shows examples of thrust data and operation data. The thrust data is data in which the propulsion distance of the drilling head 51 and the thrust applied by the advancing/retreating section 56 when advancing the drilling head 51 are recorded in association with each other, and are recorded in, for example, CSV (Comma Separated Value) format. . The propulsion distance is measured by an internal sensor 57, for example.

Similarly, the operation data is data in which the propulsion distance of the drilling head 51 and the operation performed by the operator using the control panel 37 are associated and recorded. "Start" indicates that the operator has performed an operation to start excavation. The advancing/retreating part 56 applies a predetermined thrust to the excavating head 51 to move it forward. "Right" means that the operator has performed an operation of tilting the joystick 372 in the "right" direction. "Thrust reduction by 10%" means that the operator has performed an operation to reduce the output of the advancing/retreating section 56 by 10%. "Forward" means that the operator has performed an operation of tilting the joystick 372 in the "forward" direction.

Although description of data examples is omitted, the rotational torque data is data in which the propulsion distance of the excavation head 51 and the rotational torque applied to the excavation head 51 by the rotating portion 55 are associated with each other. The survey data is data that associates the propulsion distance of the drilling head 51 and the drilling direction surveyed by the surveying device 472 . These data may be recorded in association with the time instead of the propulsion distance. Data that associates time and driving distance may be recorded in the data field during construction. Data relating the time and the external pressure applied to each portion of the propulsion pipe 52 may be recorded in the data field during construction.

The acoustic data is audio data recorded via microphones appropriately placed in the starting hole 19, around the starting hole 19, on the ground near the excavation route, and the like. Similarly, the vibration data is vibration data of the ground and excavator 50, etc., recorded via appropriately placed vibration sensors.

Although not shown, the excavation data DB 61 also records ground displacement data that records the ground displacement around the starting hole 19 and on the propulsion route measured by the ground displacement sensor.

Note that the excavation data DB 61 may have arbitrary fields for recording information on excavation work, such as fields for recording the model number and serial number of the excavation head 51, for example.

Each item shown in FIG. 4 is an example. For example, when excavating by the water pressure balance method, the pressure, amount and temperature of the face water supplied to the excavating head 51, the density of the excavated soil discharged from the excavator 50 side, etc. are recorded in the data field during construction. Also good. When excavation is performed by the impact crushing method, even if the air pressure supplied to the excavator 50, the density of the discharged soil discharged from the excavator 50 side, the particle diameter of the discharged soil, etc. are recorded in the data field during construction. good. In addition, data according to the excavation method, the type of the excavation head 51, the type of the excavator 50, the purpose of excavation, the type of the external sensor 471 that can be arranged, etc. may be recorded in the data during construction field.

FIG. 5 is a flowchart for explaining the processing flow of the program. The program shown in FIG. 5 is executed after preparatory work such as excavation of the starting hole 19, installation of the excavator 50, and alignment of the optical axis of the surveying instrument 472 is completed.

The control unit 31 transmits an activation preparation instruction to the excavation control device 40 (step S501). The control unit 41 receives the instruction (step S601). The control unit 41 transmits the construction-in-progress data obtained from the internal sensor 57, the external sensor 471 and the surveying device 472, and the data regarding the operating state of the excavator 50 to the information processing device 30 (step S602). By step S602, the control unit 41 realizes the function of the first transmission unit of the present embodiment, which transmits data to the information processing device 30 via the network.

Control unit 31 receives the data (step S502). The control unit 31 starts a subroutine for abnormality determination (step S503). The abnormality determination subroutine is a subroutine for determining whether or not there is an abnormality in the excavation state based on the data received from the excavation control device 40 . The processing flow of the abnormality determination subroutine will be described later.

The control unit 31 outputs the data received in step S502 in a format recognizable by the operator (step S504). Specifically, the control unit 31 generates an image of the target object 473 captured by the surveying device 472 based on the survey results, and displays the image on the display unit 371 . The control unit 31 outputs sound from the speaker 352 based on the sound data. The control unit 31 outputs to the display unit 351 a graph showing chronological changes in thrust and the like, as will be described later.

The control unit 31 receives an operator's instruction via the control panel 37 (step S505). The operator can use the joystick 372, the control switch 373, etc. to give instructions in the same operating procedure as at the excavation site. If the operator does not perform any operation for a predetermined time, the control unit 31 determines that it has received an instruction to "do not change the state of the excavator 50". That is, the operator does not need to operate the control panel 37, for example, to continue the excavation work under the current conditions.

The control unit 31 determines whether or not to end excavation (step S506). For example, when the operator operates a switch for instructing the end of excavation, the control unit 31 determines to end the excavation. When it is determined not to end the excavation (NO in step S506), the control unit 31 transmits the details of the operation received from the user to the excavation control device 40 (step S507). By step S507, the control unit 31 implements the function of the second transmission unit of the present embodiment, which transmits data to the excavation control device 40 via the network. After that, the control unit 31 returns to step S502.

The control unit 41 receives the operation content (step S611). The control unit 41 controls the excavator 50 to reflect the received operation content (step S612). The control unit 41 returns to step S602.

If it is determined to end excavation (YES in step S506), the control unit 31 transmits an instruction to end excavation to the excavation control device 40 (step S508). The control unit 41 receives an instruction to end excavation (step S613). The control unit 41 stops the operation of the excavator 50 by a predetermined termination process (step S614). The control unit 41 notifies the information processing device 30 that the operation of the excavator 50 has stopped (step S615).

The control unit 31 receives that the operation of the excavator 50 has stopped (step S509). The control unit 31 displays that the excavator 50 has stopped on the display unit 351 (step S510). After that, the control unit 31 terminates the process.

FIG. 6 is a flowchart for explaining the processing flow of the abnormality determination subroutine. The abnormality determination subroutine is a subroutine for determining whether or not there is an abnormality in the excavation state based on the data received from the excavation control device 40 .

The control unit 31 determines whether or not the thrust of the excavator 50 is within a predetermined normal range based on the data received from the excavation control device 40 (step S531). If it is determined that it is outside the range (NO in step S531), the control unit 31 determines that there is an abnormality in the thrust of the excavator 50, and temporarily stores it in the main storage device 32 or the auxiliary storage device 33 (step S541). ).

If it is determined that it is within the normal range (YES in step S531), or after the end of step S541, the control unit 31 controls the rotational torque of the excavator 50 to fall within the predetermined normal range based on the data received from the excavation control device 40. It is determined whether or not it is within (step S532). If it is determined that it is outside the range (NO in step S532), the control unit 31 determines that there is an abnormality in the rotational torque of the excavator 50, and temporarily stores it in the main storage device 32 or the auxiliary storage device 33 (step S542).

If it is determined that it is within the normal range (YES in step S532), or after the end of step S542, the control unit 31 determines that the traveling direction of the excavation head 51 is within the predetermined normal range based on the data received from the excavation control device 40. It is determined whether or not it is within (step S533). If it is determined that it is outside the range (NO in step S533), the control unit 31 determines that there is an abnormality in the traveling direction of the excavating head 51, and temporarily stores it in the main storage device 32 or the auxiliary storage device 33 (step S543).

If it is determined to be within the normal range (YES in step S533), or after step S543 is completed, the control unit 31 performs frequency analysis of the vibration data based on the data received from the excavation control device 40 to determine the frequency characteristic. Calculate (step S534). Algorithms for frequency analysis such as FFT (Fast Fourier Transform) are commonly used, and therefore detailed descriptions thereof are omitted.

The control unit 31 performs pattern matching between the frequency characteristics calculated in step S534 and frequency characteristics of vibration data previously measured for various grounds, and extracts grounds exhibiting similar frequency characteristics (step S535). The ground includes, for example, "cohesive soil", "soft ground", "bedrock", "concrete", and "buried pipeline".

The control unit 31 performs frequency analysis of the acoustic data based on the data received from the excavation control device 40, and calculates frequency characteristics (step S536). The control unit 31 performs pattern matching between the frequency characteristics calculated in step S536 and the frequency characteristics of acoustic data previously measured for various grounds, and extracts grounds exhibiting similar frequency characteristics (step S537). After that, the control unit 31 terminates the process.

7 and 8 are examples of screens displayed on the display unit 351. FIG. The control unit 31 outputs the screen shown in FIG. 7 or 8 to the display unit 351 in step S504 of the program described using FIG.

FIG. 7 shows an example when there is no abnormality in the excavation state. A construction condition column 71, a management condition column 72, an operating condition column 73, and an observation data column 74 are displayed on the left side of the screen. A thrust force graph column 751, a rotational torque graph column 752, and a direction graph column 753 are displayed on the right side of the screen.

The construction condition column 71 displays the construction conditions determined at the time of the construction plan. The management condition column 72 displays the management conditions determined at the time of construction planning. Construction conditions and management conditions are acquired from a database or the like in which construction plans are recorded.

The operating status column 73 displays the operating status of the excavator 50 in real time. A vibration data column 741 , an acoustic data column 742 and a notification column 748 are displayed in the observation data column 74 . In the vibration data column 741, points to which the operator should pay attention, determined based on the vibration data among the observation data acquired from the excavation control device 40 in step S502 described using FIG. 5, are briefly displayed. ing. Similarly, in the acoustic data column 742, the points to which the operator should pay attention, determined based on the acoustic data, are briefly displayed.

The notification column 748 briefly displays whether or not there is an abnormality in the observation data. FIG. 7 shows an example in which it is determined that there is no abnormality in any of steps S531 to S533, steps S535, and S537 described using FIG.

The horizontal axis of the thrust force graph column 751 is the thrust distance of the drilling head 51 . The unit of the horizontal axis is meters. The minimum value on the horizontal axis is 0 meters, and the maximum value on the horizontal axis is the planned excavation distance.

The vertical axis of the thrust force graph column 751 is the thrust applied to the excavation head 51 . The units of the vertical axis are kilonewtons. The dashed line indicates the thrust reference value. The thrust is determined to be abnormal when it exceeds the reference value. The solid line indicates the measured value of thrust.

The horizontal axis of the rotational torque graph column 752 is the same as the horizontal axis of the thrust force graph column 751 . The vertical axis of the rotational torque graph column 752 is the rotational torque applied to the excavation head 51 . The units of the vertical axis are kilonewton meters. A dashed line indicates the reference value of the rotational torque. The rotational torque is determined to be abnormal when it exceeds the reference value. A solid line indicates the measured value of the rotational torque.

The horizontal axis of the direction graph column 753 is the same as the horizontal axis of the thrust force graph column 751 . The vertical axis of the direction graph column 753 is the distance by which the target 473 located in the center of the digging head 51 is shifted from the optical axis. The units of the vertical axis are centimeters. The dashed line indicates the distance reference value. If the distance exceeds the reference value on either the positive side or the negative side, it is determined that there is an abnormality. A thin solid line indicates a measured value in the horizontal direction. A thick solid line indicates a measured value in the vertical direction.

The operator can grasp chronological changes in the excavation situation from the start of excavation to the present time by using the thrust force graph column 751 , the rotational torque graph column 752 and the direction graph column 753 . The operator controls the excavator 50 while checking the information displayed on the screen.

FIG. 8 is an example of a screen displayed on the display unit 351 by the control unit 31 when an abnormality occurs in vibration data and sound data. FIG. 8 shows the case where the drilling head 51 comes into contact with concrete. Concrete patterns are detected in both the vibration and acoustic data. A notification column 748 displays that there is an abnormality. Note that the control unit 31 may call the operator's attention by means of a warning sound, blinking of the screen, or the like.

The operator takes measures such as emergency stop of the excavator 50, and considers countermeasures after that. Note that the screens shown in FIGS. 7 and 8 are examples. Items and layouts to be displayed are not limited to those shown in FIGS.

In excavation work using the jacking method, even if the operator is at the excavation site, the operator cannot directly check the state of the excavation head 51 and the state of the ground being excavated. On the other hand, if a sufficient number of external sensors 471 and internal sensors 57 are appropriately arranged and a communication line of sufficient quality is secured, the excavation system 10 of the present embodiment can be used by the operator to perform excavation. Detailed information about the state can be obtained.

Therefore, by using the excavation system 10 of the present embodiment, the operator can make appropriate judgments based on detailed information whether the operator is at the excavation site or at a location away from the excavation site. to carry out excavation work.

The output unit 35 may include a bodysonic chair. The controller 31 vibrates the body sonic chair based on the vibration data, so that the seated operator can perceive the vibration data using the whole body.

Instead of control panel 37, output 35 and input 36 may be used. The operator can operate the excavator 50 using general hardware such as a general-purpose personal computer and software for controlling the excavator 50 without preparing special hardware.

According to this embodiment, it is possible to provide the excavation system 10 that allows the operator to operate the excavator 50 from home, office, or the like. It saves time traveling to a remote excavation site, allowing operators to work longer hours. Since the operator can work at home or in an office with air conditioning, the working conditions of the operator can be improved. As described above, the excavation system 10 of the present embodiment can contribute to solving the shortage of operators.

By mastering the excavation technique, even an operator who finds it difficult to work at an excavation site due to a disability, childcare, nursing care, or the like, can operate the excavator 50 . Therefore, the excavation system 10 of the present embodiment can contribute to the creation of employment for persons with disabilities, the improvement of work-life balance, the elimination of the gender gap, and the like, thereby contributing to the resolution of various social problems.

According to this embodiment, for example, when a plurality of microphones are arranged, the operator can appropriately switch the microphone for listening to the acoustic data. For example, when an abnormal sound is generated, the operator can properly grasp the state of the excavator 50 by sequentially switching the microphones arranged along the excavation route and confirming the sound at each position.

According to the present embodiment, when a plurality of data such as thrust force data collected during excavation exceeds a predetermined threshold value, the control unit 31 displays it on the display unit 351 to call the operator's attention. Further, the control unit 31 displays chronological changes in the state of the excavator 50 using graphs such as the thrust force graph column 751 . By doing so, it is possible to provide the excavation system 10 that supports the operator's judgment.

A plurality of information processing devices 30 may be connected to the network. For example, when an inexperienced operator uses one information processing device 30 to take charge of excavation work, a skilled operator uses another information processing device 30 to observe the state of the work and provide guidance as necessary. can. Conversely, a plurality of operators in training can use the information processing device 30 to observe the state of excavation work of a skilled operator. Therefore, it is possible to provide the excavating system 10 that is useful for operator's OJT (On the Job Training) and contributes to solving the shortage of human resources.

According to the present embodiment, information used by the operator to grasp the state of the excavator 50 during excavation work and operations performed by the operator are all computerized and transmitted via the network. A drilling system 10 can be provided that can electronically record all information regarding the progress of a drilling operation.

For example, when a sudden trouble occurs, the control unit 31 may extract information immediately before the trouble occurs from electronically recorded information and present it to the operator. The operator can appropriately estimate the cause of the trouble and plan countermeasures by reconfirming the state before the trouble occurred and when the trouble occurred. Other examples of how to utilize electronically recorded information will be described later.

The external sensors 471 may include appropriately positioned cameras, and the data during construction may include video data recorded via the cameras. It is possible to provide an excavation system 10 that appropriately records the construction status of excavation work. If the capacity of the communication line is insufficient, the control unit 41 may temporarily record the moving image data in the auxiliary storage device 43 and transmit it to the information processing device 30 after excavation is completed.

[Embodiment 2]
This embodiment relates to a drilling system 10 that uses a learning model 65 to assist the operator. Descriptions of parts common to the first embodiment are omitted.

FIG. 9 is an explanatory diagram for explaining the configuration of the learning model 65. As shown in FIG. The learning model 65 is a model that receives inputs such as data during construction that have been received from the excavation control device 40 so far, and outputs a prediction regarding the next operation that the operator will perform.

The input data input to the learning model 65 are, for example, thrust data, rotational torque data, survey data, acoustic data, vibration data and operation data described using FIG. Note that the input data may be part of the data listed above. The input data may include geotechnical results of the excavation site. The learning model 65 is stored in the auxiliary storage device 33 or an external large-capacity storage device connected to the information processing device 30 .

In the example shown in FIG. 9, the operator has an 80% chance of continuing excavation without operating the excavator 50, and a 2% chance of operating the excavator head 51 to the right. The probability of performing an operation to retract the head 51 is 1%, and the probability of pressing the emergency stop button is 1%.

The teacher data used to generate the learning model 65 is, of the data recorded in the excavation data DB 61 in the past, data relating to excavation that was "normally completed" by a skilled operator, for example. The learning model 65 is created using any machine learning technique suitable for predicting time-series data, such as RNN (Recurrent Neural Network) or LSTM (Long Short-Term Memory).

Learning model 65 may be created using any machine learning technique, such as CNN (Convolutional neural network) or random forest. The learning model 65 may be created by reinforcement learning that determines rewards based on digging time.

The learning model 65 is generated for each type of excavating head 51, such as for the press-fitting method and for the hydraulic balance method. The learning model 65 may be generated for each model of the drilling head 51 . The learning model 65 may be generated for each type of soil to be excavated. The learning model 65 may also be generated for each depth at which the propulsion pipe 52 is buried. The control unit 41 selects and uses a learning model 65 corresponding to excavation conditions.

FIG. 10 is a flow chart illustrating the flow of program processing in the machine learning stage. In the following description, the case where the information processing device 30 is used to generate the learning model 65 will be described as an example. The program of FIG. 10 may be executed by hardware different from the information processing device 30, and the learning model 65 for which machine learning has been completed may be copied to the auxiliary storage device 33 via the network. A learning model 65 learned by one piece of hardware can be used in a plurality of information processing apparatuses 30 .

Prior to executing the program of FIG. 10, an unlearned learning model 65 such as RNN or LSTM is prepared. Each parameter of the prepared learning model 65 is adjusted by the program of FIG. 10, and machine learning is performed.

A training data DB is constructed by extracting records that meet conditions from the excavation data DB 61 . The configuration of the training data DB is the same as that of the excavation data DB 61 . In the following description, records recorded in the training data DB are referred to as training records. By creating a training data DB using the excavation data DB 61 in which data is recorded in each of the plurality of information processing devices 30, a training data DB having a large number of training records can be created.

The control unit 31 acquires a training record used for one epoch training from the training data DB (step S551). The number of training records used for training in one epoch is a so-called hyperparameter and is determined appropriately.

The control unit 31 controls the input layer of the learning model 65 so that when the data during construction until the excavation head 51 advances to a certain position is input, the next operation performed by the operator is output from the output layer. The parameters of the model are adjusted (step S552).

The control unit 31 determines whether or not to end the process (step S553). For example, the control unit 31 determines to end the process when learning for a predetermined number of epochs is completed. The control unit 31 may obtain test data from the training data DB, input it to the model during machine learning, and determine to end the process when an output with a predetermined accuracy is obtained.

If it is determined not to end the process (NO in step S553), the control unit 31 returns to step S551. If it is determined to end the process (YES in step S553), the control unit 31 records the learned model parameters in the auxiliary storage device 33 (step S554). After that, the control unit 31 terminates the process. Through the above processing, the machine learning of the learning model 65 is completed.

FIG. 11 is a flow chart for explaining the flow of program processing in the use stage of the learning model 65 . The program of FIG. 11 is used during excavation work instead of the program of the first embodiment described using FIG.

The processing up to step S502 is the same as the processing flow of the program of the first embodiment described using FIG. 5, so the description is omitted. The control unit 31 selects the learning model 65 corresponding to the drilling head 51 in use. The control unit 31 inputs the construction-in-progress data received in step S502 from the start of excavation to that time into the selected learning model 65, and obtains prediction of the operation to be performed next by the operator (step S561). .

The control unit 31 outputs the data received in step S502 and the prediction obtained in step S561 in a format recognizable by the operator (step S562). The control unit 31 receives an operator's operation via the control panel 37 (step S505). Since the subsequent processing is the same as the processing flow of the program of the first embodiment described using FIG. 5, the description is omitted.

FIG. 12 shows an example of a screen displayed on the display unit 351 according to the second embodiment. The control unit 31 outputs the screen shown in FIG. 11 in step S562 of the program described using FIG. The basic configuration of the screen shown in FIG. 12 is the same as that shown in FIGS. 7 and 8, and a recommended operation column 76 is added to the driving situation column 73. The "right" operation obtained from the learning model 65 is displayed in the recommended operation column 76, indicating that a skilled operator is likely to perform the operation of turning the excavating head 51 to the right.

The control unit 31 may determine the display mode of the recommended operation column 76 based on the probability described using FIG. For example, the control unit 31 displays the recommended operation field 76 in large characters when the probability is high, and displays the recommended operation field 76 in small characters when the probability is low. When the learning model 65 outputs a plurality of operations with similar probabilities, the control unit 31 may display the plurality of operations in the recommended operation field 76 .

According to the present embodiment, it is possible to provide the excavation system 10 that assists the operator by constantly presenting predictions regarding operations performed by a skilled operator.

The information processing device 30 is arranged at the excavation site, and the operator may operate the excavator 50 at the excavation site. When the information processing device 30 is arranged at the excavation site, the information processing device 30 and the excavation control device 40 may be connected by a wire.

[Embodiment 3]
This embodiment relates to an excavation system 10 that performs automatic excavation using a learning model 65 that has been trained to output sufficiently reliable predictions. The description of the parts common to the second embodiment is omitted.

FIG. 13 is a flowchart for explaining the processing flow of the program according to the third embodiment. The processing up to step S502 is the same as the processing flow of the program of the first embodiment described using FIG. 5, so the description is omitted.

The control unit 31 determines whether or not the scheduled excavation has ended (step S571). Specifically, when the propulsion distance included in the data received in step S502 reaches the target, the control unit 31 determines that the excavation has ended.

When it is determined that the excavation has ended (YES in step S571), the control unit 31 transmits an instruction to end excavation to the excavation control device 40 (step S508). Since the subsequent processing is the same as the processing of the program of the first embodiment described using FIG. 5, the description is omitted.

If it is determined that the excavation has not been completed (NO in step S571), the control unit 31 inputs the construction-in-progress data received in step S502 from the start of excavation to that time into the learning model 65, and then the operator acquires the prediction of the operation performed by (step S572).

The control unit 31 determines whether or not the prediction obtained in step S572 indicates that the excavator 50 will stop (step S573). If it is determined to be stopped (YES in step S573), the control unit 31 notifies the operator that the excavator 50 is to be stopped even though the excavation is not completed (step S574). The notification can be made by any means such as warning sound, blinking of the screen, message transmission to the smart phone possessed by the operator, or the like.

If it is determined not to stop (NO in step S573), or after step S574 ends, the information processing device 30 transmits the operation acquired in step S572 to the excavation control device 40 (step S575). After that, the control unit 31 returns to step S502.

The control unit 41 receives the operation content (step S611). Since the subsequent processing is the same as the processing of the program of the first embodiment described using FIG. 5, the description is omitted.

According to this embodiment, it is possible to provide the drilling system 10 that automatically drills using the learning model 65 . For example, when the excavator head 51 collides with another buried pipeline, hard bedrock, concrete structure, etc., in an irregular situation, the excavator 50 is automatically stopped and the operator is notified. A drilling system 10 can be provided that

[Embodiment 4]
The present embodiment relates to a mode of realizing the excavation system 10 of the present embodiment by operating a general-purpose computer 90 and a program 97 in combination. FIG. 14 is an explanatory diagram showing the configuration of the excavation system 10 of Embodiment 4. As shown in FIG. Descriptions of parts common to the first embodiment are omitted.

The excavation system 10 of this embodiment includes a computer 90 instead of the information processing device 30 . The computer 90 includes a control section 31, a main storage device 32, an auxiliary storage device 33, a communication section 34, an output section 35, an input section 36, a control panel I/F 379, a reading section 39 and a bus. The computer 90 is an information device such as a general-purpose personal computer, tablet, or server computer.

A program 97 is recorded on a portable recording medium 96 . The control unit 31 reads the program 97 via the reading unit 39 and stores it in the auxiliary storage device 33 . Control unit 31 may also read program 97 stored in semiconductor memory 98 such as a flash memory mounted in computer 90 . Furthermore, the control unit 31 may download the program 97 from another server computer (not shown) connected via the communication unit 34 and a network (not shown) and store it in the auxiliary storage device 33 .

A part of the program 97 that is executed by the control unit 31 is installed as a control program of the computer 90, loaded into the main storage device 32, and executed. A portion of the program 97 executed by the control unit 41 is transmitted to the excavation control device 40 via the network and stored in the auxiliary storage device 43 of the excavation control device 40 . The stored program is installed as a control program of the excavation control device 40, loaded into the main storage device 42 and executed.

As described above, the computer 90 and the excavation control device 40 cooperate to function as the excavation system 10 described above.

[Embodiment 5]
FIG. 15 is a functional block diagram of excavation system 10 of Embodiment 5. As shown in FIG. The excavation system 10 includes an excavator 50 for a jacking method, an excavation control device 40 , and an information processing device 30 . The excavation control device 40 includes an acquisition section 83 and a first transmission section 81 . The information processing device 30 includes a display section 84 , a reception section 85 and a second transmission section 82 .

The acquisition unit 83 acquires observation data observed during excavation work by the excavator 50 . The first transmission unit 81 transmits observation data to the information processing device 30 via the network. The display unit 84 displays observation data. The receiving unit 85 receives instructions for the excavator 50 . The second transmission unit 82 transmits the instruction to the excavation control device 40 via the network. The excavation control device 40 controls the excavator 50 based on the instructions.

The technical features (constituent elements) described in each embodiment can be combined with each other, and new technical features can be formed by combining them.
The embodiments disclosed this time are illustrative in all respects and should be considered not restrictive. The scope of the present invention is indicated by the scope of the claims rather than the meaning described above, and is intended to include all modifications within the scope and meaning equivalent to the scope of the claims.

10 excavation system 19 starting hole 30 information processing device 31 control unit 32 main storage device 33 auxiliary storage device 34 communication unit 35 output unit 351 display unit 352 speaker 36 input unit 37 control panel 371 display unit 372 joystick 373 control switch 379 control panel I /F
39 reading unit 40 excavation control device 41 control unit 42 main storage device 43 auxiliary storage device 44 communication unit 46 excavator I/F
47 Measurement I/F
471 external sensor 472 surveying device 473 target body 50 excavator 51 excavation head 52 propulsion pipe 55 rotating part 56 advance/retreat part 57 internal sensor 61 excavation data DB
65 learning model 71 construction condition column 72 management condition column 73 operating condition column 74 observation data column 741 vibration data column 742 acoustic data column 748 notification column 751 thrust force graph column 752 rotating torque graph column 753 direction graph column 76 recommended operation column 81 first Transmission unit 82 Second transmission unit 83 Acquisition unit 84 Display unit 85 Reception unit 90 Computer 96 Portable recording medium 97 Program 98 Semiconductor memory

Claims (15)

Acquiring observation data observed during excavation work by jacking method,
An information processing method in which a computer executes a process of inputting acquired observation data into a model that outputs a prediction about an excavation work when observation data is input, and displaying information based on the prediction output from the model. 2. The information processing method according to claim 1, wherein said observation data includes data regarding the orientation of a hole being drilled. The information processing method according to claim 1 or 2, wherein the observation data includes data relating to sound of excavation work. Get information about the drilling head in use,
Select one model based on the acquired information from a plurality of models that output predictions regarding excavation work when observation data is input. The information processing method according to any one of claims 1 to 3, wherein information based on prediction is displayed. 5. The information processing method according to any one of claims 1 to 4, further comprising: displaying the relationship between the driving distance of the excavation work and the thrust during excavation based on the obtained observation data. Display observation data obtained through the network and observed during excavation work using jacking method,
receive instructions for the excavator,
An information processing method in which a computer executes a process of transmitting the instruction to the excavator via a network. 7. The information processing method according to claim 6, wherein the state of excavation work is determined based on the observation data. 8. Information according to claim 6 or claim 7, wherein the obtained observation data is input to a model that outputs a prediction about excavation work when observation data is input, and information based on the prediction output from the model is displayed. Processing method. 9. The information processing method according to claim 7 or 8, wherein when the excavation work status indicates that it is necessary to operate the excavator, information on the required operation is displayed. 10. The information processing method according to any one of claims 6 to 9, wherein the observation data includes data regarding the direction of the hole being drilled. The information processing method according to any one of claims 6 to 10, wherein the observation data includes data relating to sound of excavation work. Observation data including a chronological relationship between an operation of an excavator for a jacking method and an operating state of the excavator is acquired as training data recorded for each of a plurality of excavations,
Observation data up to a certain point in time is input, the next operation of the excavator is output, and a learning model is generated that outputs a prediction about the next operation of the excavator when observation data is input. Method of generating a learning model. Acquiring observation data observed during excavation work by jacking method,
A program that causes a computer to execute a process of inputting acquired observation data into a model that outputs predictions regarding the state of excavation work when observation data is input, and displaying information based on predictions output from the said model. an acquisition unit that acquires observation data observed during excavation work by the jacking method;
An information processing apparatus comprising: a display unit for inputting acquired observation data into a model for outputting a prediction regarding the state of excavation work when observation data is input, and for displaying information based on the prediction output from the model. In an excavation system comprising an excavator for a jacking method, an excavation control device, and an information processing device,
The excavation control device includes:
an acquisition unit that acquires observation data observed during excavation work by the excavator;
a first transmission unit that transmits the observation data to the information processing device via a network;
The information processing device is
a display unit that displays the observation data;
a reception unit that receives instructions for the excavator;
and a second transmitter that transmits the instruction to the excavation control device via a network, wherein the excavation control device controls the excavator based on the instruction.

Images