Excavation method

Abstract

To provide an excavation method that enables efficient excavation. [Solution] The present invention provides an excavation method in which a computer performs a process that includes an excavation step of performing excavation while moving a construction machine, acquires first data showing the change in the amount of soil loaded onto the construction machine over time in the excavation step of the first process, and uses the acquired first data to set at least one of the excavation time and excavation amount based on the change in the amount of soil loaded onto the construction machine over time.

Description

Technical Field

[0006] , , ,

[0005] , ,

[0001] The present invention relates to an excavation method.

Background Art

[0002] Conventionally, a scraper vehicle equipped with a scraper for excavating the ground or the like has been used at a civil engineering site. The scraper vehicle is provided with a bowl for accommodating the excavated material excavated by the scraper. Measuring the weight of the excavated material accommodated in this bowl is described in Patent Document 1.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] As the excavated material excavated by the scraper is loaded into the bowl, the earth pressure in the bowl increases as the loading amount in the bowl increases, making it difficult to load the excavated material into the bowl. Therefore, if an attempt is made to load the excavated material until the bowl is full, the loading efficiency may deteriorate.

[0005] On one side, an object of the present invention is to provide an excavation method capable of efficiently performing excavation.

Means for Solving the Problems

[0006] The present invention provides an excavation method in which a computer performs a process that includes an excavation step in which excavation is performed while moving a construction machine, acquires first data showing the change in the amount of soil loaded onto the construction machine over time in the excavation step of the first process, and uses the acquired first data to set at least one of the excavation time and excavation amount based on the change in the amount of soil loaded onto the construction machine over time. [Effects of the Invention]

[0007] The drilling method of the present invention has the effect of enabling efficient drilling. [Brief explanation of the drawing]

[0008] [Figure 1] Figure 1 shows a towing vehicle and a scraper vehicle according to one embodiment. [Figure 2] Figure 2 is a block diagram showing the control system of a towing vehicle and construction machinery according to one embodiment. [Figure 3] Figure 3 illustrates one cycle of processing performed when a scraper vehicle is towed by a towing vehicle. [Figure 4] Figure 4 is a graph showing the time-dependent change in loading volume when the bowl of a scraper vehicle is fully loaded with excavated material. [Figure 5] Figures 5(a) and 5(b) illustrate the method for determining the optimal drilling volume and optimal drilling time. [Figure 6] Figure 6 is a flowchart showing the processing steps of the control device of a scraper vehicle. [Figure 7] Figure 7 is a flowchart showing the specific processing steps in step S30 of Figure 6. [Figure 8] Figures 8(a) to 8(c) are diagrams (part 1) illustrating the processing content of step S48 in Figure 7. [Figure 9] Figures 9(a) to 9(c) are diagrams (part 2) illustrating the processing content of step S48 in Figure 7. [Figure 10]Figure 10 is a diagram (part 3) illustrating the processing content of step S48 in Figure 7. [Modes for carrying out the invention]

[0009] One embodiment will be described in detail below with reference to the drawings. However, the present invention is not limited to the embodiment described below.

[0010] Figure 1 shows a towing vehicle 1 and a scraper vehicle 20 as a construction machine according to one embodiment. The scraper vehicle 20 is a towed vehicle that is towed by the towing vehicle 1. In Figure 1, the front of the towing vehicle 1 in the direction of travel is the -X direction, the rear is the +X direction, the left side of the towing vehicle 1 in the direction of travel is the -Y direction, the right side is the +Y direction, the upper vertical side is the +Z direction, and the lower vertical side is the -Z direction.

[0011] (Towing vehicle 1) The towing vehicle 1 tows the scraper vehicle 20, and the scraper vehicle 20 is connected to the towing vehicle 1 via a coupling part (hitch 21) on the scraper vehicle 20. A flexible ball joint 22 is provided at one end of the hitch 21 (the -X side end), and the ball joint 22 is detachably attached to the towing vehicle 1. A flexible ball joint (not shown) is also provided at the other end of the hitch 21 (the +X side end).

[0012] (Scraper vehicle 20) As shown in Figure 1, the scraper vehicle 20 comprises the aforementioned hitch 21, frame 23, bowl 24, scraper 25, axle 26, and wheel 27.

[0013] Frame 23 is a metal frame that supports structures such as bowl 24. Bowl 24 is open on the top (+Z side) and is used to contain excavated materials such as soil and sand excavated by scraper 25.

[0014] The scraper 25, also called a cutter, is a blade-shaped or spatula-shaped member for scraping off earth and sand on a traveling surface such as the ground surface. In the present embodiment, the scraper 25 is provided integrally with the bowl 24 at the bottom of the bowl 24.

[0015] Since the bowl 24 and the scraper 25 are provided integrally, by inclining the bowl 24 toward the ground by the cutter drive device (hydraulic cylinder) 32 in FIG. 2, the scraper 25 can be made to bite into the ground to excavate earth and sand. Also, by adjusting the degree to which the scraper 25 bites into the ground by the cutter drive device 32, the excavation degree and the speed of the scraper 25 can be adjusted. Further, the bowl 24 is provided with an opening (not shown), and in a state where the bowl 24 is inclined toward the ground, the excavated material excavated by the scraper 25 is accommodated in the bowl 24 through the opening.

[0016] When the excavation by the scraper 25 is completed, by inclining the bowl 24 away from the ground by the cutter drive device 32, the scraper 25 is in a state of being separated from the ground.

[0017] The bowl 24 is provided with a gate (not shown) for discharging (unwinding) the excavated material accommodated in the bowl 24 to the outside of the bowl 24, and the gate is opened and closed by the gate drive device (hydraulic cylinder) 34 in FIG. 2. Further, an ejector (not shown) for pushing out the excavated material accommodated in the bowl 24 to the outside of the bowl 24 is provided in the bowl 24. The ejector is driven in the X-axis direction by the ejector drive device (hydraulic cylinder) 36 in FIG. 2.

[0018] The axle 26 rotates by the traction force of the towing vehicle 1. Wheels 27 are connected to both ends of the axle 26. The wheels 27 are a pair of driven wheels that rotate as the axle 26 rotates. Note that a plurality of pairs of the wheels 27 may be provided in front of and behind the scraper vehicle 20 (that is, front wheels and rear wheels may be provided).

[0019] Figure 2 is a block diagram showing the control systems for the towing vehicle 1 and the scraper vehicle 20.

[0020] As shown in Figure 2, the scraper vehicle 20 includes a control device 30, a cutter drive device 32, a gate drive device 34, an ejector drive device 36, a weight measuring device 38, a speed measuring device 40, a timing device 42, a cylinder stroke detection device 44, a communication device 46, and a storage device 48.

[0021] The control device 30 comprehensively controls the operation of each part of the scraper vehicle 20. The weight measuring device 38 is a sensor such as a strain gauge that measures the weight of the excavated material in the bowl 24. The speed measuring device 40 is a sensor such as a GNSS (Global Navigation Satellite System) sensor or a sensor that detects the rotation speed of the axle 26, and measures the speed of the scraper vehicle 20. The timing device 42 measures the time from when the control device 30 gives an instruction to start timing until when it gives an instruction to end timing. The cylinder stroke detection device 44 is a sensor that detects the stroke of the hydraulic cylinder provided on the scraper vehicle 20. The communication device 46 exchanges information with the communication device 50 of the towing vehicle 1. The storage device 48 is a non-volatile memory, etc., and stores and manages time change data of the loading amount acquired by the control device 30, as well as information on the optimal loading amount.

[0022] As shown in Figure 2, the towing vehicle 1 is equipped with a communication device 50, a display device 52, an operating device 54, a drive device 56, and a rotation speed detection device 58.

[0023] The communication device 50 exchanges information with the communication device 46 of the scraper vehicle 20. The display device 52 displays the information transmitted from the scraper vehicle 20 via the communication devices 46 and 50. The scraper vehicle 20 transmits instruction information to the operator driving the towing vehicle 1 (such as how much to tilt the bowl 24, how much to press the accelerator, the optimal excavation amount, and the optimal excavation time).

[0024] The operating device 54 includes a handle operated by the operator, an accelerator, and buttons for operating the scraper vehicle 20. Operating information for the scraper vehicle 20 is notified to the control device 30 of the scraper vehicle 20 via communication devices 50 and 46.

[0025] The drive unit 56 drives various parts of the towing vehicle 1 in response to the operator's steering wheel and accelerator pedal operations. The rotation speed detection device 58 is a sensor that detects the rotation speed of the axle on which the wheels of the towing vehicle 1 are attached.

[0026] In this embodiment, the scraper vehicle 20 is towed by the towing vehicle 1 and moves along the path shown in Figure 3 to perform one cycle of processing (operation).

[0027] In the "movement" region of Figure 3, the scraper vehicle 20, being towed by the towing vehicle 1, is unloaded, and the tip of the scraper 25 is not in contact with the ground, meaning it is not scraping the ground (the scraper 25 is retracted). The movement region in Figure 3 can also be considered a transport region where the scraper vehicle 20 is moved, and the movement process of the scraper vehicle 20 moving through the movement region can also be considered a transport process.

[0028] In the "excavation" area shown in Figure 3, the scraper vehicle 20 moves while being towed by the towing vehicle 1, leveling the running surface by digging the tip of the scraper 25 into the ground and excavating it. In this excavation area, the scraper vehicle 20 moves while collecting the excavated soil and other materials into the bowl 24. It can be said that the scraper vehicle 20 is performing the excavation process in the excavation area.

[0029] In the "transportation" area of ​​Figure 3, the bowl 24 of the scraper vehicle 20, which is being towed by the towing vehicle 1, contains excavated material such as soil and sand. The scraper vehicle 20 transports the excavated material to the unloading point (unloading area (soil and sand collection area)). In this transport area, the tip of the scraper 25 does not come into contact with the ground and does not scrape the ground (scraper 25 is retracted). In this transport area, the scraper vehicle 20 can be said to be performing the transport process.

[0030] In the "unwinding" section of Figure 3, the scraper vehicle 20 lowers the excavated material from the bowl 24 and unwinds (spreads out) the lowered excavated material. In addition, in the unwinding section of Figure 3, the scraper vehicle 20 is performing the unwinding process, and the unwinding process can also be described as a discharge process in which the excavated material is discharged from the scraper vehicle 20.

[0031] Here, the excavation process → transportation process → unloading process → movement process constitutes one cycle of processing (first process), and the time required for one cycle of processing is called the cycle time. When the bowl 24 is filled to capacity during the excavation process (when the loading amount is set to S), the time change data of the loading amount in the bowl 24 is as shown in Figure 4. As shown in Figure 4, in the initial stage of the excavation process (the start of excavation), the bowl 24 is nearly empty, so the loading amount per unit time is large, and loading is carried out efficiently. In contrast, in the latter half of the excavation process, the loading amount (loading capacity) in the bowl 24 increases, so the soil pressure inside the bowl 24 increases, making loading more difficult. For this reason, the loading amount per unit time decreases. Note that in the transportation process following the excavation process, the loading amount does not change, and in the unloading process, the loading amount decreases linearly, for example, to 0.

[0032] As described above, the amount of material loaded per unit time decreases in the later stages of the excavation process, so it may be more efficient not to load the excavated material into bowl 24 until it is fully loaded (load capacity = S). Figure 5(a) is a diagram illustrating how to determine the most efficient load capacity (optimal load capacity) S' and the time (optimal excavation time) T until the optimal load capacity S' is loaded, in the example shown in Figure 4.

[0033] As shown in Figure 5(a), from Figure 4, the time a (time required for the transport, unwinding, and moving processes) is identified from the cycle time excluding the time required for the excavation process, and point a is set by time a from the start time of the excavation process. Then, the amount of material loaded at the point of tangency when a tangent line AS is drawn from point a to the time change data of the loading amount during the excavation process (called the loading curve) is defined as the optimal loading amount S'. The time required for the excavated material of the optimal loading amount S' to be loaded into the bowl 24 is defined as the optimal excavation time T. When the optimal loading amount S' and optimal excavation time T are set in this way, the time change data of the loading amount will be as shown in Figure 5(b).

[0034] Here, the ratio of the loading amount to the cycle time (slope β) in Figure 5(b) is greater than the ratio of the loading amount to the cycle time (slope α) in Figure 4 (α < β), indicating that the processing in Figure 5(b) is efficient. Note that even if the loading curve for the excavation process is the same, the optimal loading amount S' and optimal excavation time T will change if the time required for the transport process, unloading process, and moving process changes. In other words, in this embodiment, the time change data of the loading amount obtained in the excavation process of one cycle (first data) and the time required for the transport process, unloading process, and moving process (second data) are used to set the loading amount (excavation amount) and excavation time (time required for one excavation process) so that the time change of the loading amount is small (the loading amount per unit time does not change much while the excavation process is being carried out, i.e., the loading speed does not change much while the excavation process is being carried out). Note that the loading curve in the excavation process is thought to change depending on the site conditions (soil type, etc.). As mentioned above, as excavation progresses, the soil pressure in bowl 24 increases, reducing the amount of material that can be loaded into bowl 24. Therefore, in order to efficiently carry out the excavation process, at least one of the following should be done based on the time-varying data (first data) of the amount of material loaded during the excavation process: shorten the excavation time or reduce the amount of material excavated (the amount excavated in one excavation process). Alternatively, the excavation time should be set shorter or the amount of material excavated (the amount excavated in one excavation process) should be reduced so that the amount of material loaded into bowl 24 per unit time does not fall below a predetermined amount. For example, the excavation process should be stopped when the slope of the tangent to the loading curve during the excavation process falls below a predetermined value or before it falls below a predetermined value.

[0035] (Regarding the processing of the control device 30 of the scraper vehicle 20) Next, the processing of the control device 30 will be explained in detail according to the flowcharts in Figures 6 and 7, with reference to other drawings. In the processing shown in Figure 6, the first round of work is performed when the towing vehicle 1 enters the excavation site, based on the detection results of the GNSS (not shown) of the towing vehicle 1.

[0036] When the process shown in Figure 6 begins, in step S10, the control device 30 first determines whether it is the first cycle or not. If the determination in step S10 is affirmative, the process proceeds to step S12.

[0037] When the process moves to step S12, the control device 30 waits until the towing vehicle 1 reaches a specific location (the excavation site in the first lap). The control device 30 may also use the entry of the towing vehicle 1 into the excavation site as a trigger for the decision in step S12 from the second lap onward, or alternatively, it may use the start of movement from the unwinding site as a trigger for the decision in step S12. Once the towing vehicle 1 reaches the specific location (the excavation site in the first lap), the process moves to step S14, and the control device 30 begins measuring the cycle time.

[0038] Next, in step S16, the control device 30 starts measuring the loading amount using the weight measuring device 38. Then, in step S18, the control device 30 waits until unloading is complete. Specifically, the control device 30 waits until the value of the weight measuring device 38 changes from 0 to S (full load) and then back to 0. Since unloading is considered complete when the value of the weight measuring device 38 returns to 0, the control device 30 proceeds to step S20 at this stage.

[0039] When the process moves to step S20, the control device 30 terminates the measurement of the cycle time. Next, in step S22, the control device 30 stores the time-varying data of the loading amount in the storage device 48. Here, in the first cycle, the processes are carried out in the order of moving process → excavation process → transport process → unloading process, but for convenience, the time-varying data of the loading amount is shown in the order of excavation process → transport process → unloading process → moving process, as shown in Figure 4.

[0040] Next, in step S24, the control device 30 calculates the optimal loading amount S' and the optimal excavation time T. For example, if time-varying data of the loading amount as shown in Figure 4 is obtained, the optimal loading amount S' and the optimal excavation time T are determined through the process shown in Figure 5(a) and as shown in Figure 5(b). After that, the process returns to step S10.

[0041] When the system proceeds to step S10, the control device 30 determines again whether it is the first lap or not. If the determination here is negative, the control device 30 proceeds to step S26.

[0042] When the process moves to step S26, the control device 30 determines whether it is the 2nd to nth cycle. Note that n is a predetermined value (2 or more). If the determination in step S26 is affirmative, the process moves to step S28.

[0043] When the process moves to step S28, the control device 30 sets the optimal loading amount S' and optimal excavation time T determined in the previous cycle (step S24) as loading targets, and displays the optimal loading amount S' and optimal excavation time T on the display device 52 via the communication devices 46 and 50. In the second cycle, the worker refers to the optimal loading amount S' and optimal excavation time T displayed on the display device 52 and performs the excavation process so that the amount of excavated material loaded into the bowl 24 is S'. In this case, the control device 30 also displays the measured value of the weight measuring device 38 and the elapsed time since the excavation process was performed on the display device 52 in real time. This makes it possible for the worker to easily perform an appropriate (efficient) excavation process.

[0044] After that, the processes in steps S12 to S24 are executed in the same manner as described above. By going through each process in the second cycle, it is possible to obtain time-varying data of the loading amount in the second cycle. Therefore, in step S24, the optimal loading amount S' and optimal excavation time T are newly calculated based on the time-varying data of the loading amount obtained in the second cycle. As a result, up to the nth cycle, the excavation process can be carried out with the optimal loading amount S' and optimal excavation time T calculated in the previous cycle as the loading target. Note that in step S12 from the second cycle onward, the judgment in step S12 may be affirmed when the value (loading amount) of the weight measuring device 38 becomes 0, or the judgment in step S12 may be affirmed when it is confirmed from the detection value of the cylinder stroke detection device 44 that the unwinding process has been completed, or when a drive instruction is issued from the operating device 54 to the cutter drive device 32.

[0045] After the above process is repeatedly executed, if the judgment in step S26 is rejected (from the n+1th cycle onward), the control device 30 proceeds to step S30. Upon proceeding to step S30, the control device 30 executes a cycle processing (Figure 7) that utilizes the accumulated data.

[0046] In the process shown in Figure 7, first, in step S40, the control device 30 waits until the scraper vehicle 20 begins to move from the unwinding point. In this case, the decision in step S40 may be affirmed when the value (loading amount) of the weight measuring device 38 becomes 0, or when it is confirmed from the detection value of the cylinder stroke detection device 44 that the unwinding process has been completed, or when a drive instruction is issued from the operating device 54 to the cutter drive device 32. If the decision in step S40 is affirmed, the process proceeds to step S42.

[0047] When the process moves to step S42, the control device 30 starts measuring the cycle time using the timing device 42. Then, in step S44, the control device 30 starts measuring the amount loaded into the bowl 24 using the weight measuring device 38.

[0048] Next, in step S46, the control device 30 waits until a predetermined time elapses after the loading amount starts to change (becomes greater than 0). The predetermined time is, for example, about 1 / 2 to 1 / 3 of the time required for the excavation process. When the predetermined time elapses, the control device 30 proceeds to step S48. In the next step S48, the control device 30 is to predict the time change data of the loading amount after the predetermined time using the time change data of the loading amount obtained at the predetermined time as shown in FIG. 8(a).

[0049] When proceeding to step S48, the control device 30 predicts the time change data of the loading amount after the predetermined time based on the time change data of the loading amount obtained within the predetermined time (FIG. 8(a)) and the time change data of the past loading amount.

[0050] Here, any one of the following three methods is used.

[0051] (1) When the time change data of the loading amount has not been collected much under the same conditions as the current week (when the number of data < M (M is a predetermined natural number)) In this case, the control device 30 uses the time change data of the loading amount obtained in the immediately previous week to predict the time change of the loading amount hereafter. For example, it is assumed that the time change data of the loading amount as shown in FIG. 8(b) was obtained in the immediately previous week and stored in the storage device 48. In this case, by fitting the data of FIG. 8(b) to the data of FIG. 8(a), the time change data of the loading amount as shown by the broken line in FIG. 8(c) is predicted. When multiple time change data of the loading amount are obtained under the same conditions as the current week, it is also possible to obtain the average of the multiple time change data of the loading amount and use the average to predict the time change data of the loading amount after the predetermined time.

[0052] (2) When the time change data of the loading amount of a predetermined number or more has been collected under the same conditions (when the number of data ≧ M) In this case, the control device 30 selects data from the time-varying data of loading volume obtained in the past under the same conditions that is similar to the time-varying data of loading volume obtained within a predetermined time (Figure 8(a)). The control device 30 also makes a prediction by fitting the selected data to the time-varying data of loading volume obtained within a predetermined time (Figure 8(a)). For example, suppose that data like the one shown in Figure 9(a) is stored in the storage device 48 as time-varying data of loading volume obtained in the past under the same conditions. Then, when checking the similarity with the time-varying data of loading volume obtained within a predetermined time (Figure 8(a)), suppose that the third data from the left has the highest similarity, as shown in Figure 9(b). In this case, the control device 30 predicts the time-varying data of loading volume shown by the dashed line in Figure 9(c) by fitting the data with the highest similarity to the data in Figure 8(a).

[0053] (3) When a large amount of time-varying data on loading volume can be collected, including under different conditions. In this case, machine learning is used to predict the time-dependent changes in loading volume. Figure 10 schematically shows an example of a method for predicting time-varying data of loading volume using machine learning. In Figure 10, a learning phase and an inference phase are performed. In the learning phase, a large amount of training data is prepared, consisting of combinations of past time-varying data of loading volume stored in the memory device 48 and excavation conditions such as the travel speed of the scraper vehicle 20 and site information (type of soil, etc.) at the time the data was obtained. A trained model is generated by machine learning on this large amount of training data. This trained model is a model for predicting time-varying data of loading volume after a predetermined time, based on the excavation conditions (travel speed and site information) and time-varying data of loading volume obtained within a predetermined time. On the other hand, in the inference phase, when a combination of time-varying data of loading volume obtained within a predetermined time and excavation conditions (travel speed and site information) is input to the control device 30, the control device 30 uses the trained model to predict time-varying data of loading volume (output data) after a predetermined time.

[0054] After predicting the time-dependent change data of the loading volume after a predetermined time using one of the three methods described above, the process proceeds to step S50.

[0055] When the system moves to step S50, the control device 30 determines whether it is necessary to control the towing vehicle 1, the scraper vehicle 20, or other devices based on the prediction results (time-varying data of the loading amount). For example, the time-varying data of the predicted loading amount may not allow for the desired loading amount. In such cases, the control device 30 determines that it is necessary to either dig the scraper 25 deeper into the ground, or conversely, reduce the amount it digs into the ground, or press the accelerator more, or prepare a device (pusher) to push the scraper vehicle 20 from behind near the excavation site in order to obtain the desired loading amount.

[0056] If the judgment in step S50 is affirmed, the control device 30 proceeds to step S52. Upon proceeding to step S52, the control device 30 outputs control information. For example, it may display a message on the display device 52 to notify the operator that the above control is necessary, or, if the control device 30 can automatically control each device, it may control each device. If it is determined that control is necessary, the control device 30 takes the control content into consideration and predicts the time change data of the loading amount after the control has been performed. After that, it proceeds to step S54. If the judgment in step S50 is denied (control is not necessary), it proceeds to step S54 without going through step S52.

[0057] When the process moves to step S54, the control device 30 calculates the optimal loading amount S' and optimal excavation time T based on the predicted time-varying data of the loading amount and sets them as loading targets. Next, in step S56, the control device 30 waits until unwinding is complete. Specifically, the control device 30 waits until the value of the weight measuring device 38 goes from 0 to a larger value and then returns to 0. When the value of the weight measuring device 38 returns to 0, it is considered that unwinding is complete, and the process moves to step S58. Alternatively, the control device 30 may determine the completion of unwinding based on the detection result of the cylinder stroke detection device 44.

[0058] When the system moves to step S58, the control device 30 terminates the measurement of the cycle time. Then, in step S60, the control device 30 stores the time-varying data of the loading amount in the storage device 48.

[0059] As a result, the process in Figure 7 is completed, and step S30 in Figure 6 is finished. After that, step S30 is executed repeatedly. Note that the process in Figure 6 is terminated when the operator inputs the end of the work on the operating device 54.

[0060] As described in detail above, according to this embodiment, when the scraper vehicle 20 performs one cycle of processing (movement process, excavation process, transport process, unloading process), the control device 30 acquires time change data of the amount of excavated material loaded onto the scraper vehicle 20 during the excavation process of that cycle (loading curve in Figure 5(a)) and the time a required for processes other than the excavation process (movement process, transport process, unloading process). The control device 30 then uses the acquired loading curve and time a to set the optimal loading amount S' and the optimal excavation time T. As a result, in this embodiment, excavation can be performed efficiently.

[0061] Furthermore, in this embodiment, the time-varying data of the loading volume for one cycle of processing performed at the same site after one cycle of processing is predicted based on multiple time-varying data of loading volumes stored in the storage device 48. This makes it possible to accurately predict the time-varying data of the loading volume based on time-varying data of loading volumes obtained under the same conditions in the past.

[0062] Furthermore, in this embodiment, data similar to the time-varying data of loading amounts obtained within a predetermined time is identified from among the multiple time-varying data of loading amounts stored in the storage device 48, and the time-varying data of loading amounts after the predetermined time is predicted based on the identified data. This makes it possible to accurately predict the time-varying data of loading amounts after the predetermined time.

[0063] Furthermore, in this embodiment, the towing vehicle 1, scraper vehicle 20, and other devices (such as pushers) are controlled based on the time change of the loading amount after a predetermined time has been predicted. This allows each device to be controlled so that appropriate time change data for the loading amount is obtained (so that the desired loading amount is achieved).

[0064] Furthermore, in this embodiment, by inputting excavation conditions and other data into the trained model shown in Figure 10, the time-varying data of the loading volume after a predetermined time is predicted. This makes it possible to accurately predict the time-varying data of the loading volume after a predetermined time.

[0065] In the above embodiment, the processing cycle may be executed multiple times to obtain multiple time-varying data on the amount of excavated material loaded and the time a required for processes other than the excavation process. These can then be used to set the optimal loading amount S' and optimal excavation time T so that the processing cycle is performed efficiently. This allows for setting appropriate values ​​for the optimal loading amount S' and optimal excavation time T.

[0066] In the above embodiment, time-varying data of the loading volume obtained within a predetermined time and excavation conditions (travel speed and soil information) were input to the trained model shown in Figure 10 to predict the time-varying data of the loading volume obtained after a predetermined time. However, the embodiment is not limited to this. For example, it is not necessary to input the time-varying data of the loading volume obtained within a predetermined time, and the time-varying data of the loading volume can be predicted by inputting only the excavation conditions.

[0067] In the above embodiment, the trained model shown in Figure 10 may be generated on an external server. This reduces the processing load on the control device 30. Furthermore, performing the training phase on an external server makes it easier to collect training data from a large number of scraper vehicles.

[0068] In the above embodiment, the control device 30 may use only the time-varying data of the amount of excavated material loaded onto the scraper vehicle 20 during the excavation process, and set the loading amount (excavation amount) and excavation time (time required for one excavation process) so that the time-varying change of the loading amount is small (so that the loading amount per unit time does not change much while the excavation process is being carried out, i.e., so that the loading speed does not change much while the excavation process is being carried out). In this case, it is sufficient to shorten the excavation time and reduce the excavation amount (the amount excavated in one excavation process). Alternatively, the excavation time may be shortened or the excavation amount reduced so that the amount loaded into the bowl 24 per unit time does not fall below a predetermined amount. For example, the excavation process may be stopped when the slope of the tangent to the loading curve in the excavation process falls below a predetermined value or before it falls below a predetermined value.

[0069] In the above embodiment, the case in which the towing vehicle 1 is driven by a worker has been described, but the towing vehicle 1 is not limited to this, and may be a vehicle capable of automatic driving or remote control. If the towing vehicle 1 is capable of automatic driving, instead of giving driving and control instructions to the worker, instructions for automatic driving should be output to the towing vehicle 1.

[0070] The embodiments described above are preferred examples of the present invention. However, the invention is not limited thereto, and various modifications are possible without departing from the spirit of the invention. [Explanation of symbols]

[0071] 1. Towing vehicle 20. Scraper vehicles (construction machinery) 30 Control device 32 Cutter drive unit 34 Gate drive unit 36 Ejector drive unit 38. Weight measuring device 40 Speed ​​measuring device 42 Timing device 44 Cylinder Stroke Detection Device 46 Communication equipment 48 Storage device

Claims

1. A first process is performed which includes an excavation step in which excavation is carried out while moving construction machinery, and first data is obtained which shows the change in the amount of soil loaded onto the construction machinery in the excavation step of the first process over time. An excavation method in which a computer performs a process to set at least one of the excavation time and excavation amount based on the time change of the amount of soil loaded into the construction machine, using the acquired first data.

2. The excavation method according to claim 1, wherein the setting process involves setting the excavation time and the amount of excavation so as to shorten the time required for the excavation process.

3. The drilling method according to claim 1, wherein the drilling time and the amount of drilling are set in the setting process such that the amount of drilling is reduced.

4. The excavation method according to claim 1, wherein the process of setting the above-mentioned amount of soil loaded into the construction machine is set such that the time change in the amount of soil loaded into the construction machine is small.

5. The first process is executed in one cycle, which includes the excavation step, the movement step of moving the construction machine from the excavation site to the unwinding site from which the excavated material is unwound, the discharge step of having the construction machine discharge the excavated material at the unwinding site, and the return step of moving the construction machine from the unwinding site back to the excavation site, and second data indicating the time taken for the first process other than the excavation step is obtained, The excavation method according to claim 1, wherein the setting process involves setting at least one of the excavation time and the amount of excavation based on the time change in the amount of soil loaded into the construction machine, using the first data and the second data.

6. In the acquisition process described above, the first process is executed multiple times to acquire multiple copies of the first data and the second data. The drilling method according to claim 5, wherein the setting process uses a plurality of first data and a plurality of second data to set the drilling time and the drilling amount so that the time required for the drilling process is shortened.

7. The excavation method according to claim 5, wherein the time change of the amount of soil loaded onto the construction machine in the excavation step of the second process of the cycle, which is performed at the same site as the first process after the first process, is predicted based on a plurality of acquired first data.

8. The excavation method according to claim 7, wherein in the prediction process, a first data set similar to the time-varying data obtained after the start of the excavation process of the second process is extracted from a plurality of first data sets, and the time-varying data of the amount of soil to be loaded into the construction machine in the excavation process of the second process is predicted based on the similar first data set.

9. The excavation method according to claim 7 or 8, wherein the predicted results are used for controlling the construction machine or a device that assists the operation of the construction machine.

10. The excavation method according to claim 5, wherein the excavation conditions in the excavation process of the second process are input into a trained model, which has been trained using multiple combinations of excavation conditions in the excavation process of the first process of the first cycle and the first data as training data, in order to predict the time change in the amount of soil loaded into the construction machine in the excavation process of the second process.

Images