Muck transportation control method, device and equipment for heading machine, medium and program product

Abstract

The embodiment of the invention provides a muck transportation control method, device and equipment for a heading machine, a medium and a program product, which are applied to a muck transportation control system, and are characterized in that video data of a muck discharge belt regulation and control system is acquired through image acquisition equipment, and the video data is divided into training set data, verification set data and test set data; the method comprises the following steps: generating a training set thermodynamic diagram, a verification set thermodynamic diagram and a test set thermodynamic diagram by using an image labeling tool and transfer function labeling, acquiring a carriage slag receiving image of a slag transporting vehicle by using image acquisition equipment, and training a pre-created image processing model according to the training set thermodynamic diagram, the verification set thermodynamic diagram and the carriage slag receiving image; the precision of the trained model is verified according to the test set thermodynamic diagram, if the trained model meets the preset precision, the trained model is applied to a deslagging belt regulation and control system, an image acquisition device acquires a work site image in real time, and the deslagging belt regulation and control system configuring the model identifies the work site image to control the operation of the deslagging belt. A worker does not need to monitor the slag conveying process, and the slag conveying efficiency is improved.

Description

Technical Field

[0001] The present application relates to the technical field of tunnel construction, and particularly relates to a control method, device, equipment, medium and program product for muck transportation of a tunneling machine. Background Art

[0002] An earth pressure balance shield machine is a device used for tunnel construction. During the construction process, the muck discharging system of the earth pressure balance shield machine discharges the muck cut by the cutter head to the belt conveyor through a screw conveyor, and then the belt conveyor transports it to the muck truck.

[0003] In the prior art, during the tunneling process of an earth pressure balance shield machine, most rely on muck transport trolleys to transport the muck on the belt to outside the tunnel, and the muck loading process completely depends on manual monitoring and regulation. Specifically, the staff observes the states of the material dropping opening and the muck transport trolley in real time through the camera video. When it is observed that there is a muck truck under the material dropping opening and the carriage is not full, the belt is started, and the muck drops into the carriage. When the carriage is full of muck, the staff reminds the muck truck driver through a walkie-talkie, controls the muck truck to move forward, and uses the next carriage to receive the material.

[0004] However, the solution of the prior art leads to the need to equip personnel to observe and remind through dialogue throughout the muck transport process. The long-term muck discharging and transport greatly occupy the time of the staff, resulting in a reduction in the muck transport efficiency. Summary of the Invention

[0005] Embodiments of the present application provide a control method, device, equipment, medium and program product for muck transportation of a tunneling machine, so as to solve the problem of reduced muck transport efficiency existing in the prior art.

[0006] In a first aspect, an embodiment of the present application provides a control method for muck transportation of a tunneling machine, which is applied to a muck transportation control system. The muck transportation control system includes an image acquisition device and a muck discharging belt regulation system; the method includes:

[0007] Obtaining video data of the muck discharging belt regulation system through the image acquisition device;

[0008] Dividing the video data into training set data, validation set data and test set data;

[0009] Annotating the boundary boxes and target categories of the training set data, the validation set data and the test set data through an image annotation tool to obtain the annotated training set data, the annotated validation set data and the annotated test set data;

[0010] Converting the annotated training set data, the annotated validation set data and the annotated test set data into corresponding training set heatmaps, validation set heatmaps and test set heatmaps through a conversion function;

[0011] Obtain the image data of the slag transport vehicle through the image acquisition device, and generate the image of the carriage slag receiving of the slag transport vehicle according to the image data of the slag transport vehicle;

[0012] Train the pre-created image processing model according to the training set heat map, the validation set heat map and the carriage slag receiving image to generate the trained image processing model;

[0013] Verify the model accuracy of the trained image processing model according to the test set heat map;

[0014] If the trained image processing model meets the preset model accuracy, configure the trained image processing model in the slag discharge belt control system to obtain the slag discharge belt control system configured with the image processing model;

[0015] Obtain the image information of the work site through the image acquisition device, and input the image information of the work site into the slag discharge belt control system configured with the image processing model;

[0016] Control the operation of the slag discharge belt through the slag discharge belt control system configured with the image processing model according to the image information of the work site.

[0017] In a possible implementation manner, the training of the pre-created image processing model according to the training set heat map, the validation set heat map and the carriage slag receiving image to generate the trained image processing model includes: performing image preprocessing on the training set heat map, the validation set heat map and the carriage slag receiving image to generate the processed training set heat map, the processed validation set heat map and the processed carriage slag receiving image; training the pre-created image processing model according to the processed training set heat map and the processed carriage slag receiving image according to the set number of iterations to obtain multiple sets of model weights of the trained image processing model; verifying the model weights of the multiple sets of trained image processing models according to the processed validation set heat map to obtain the optimal weights; generating the trained image processing model according to the optimal weights.

[0018] In a possible implementation manner, the pre-created image processing model includes: a spatial module, a context module and a detection head module; the loss function of the pre-created image processing model is:

[0019] Loss=λ 1 L k +λ 2 L size +λ 3 L off +λ 4 L cls

[0020] In the formula, L k represents the key point loss; L size represents the bounding box loss; L off represents the center point offset loss; L cls represents the target class classification loss; λ 1 represents the key point loss weight; λ 2 represents the bounding box loss weight; λ 3 represents the center point offset loss weight; λ 4 represents the target class classification loss weight; λ 1 +λ 2 +λ 3 +λ 4 = 1.

[0021] In a possible implementation manner, the method of obtaining the slag truck image data by the image acquisition device and generating the carriage slag receiving image of the slag truck according to the slag truck image data includes: obtaining first image data when the front baffle of the carriage of the slag truck is aligned with the material dropping port by the image acquisition device; annotating the bounding box of the first image data by an image annotation tool to obtain first image position information; obtaining second image data when the rear baffle of the carriage of the slag truck is aligned with the material dropping port by the image acquisition device; annotating the bounding box of the second image data by an image annotation tool to obtain second image position information; and generating the carriage slag receiving image of the slag truck according to the first image position information and the second image position information.

[0022] In a possible implementation manner, the method of controlling the operation of the slag discharging belt by the slag discharging belt control system configured with the image processing model according to the image information of the work site includes: identifying the position information of the carriage of the slag truck according to the image information of the work site by the image processing model; if the position information of the carriage of the slag truck is within the data range of the carriage slag receiving image, determining whether the slag in the carriage of the slag truck is fully loaded; and if the slag in the carriage of the slag truck is not fully loaded, controlling the operation of the slag discharging belt by the slag discharging belt control system configured with the image processing model.

[0023] In a possible implementation, if the muck in the muck truck carriage is not full, after controlling the operation of the slag discharge belt through the slag discharge belt control system configured with the image processing model, it further includes: if the muck in the muck truck carriage is in a full load state, send a warning message to the muck truck to make the muck truck move forward a distance of one carriage, and record the number of full load carriages; if the muck in all carriages of the muck truck is in a full load state, control the slag discharge belt to stop running through the slag discharge belt control system, and send a departure instruction to the muck truck; if the position information of the muck truck carriage is not within the data range of the carriage slag receiving image, send the collected image information to the industrial control computer, and the industrial control computer sends a stop slag discharge instruction to the slag discharge belt control system to make the slag discharge belt control system control the slag discharge belt to stop running.

[0024] In a second aspect, an embodiment of the present application provides a tunneling machine muck transportation control device, which is applied to a muck transportation control system. The muck transportation control system includes an image acquisition device and a slag discharge belt control system; the device includes:

[0025] A first acquisition module, configured to acquire video data of the slag discharge belt control system through the image acquisition device;

[0026] A division module, configured to divide the video data into training set data, validation set data, and test set data;

[0027] A labeling module, configured to label the bounding boxes and target categories of the training set data, the validation set data, and the test set data through an image labeling tool to obtain labeled training set data, labeled validation set data, and labeled test set data;

[0028] A conversion module, configured to convert the labeled training set data, the labeled validation set data, and the labeled test set data into corresponding training set heatmaps, validation set heatmaps, and test set heatmaps through a conversion function;

[0029] A second acquisition module, configured to acquire muck truck image data through the image acquisition device and generate a carriage slag receiving image of the muck truck according to the muck truck image data;

[0030] A training module, configured to train a pre-created image processing model according to the training set heatmap, the validation set heatmap, and the carriage slag receiving image to generate a trained image processing model;

[0031] A verification module, configured to verify the model accuracy of the trained image processing model according to the test set heatmap;

[0032] A configuration module, configured to, if the trained image processing model meets a preset model accuracy, configure the trained image processing model in the slag discharge belt control system to obtain a slag discharge belt control system configured with the image processing model;

[0033] A third acquisition module, configured to acquire image information of a work site through the image acquisition device, and input the image information of the work site into the slag discharge belt control system configured with the image processing model;

[0034] A control module, configured to control the operation of the slag discharge belt according to the image information of the work site through the slag discharge belt control system configured with the image processing model.

[0035] In a third aspect, an embodiment of the present application provides a tunneling machine slag transportation control device, including:

[0036] At least one processor and a memory;

[0037] The memory stores computer-executable instructions;

[0038] The at least one processor executes the computer-executable instructions stored in the memory, so that the at least one processor executes the first aspect and / or various possible implementation manners of the first aspect as above.

[0039] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which computer-executable instructions are stored, and when the computer-executable instructions are executed by a processor, they are used to implement the first aspect and / or various possible implementation manners of the first aspect as above.

[0040] In a fifth aspect, an embodiment of the present application provides a computer program product, including a computer program, and when the computer program is executed by a processor, it implements the first aspect and / or various possible implementation manners of the first aspect as above.

[0041] The tunneling machine muck transportation control method, device, equipment, medium and program product provided by the embodiments of the present application obtain the video data of the muck belt regulation system through an image acquisition device, divide the video data into training set data, validation set data and test set data, use an image annotation tool and a conversion function to annotate and generate a training set heat map, a validation set heat map and a test set heat map, collect the carriage slag receiving image of the slag transport vehicle through the image acquisition device, train a pre-created image processing model according to the training set heat map, the validation set heat map and the carriage slag receiving image, verify the accuracy of the trained model according to the test set heat map, if the trained model meets the preset accuracy, apply the trained model to the muck belt regulation system, collect the working site image in real time through the image acquisition device, identify the working site image through the muck belt regulation system configured with the model, and control the operation of the muck belt. Compared with the prior art, it does not require staff to monitor the muck transportation process, improving the muck transportation efficiency. Description of the Drawings

[0042] The drawings herein are incorporated into the specification and form a part of this specification, showing embodiments consistent with the present application, and are used together with the specification to explain the principles of the present application.

[0043] Figure 1 It is a schematic diagram of the application scenario of the tunneling machine muck transportation control method provided by the embodiments of the present application;

[0044] Figure 2 It is a schematic flowchart of the tunneling machine muck transportation control method provided by the present application;

[0045] Figure 3 It is a schematic diagram of the spatial module structure provided by the embodiments of the present application;

[0046] Figure 4 It is a schematic diagram of the structure of the context module stage1 provided by the embodiments of the present application;

[0047] Figure 5 It is a schematic diagram of the structure of the NoUpSampling-GE layer provided by the embodiments of the present application;

[0048] Figure 6 It is a schematic diagram of the structure of the UpSampling-GE layer provided by the embodiments of the present application;

[0049] Figure 7 It is a schematic diagram of the structure of the tunneling machine muck transportation control device provided by the present application;

[0050] Figure 8 It is a schematic diagram of the structure of the tunneling machine muck transportation control equipment provided by the present application.

[0051] Through the above-mentioned drawings, specific embodiments of the present application have been shown, and more detailed descriptions will be provided hereinafter. These drawings and textual descriptions are not intended to limit the scope of the concept of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. Detailed Description of the Embodiments

[0052] Exemplary embodiments will be described in detail herein, and examples thereof are shown in the drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0053] The earth pressure balance shield machine is a device used for tunnel construction. During the construction process, the muck discharging system of the earth pressure balance shield machine discharges the muck cut by the cutter head to the belt conveyor through a screw conveyor, and then transports it to the muck truck by the belt conveyor. In the prior art, during the tunneling process of the earth pressure balance shield machine, most of the muck on the belt is transported to the outside of the tunnel by a muck transport trolley, and the muck loading process completely relies on manual monitoring and regulation. Specifically, the staff observes the states of the feeding port and the muck transport trolley in real time through the camera video. When it is observed that there is a muck truck under the feeding port and the carriage is not full, the belt is started, and the muck falls into the carriage. When the carriage is full of muck, the staff reminds the muck truck driver through a walkie-talkie, controls the muck truck to move forward, and uses the next carriage to receive the material. However, the solution of the prior art requires personnel to observe and remind through dialogue throughout the muck transportation process. The long-term muck discharging and transportation greatly occupy the time of the staff, resulting in a reduction in the muck transportation efficiency.

[0054] To solve the above technical problems, the embodiments of the present application propose the following technical concepts: The inventor considered obtaining the video data of the slag discharge belt control system through an image acquisition device, dividing the video data into a training set, a validation set, and a test set, annotating the bounding boxes and target categories of the training set, validation set, and test set data through an image annotation tool, converting the annotated training set, annotated validation set, and annotated test set into corresponding training set heatmaps, validation set heatmaps, and test set heatmaps through a conversion function, obtaining the image data of the slag transport vehicle through the image acquisition device and generating the carriage slag receiving image of the slag transport vehicle, training a pre-created image processing model using the training set heatmap, validation set heatmap, and carriage slag receiving image, generating a trained image processing model, and verifying the model accuracy of the trained image processing model according to the test set heatmap. If the preset model accuracy is met, the trained model is configured into the slag discharge belt control system, the image information of the work site is obtained through the image acquisition device, and the image information of the work site is input into the slag discharge belt control system configured with the image processing model to control the operation of the slag discharge belt. The following will be described in detail with specific embodiments.

[0055] Figure 1 FIG. is a schematic diagram of the application scenario of the tunneling machine slag transportation control method provided by the embodiments of the present application. As Figure 1 shown, this scenario includes: an image acquisition device 10, a light source 20, an industrial control computer 30, and a slag discharge belt control system 40. The slag discharge belt control system 40 includes: a slag conveying belt 401, a material dropping port 402, and a slag transport vehicle 403.

[0056] As Figure 1 shown, the image acquisition device 10 obtains the video data of the slag discharge belt control system 40. The slag transportation control system divides the video data into training set data, validation set data, and test set data, annotates the bounding boxes and target categories of the training set data, validation set data, and test set data through an image annotation tool to obtain the annotated training set data, validation set data, and test set data, generates a training set heatmap, a validation set heatmap, and a test set heatmap through a conversion function, obtains the image data of the slag transport vehicle 403 through the image acquisition device 10, generates the carriage slag receiving image of the slag transport vehicle 403, trains a pre-created image processing model according to the training set heatmap, validation set heatmap, and carriage slag receiving image to generate a trained image processing model. After the trained model meets the model accuracy, the trained model is configured in the slag discharge belt control system 40. The image information of the work site is obtained through the image acquisition device 10, and the image information of the work site is input into the slag discharge belt control system 40 configured with the image processing model to control the operation of the slag discharge belt.

[0057] Figure 2It is a schematic flowchart of the control method for the muck transportation of the tunneling machine provided by this application. The execution subject of this embodiment can be the muck transportation control system, and no special limitation is made here in this embodiment. As Figure 2 shown, this method includes:

[0058] S201: Obtain the video data of the slag discharge belt regulation system through the image acquisition device.

[0059] In this embodiment, the image acquisition device includes, but is not limited to, a video recorder, a video camera, and an intelligent device equipped with a camera.

[0060] Specifically, the image acquisition device collects the video data of the heavy belt slag dropping area of the slag discharge belt regulation system in real time, performs frame division on the video data, and extracts key frames to obtain the muck truck status monitoring data set E.

[0061] S202: Divide the video data into training set data, validation set data, and test set data.

[0062] In this embodiment, the proportions of the training set data, validation set data, and test set data in the video data are 70%, 15%, and 15% respectively.

[0063] S203: Mark the bounding boxes and target categories of the training set data, validation set data, and test set data through an image annotation tool to obtain the labeled training set data, labeled validation set data, and labeled test set data.

[0064] Specifically, mark the target category name cls, the horizontal pixel coordinate r of the center point of the bounding box, x the vertical pixel coordinate r of the center point of the bounding box, y the width w of the bounding box, and the height h of the bounding box of the training set data, validation set data, and test set data through the image annotation tool.

[0065] S204: Convert the labeled training set data, labeled validation set data, and labeled test set data into corresponding training set heatmaps, validation set heatmaps, and test set heatmaps through a conversion function.

[0066] Specifically, convert the labeled bounding box into a key point heatmap through a Gaussian kernel function. The input image size is W×H×3, and the corresponding key point heatmap label size is where C is the total number of target categories, and the heatmap channel order is the target category number order.

[0067] S205: Obtain the muck truck image data through the image acquisition device, and generate the carriage slag receiving image of the muck truck according to the muck truck image data.

[0068] Specifically, through an image acquisition device, first image data is acquired when the front baffle of the slag transport vehicle is aligned with the blanking port. The bounding box of the first image data is marked, and the center point is determined based on the bounding box to obtain the first image position information. The slag transport vehicle is moved, and second image data is acquired when the rear baffle of the slag transport vehicle is aligned with the blanking port. The bounding box of the second image data is marked, and the center point is determined based on the bounding box to obtain the second image position information. The slag receiving range of the carriage is determined from the first image position information and the second image position information, and a slag receiving image of the carriage of the slag transport vehicle is generated.

[0069] S206: According to the training set heat map, the validation set heat map, and the slag receiving image of the carriage, a pre-created image processing model is trained to generate a trained image processing model.

[0070] Specifically, the training set heat map, the validation set heat map, and the slag receiving image of the carriage are subjected to image preprocessing. According to the set number of iterations, the pre-created image processing model is trained with the preprocessed data to obtain multiple sets of model weights of the trained image processing model, the optimal weight is verified, and a trained image processing model is generated based on the optimal weight.

[0071] In this embodiment, Pytorch is used to build a BiseNet spatial-context dual-path deep learning model to generate a pre-created image processing model.

[0072] Among them, the image processing model includes a spatial module Spatial Module, a context module Context Module, and a detection head module Head Module.

[0073] Figure 3 It is a schematic diagram of the spatial module structure provided by the embodiment of the present application.

[0074] Such as Figure 3As shown in the figure, the spatial module consists of 5 stages. Stage1 is a 2x downsampling residual module. The residual module consists of a main branch composed of 1x1Conv-BN-Relu-3x3Conv-BN-Relu-1x1Conv-BN and a shortcut branch containing 1x1Conv-BN. In the main branch, the convolution stride of 3x3Conv is 2, and the convolution strides of the rest are 1. In the shortcut branch, the convolution stride is 2. The outputs of the main branch and the shortcut branch are fused through the Add operation of channel-wise feature map addition. Stage2 contains two 2x downsampling residual modules. The structure of the residual module is the same as that in Stage1. The input of the first residual module comes from the output of Stage1. The input of the second residual module comes from the output of Stage1 and the fused result of the output of the first residual module in Stage2 after 2x downsampling. The fusion operation is Add. In Stage3, there are three residual modules. The input of the first residual module comes from the output of the first residual module in Stage2. The input of the second residual module is fused by the output of the first residual module in Stage3 after downsampling, the output of the first residual module in Stage2, and the output of the second residual module in Stage2 after upsampling through the Add operation. The upsampling method is implemented by 3x3DeConv transposed convolution. The same applies to Stage4 and Stage5. Among them, the output of the third residual module in Stage5 is the final output of the spatial module. The size of the output feature map is 1 / 8 of the size of the original input image, and the number of channels is 128.

[0075] In this embodiment, the context module consists of 4 stages. Stage1 contains four different parallel branches for downsampling the feature map.

[0076] Figure 4 It is a schematic diagram of the structure of Stage1 of the context module provided by the embodiment of the present application.

[0077] As Figure 4 shown, the first branch consists of 3x3Conv-BN-Relu with a stride of 2 and 1x1Conv-BN-Relu with a stride of 1. The second branch consists of 3x3DConv-BN-Relu with a stride of 2 and a dilation rate s of 2 and 1x1Conv-BN-Relu with a stride of 1. The third branch consists of 3x3DConv-BN-Relu with a stride of 2 and a dilation rate of 3 and 1x1Conv-BN-Relu with a stride of 1. The fourth branch consists of 1x1Conv-BN-Relu with a stride of 1 and 3x3AvgPool-BN-Relu. Finally, the outputs of the four branches are fused through the concatenate feature map stacking operation.

[0078] In this embodiment, stage2 of the context module is composed of a connection of a NoUpSampling-GE layer and an UpSampling-GE layer.

[0079] Figure 5 It is a schematic structural diagram of the NoUpSampling-GE layer provided by the embodiment of the present application.

[0080] As Figure 5 shown, the left branch structure is 1x1Conv-BN-Relu-3x3DConv-BN-Relu-3x3DConv-BN-Relu-Soft Attention Module-Relu, where the stride of the 3x3DConv dilated convolution is 1 and the dilation rate is 2, and the SoftAttention Module is a soft attention module. The right branch is a skip connection, and the output feature map of the left branch and the output feature map of the right branch are fused through a concatenate operation.

[0081] Figure 6 It is a schematic structural diagram of the UpSampling-GE layer provided by the embodiment of the present application.

[0082] As Figure 6 shown, the main branch has three 3x3 dilated convolutions connected in sequence, and the skip connection branch is composed of 1x1Conv-BN-Relu-3x3DConv-BN, where the stride of the 3x3DConv is 2 and the dilation rate is 2.

[0083] In this embodiment, the output of stage4 of the context module is the final output of the context module, and its size is 1 / 32 of the input image size.

[0084] In this embodiment, the four parallel output branches of the detection head module respectively output different prediction quantities. The first branch is composed of 3x3Conv-BN-Relu-1x1Conv-BN-Relu, and the prediction quantity is the heat map of the target center point coordinates. The size of the heat map is where W is the width of the input image, H is the height of the input image, and C is the total number of categories predicted by the model. The second branch is the same as the first branch, and the size of the heat map of the offset of the single output center point is The third branch is the same as the second branch, and the output is the heat map of the width and height of the bounding box. The fourth branch is used to predict the probability size of the target belonging to each category, and is composed of 3x3Conv-BN-Relu-1x1Conv-BN-Relu-FC(1024)-FC(C)-Softmax, and the output size is (1×1×(C + 1)), where C is the total number of categories predicted by the model.

[0085] S207: Verify the model accuracy of the trained image processing model according to the heat map of the test set.

[0086] Specifically, input the heat map of the test set into the trained image processing model to obtain the image processing result output by the trained image processing model. If the error range between the output image processing result and the heat map of the test set is within the set range, the model accuracy of the trained image processing model meets the set requirements.

[0087] S208: If the trained image processing model meets the preset model accuracy, configure the trained image processing model in the slag discharge belt control system to obtain the slag discharge belt control system configured with the image processing model.

[0088] Specifically, configure the trained image processing model on the industrial control computer of the slag discharge belt control system, and control the slag discharge of the slag discharge belt control system through the industrial control computer.

[0089] S209: Obtain the image information of the work site through the image acquisition device, and input the image information of the work site into the slag discharge belt control system configured with the image processing model.

[0090] In this embodiment, deploy the image processing model to the industrial control computer of the slag discharge belt control system, obtain the image information of the work site in real time through the image acquisition device, and the industrial control computer controls the slag discharge of the system according to the image information.

[0091] S210: Control the operation of the slag discharge belt through the slag discharge belt control system configured with the image processing model according to the image information of the work site.

[0092] Specifically, identify the position of the slag truck carriage through the image processing model. If the carriage position is within the data range of the carriage slag receiving image, determine whether the slag in the carriage is full. If it is not full, control the slag discharge through the slag discharge belt control system. When the slag in the carriage is full, give a prompt to the slag truck through the slag discharge belt control system.

[0093] As can be seen from the above embodiments, video data of the slag discharge belt control system is obtained through an image acquisition device, the video data is divided into training set data, validation set data, and test set data, training set heatmaps, validation set heatmaps, and test set heatmaps are generated through annotation using an image annotation tool and a conversion function, the carriage slag receiving image of the slag transport vehicle is acquired through the image acquisition device, a pre-created image processing model is trained based on the training set heatmap, validation set heatmap, and carriage slag receiving image, the accuracy of the trained model is verified according to the test set heatmap, if the trained model meets the preset accuracy, the trained model is applied to the slag discharge belt control system, the working site image is acquired in real time through the image acquisition device, the working site image is recognized by the slag discharge belt control system configured with the model, and the operation of the slag discharge belt is controlled. Compared with the prior art, it is not necessary for staff to monitor the slag transport process, and the slag transport efficiency is improved.

[0094] In an embodiment of the present application, step S206 includes:

[0095] S2061: Perform image preprocessing on the training set heatmap, validation set heatmap, and carriage slag receiving image to generate a processed training set heatmap, a processed validation set heatmap, and a processed carriage slag receiving image.

[0096] Specifically, image enhancement technology is used to preprocess the input training set heatmap, validation set heatmap, and carriage slag receiving image.

[0097] Among them, the image enhancement technology includes, but is not limited to, rotation, scaling, cropping, brightness change, and adding noise.

[0098] In an embodiment of the present application, the loss function of the pre-created image processing model is:

[0099] Loss=λ 1 L k +λ 2 L size +λ 3 L off +λ 4 L cls

[0100] In the formula, L k represents the key point loss; L size represents the bounding box loss; L off represents the center point offset loss; L cls represents the target class classification loss; λ 1 represents the key point loss weight; λ 2 represents the bounding box loss weight; λ 3 represents the center point offset loss weight; λ 4 represents the target class classification loss weight; λ 1 +λ2 +λ 3 +λ 4 = 1。

[0101] S2062: Train the pre-created image processing model according to the processed training set heat map and the processed carriage slag receiving image based on the set number of iterations to obtain the model weights of multiple groups of trained image processing models.

[0102] Specifically, through the Stochastic Gradient Descent (SGD) algorithm, set the momentum to 0.9, the weight decay to 0.0001, combine with transfer learning technology, and use the learning rate optimization method of cosine annealing to iteratively train the model offline until the model converges to obtain the model weights including the optimal weight ω.

[0103] S2063: Verify the model weights of multiple groups of trained image processing models according to the processed validation set heat map to obtain the optimal weight.

[0104] Specifically, verify the data output by the model corresponding to each model weight through the validation set heat map, compare the errors between multiple groups of data and the validation set heat map, obtain the model weight with the minimum error, and get the optimal weight.

[0105] S2064: Generate the trained image processing model according to the optimal weight.

[0106] Specifically, test the model accuracy on the trained image processing model through the test set heat map.

[0107] As can be seen from the above embodiments, by performing image preprocessing on the training set heat map, the validation set heat map, and the carriage slag receiving image, the robustness and generalization of the training model are improved. According to the set number of iterations, train the pre-created image processing model with the processed training set heat map and the processed carriage slag receiving image, obtain the optimal weight after training, and determine the trained image processing model according to the optimal weight, thereby improving the prediction accuracy of the image processing model.

[0108] In an embodiment of the present application, step S205 includes:

[0109] S2051: Obtain the first image data when the front baffle of the carriage of the slag transport vehicle is aligned with the blanking port through an image acquisition device.

[0110] Specifically, when a certain carriage of the slag transport vehicle is parked at the blanking port of the slag discharging belt control system and the front baffle of the carriage is aligned with the blanking port, obtain the first image data P through the image acquisition device. 1 。

[0111] S2052: Mark the bounding box of the first image data through an image annotation tool to obtain the first image position information.

[0112] Specifically, the boundary box of the slag transport vehicle is marked by the image annotation tool Labelme, and the first image position information (x 1 , y 1 ) of the center point of the slag transport vehicle in the image is obtained.

[0113] S2053: Obtain the second image data when the rear baffle of the carriage of the slag transport vehicle is aligned with the blanking port through the image acquisition device.

[0114] Specifically, move the slag transport vehicle to align the rear baffle of the carriage of the slag transport vehicle with the blanking port, and collect the second image data P at this time through the image acquisition device. 2 .

[0115] S2054: Mark the boundary box of the second image data through the image annotation tool to obtain the second image position information.

[0116] Specifically, the boundary box of the slag transport vehicle is marked by the image annotation tool Labelme, and the second image position information (x 2 , y 2 ) of the center point of the slag transport vehicle in the image is obtained.

[0117] S2055: Generate the slag receiving image of the carriage of the slag transport vehicle according to the first image position information and the second image position information.

[0118] In this embodiment, the range of the slag receiving image of the carriage of the slag transport vehicle is the rectangular area between the first image position information and the second image position information.

[0119] As can be seen from the above embodiments, the first image data when the front baffle of the carriage is aligned with the blanking port is marked by the image annotation tool, and the first image position information of the center point of the first image data is determined according to the first image data. The second image data when the rear baffle of the carriage is aligned with the blanking port is marked by the image annotation tool, and the second image position information of the center point of the second image data is determined according to the second image data. The slag receiving range of the carriage is determined according to the first image position information and the second image position information, improving the accuracy of the slag receiving image of the carriage.

[0120] In an embodiment of the present application, step S210 includes:

[0121] S2101: Identify the position information of the carriage of the slag transport vehicle according to the image information of the work site through the image processing model.

[0122] Specifically, based on the image information of the work site, the image processing model determines whether the carriage of the slag transport vehicle is located at the material discharge opening of the slag discharge belt control system. If the slag transport vehicle is not parked under the slag discharge opening, the information is sent to the industrial control computer, and the industrial control computer control system automatically controls the slag transport belt to stop running until it is recognized that the slag transport vehicle is within the range of the slag discharge opening, and then starts the slag transport belt.

[0123] S2102: If the position information of the carriage of the slag transport vehicle is within the data range of the carriage slag receiving image, it is determined whether the slag in the carriage of the slag transport vehicle is fully loaded.

[0124] Specifically, if the carriage of the slag transport vehicle is at the material discharge opening of the slag discharge belt control system, the image acquisition device is used to collect whether the slag in the carriage is in a fully loaded state.

[0125] S2103: If the slag in the carriage of the slag transport vehicle is not fully loaded, the slag discharge belt control system configured with the image processing model controls the slag discharge belt to run.

[0126] Specifically, if the slag in the carriage is not fully loaded, the slag discharge belt control system controls the slag transport belt to start, drops the slag on the belt onto the slag transport vehicle, and the image acquisition device is used to collect the situation of the slag in the carriage in real time. When the slag in the carriage is fully loaded, a voice warning is issued to prompt the slag transport vehicle to drive away and prompt the next slag transport vehicle to move forward to the slag discharge position.

[0127] As can be seen from the above embodiments, the image information of the work site is obtained through the image acquisition device, and the slag discharge belt control system configured with the image processing model controls the slag discharge of the slag transport belt according to the image information of the work site. When the slag transport vehicle is fully loaded with slag, a prompt is given to the slag transport vehicle, eliminating the need for manual full-time observation and improving the slag transport efficiency.

[0128] In an embodiment of the present application, after step S2103, it further includes:

[0129] S2104: If the slag in the carriage of the slag transport vehicle is in a fully loaded state, a warning message is sent to the slag transport vehicle to cause the slag transport vehicle to move forward a distance of one carriage, and the number of fully loaded carriages is recorded.

[0130] Specifically, the video acquisition device is used to collect the slag loading state in the carriage. If the carriage is full of slag, a voice prompt message is sent to prompt the slag transport vehicle to move forward a distance of one carriage, and at the same time, the number of fully loaded carriages is recorded.

[0131] S2105: If the slag in all carriages of the slag transport vehicle is in a fully loaded state, the slag discharge belt control system controls the slag discharge belt to stop running and sends a departure command to the slag transport vehicle.

[0132] Specifically, when all carriages are full, the muck conveyor belt control system automatically controls the belt to stop running and uses voice prompts to direct the muck truck to leave the tunnel.

[0133] S2106: If the position information of the muck truck carriage is not within the data range of the carriage slag receiving image, the collected image information is sent to the industrial control computer, and the industrial control computer sends a stop slag discharge instruction to the muck conveyor belt control system, so that the muck conveyor belt control system controls the muck conveyor belt to stop running.

[0134] In this embodiment, when the position of the muck truck carriage is not aligned with the slag dropping opening or no carriage is recognized within the slag dropping opening range, the industrial control computer sends a stop slag discharge instruction to the muck conveyor belt control system, so that the muck conveyor belt control system controls the muck conveyor belt to stop running.

[0135] As can be seen from the above embodiments, by using the image acquisition device to collect the image at the slag dropping opening, when the carriage of the muck truck is in a full load state, the muck truck is prompted to move forward by the distance of one carriage and the number of full load carriages is recorded. When all carriages are full, the muck truck is prompted to leave. When the muck truck is not parked at the slag dropping opening, the information is sent to the industrial control computer, and the belt is controlled to stop running by controlling the muck conveyor belt control system until the muck truck reaches the slag dropping opening and the belt is automatically started. There is no need for manual monitoring of the muck transportation status, which improves the efficiency of muck transportation.

[0136] Figure 7 The following is a schematic structural diagram of the tunneling machine muck transportation control device provided by the present application. As Figure 7 shown, the tunneling machine muck transportation control device 70 provided in this embodiment includes: a first acquisition module 701, a division module 702, a marking module 703, a conversion module 704, a second acquisition module 705, a training module 706, a verification module 707, a configuration module 708, a third acquisition module 709, and a control module 710.

[0137] The first acquisition module 701 is used to obtain the video data of the muck conveyor belt control system through the image acquisition device.

[0138] The division module 702 is used to divide the video data into training set data, verification set data, and test set data.

[0139] The marking module 703 is used to mark the bounding boxes and target categories of the training set data, verification set data, and test set data through an image marking tool, and obtain the marked training set data, marked verification set data, and marked test set data.

[0140] The conversion module 704 is used to convert the marked training set data, marked verification set data, and marked test set data into corresponding training set heatmaps, verification set heatmaps, and test set heatmaps through a conversion function.

[0141] The second acquisition module 705 is configured to acquire the image data of the slag truck through an image acquisition device, and generate a slag receiving image of the carriage of the slag truck according to the image data of the slag truck.

[0142] The training module 706 is configured to train a pre-created image processing model according to the training set heat map, the validation set heat map, and the slag receiving image of the carriage, so as to generate a trained image processing model.

[0143] The verification module 707 is configured to verify the model accuracy of the trained image processing model according to the test set heat map.

[0144] The configuration module 708 is configured to, if the trained image processing model meets the preset model accuracy, configure the trained image processing model in the slag discharge belt control system, so as to obtain a slag discharge belt control system with the configured image processing model.

[0145] The third acquisition module 709 is configured to acquire the image information of the work site through an image acquisition device, and input the image information of the work site into the slag discharge belt control system with the configured image processing model.

[0146] The control module 710 is configured to control the operation of the slag discharge belt through the slag discharge belt control system with the configured image processing model according to the image information of the work site.

[0147] In a possible implementation manner, the training module 706 includes:

[0148] The preprocessing unit 7061 is configured to perform image preprocessing on the training set heat map, the validation set heat map, and the slag receiving image of the carriage, so as to generate a processed training set heat map, a processed validation set heat map, and a processed slag receiving image of the carriage.

[0149] The training unit 7062 is configured to train a pre-created image processing model according to the processed training set heat map and the processed slag receiving image of the carriage according to the set number of iterations, so as to obtain multiple sets of model weights of the trained image processing model.

[0150] The verification unit 7063 is configured to verify the model weights of multiple sets of trained image processing models according to the processed validation set heat map, so as to obtain the optimal weights.

[0151] The first generation unit 7064 is configured to generate a trained image processing model according to the optimal weights.

[0152] In a possible implementation manner, the loss function of the pre-created image processing model is:

[0153] Loss=λ 1 L k +λ 2 Lsize +λ 3 L off +λ 4 L cls

[0154] Wherein, L k represents the key point loss; L size represents the bounding box loss; L off represents the center point offset loss; L cls represents the target class classification loss; λ 1 represents the key point loss weight; λ 2 represents the bounding box loss weight; λ 3 represents the center point offset loss weight; λ 4 represents the target class classification loss weight; λ 1 +λ 2 +λ 3 +λ 4 = 1.

[0155] In a possible implementation manner, the second acquisition module 705 includes:

[0156] The first acquisition unit 7051 is configured to acquire first image data when the front baffle of the slag truck carriage is aligned with the blanking port through an image acquisition device.

[0157] The first annotation unit 7052 is configured to annotate the bounding box of the first image data through an image annotation tool to obtain first image position information.

[0158] The second acquisition unit 7053 is configured to acquire second image data when the rear baffle of the slag truck carriage is aligned with the blanking port through an image acquisition device.

[0159] The second annotation unit 7054 is configured to annotate the bounding box of the second image data through an image annotation tool to obtain second image position information.

[0160] The second generation unit 7055 is configured to generate a slag receiving image of the slag truck carriage according to the first image position information and the second image position information.

[0161] In a possible implementation manner, the control module 710 includes:

[0162] The recognition unit 7101 is configured to recognize the position information of the slag truck carriage according to the image information of the work site through an image processing model.

[0163] The judgment unit 7102 is configured to judge whether the slag in the slag truck carriage is fully loaded if the position information of the slag truck carriage is within the data range of the slag receiving image.

[0164] The first control unit 7103 is configured to control the operation of the slag discharge belt through the slag discharge belt control system configured with an image processing model if the soil in the carriage of the slag truck is not fully loaded.

[0165] In a possible implementation, the control module 710 further includes:

[0166] The first sending unit 7104 is configured to send a warning message to the slag truck if the soil in the carriage of the slag truck is in a fully loaded state, so that the slag truck moves forward a distance of one carriage and records the number of fully loaded carriages.

[0167] The second control unit 7105 is configured to control the slag discharge belt to stop running through the slag discharge belt control system and send a departure instruction to the slag truck if the soil in all the carriages of the slag truck is in a fully loaded state.

[0168] The second sending unit 7106 is configured to send the collected image information to the industrial control computer if the position information of the carriage of the slag truck is not within the data range of the carriage slag receiving image, and the industrial control computer sends a stop slag discharge instruction to the slag discharge belt control system to control the slag discharge belt control system to stop the operation of the slag discharge belt.

[0169] The roadheader soil transportation control device provided in this embodiment can execute the method provided in the above method embodiment, and its implementation principle and technical effect are similar, which will not be elaborated here in this embodiment.

[0170] Figure 8 It is a schematic structural diagram of the roadheader soil transportation control device provided in this application. As Figure 8 shown, the roadheader soil transportation control device 80 provided in this embodiment includes: at least one processor 801 and a memory 802. Optionally, the device 80 further includes a communication component 803. Among them, the processor 801, the memory 802, and the communication component 803 are connected through a bus 804.

[0171] In a specific implementation process, at least one processor 801 executes the computer execution instructions stored in the memory 802, so that at least one processor 801 executes the above-mentioned roadheader soil transportation control method.

[0172] The specific implementation process of the processor 801 can refer to the above method embodiment, and its implementation principle and technical effect are similar, which will not be elaborated here in this embodiment.

[0173] In the above embodiments, it should be understood that the processor may be a central processing unit (CPU for short), or other general-purpose processors, digital signal processors (DSP for short), application specific integrated circuits (ASIC for short), etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The steps of the method disclosed in combination with the invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules in the processor.

[0174] The memory may include a random access memory (RAM), and may also include a non-volatile memory (NVM), such as at least one disk memory.

[0175] The bus may be an industry standard architecture (ISA) bus, a peripheral component interconnect (PCI) bus, an extended industry standard architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For the sake of convenience in representation, the bus in the drawings of this application is not limited to only one bus or one type of bus.

[0176] This application also provides a computer program product, including a computer program, which when executed by a processor implements the above-mentioned control method for the muck transportation of a roadheader.

[0177] This application also provides a computer-readable storage medium, in which computer-executable instructions are stored. When the processor executes the computer-executable instructions, the above-mentioned control method for the muck transportation of a roadheader is implemented.

[0178] The above-mentioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as a static random access memory (SRAM), an electrically erasable programmable read-only memory (EEPROM), an erasable programmable read-only memory (EPROM), a programmable read-only memory (PROM), a read-only memory (ROM), a magnetic memory, a flash memory, a magnetic disk or an optical disc. The readable storage medium can be any available medium accessible by a general-purpose or special-purpose computer.

[0179] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be part of the processor. The processor and the readable storage medium can be located in an Application Specific Integrated Circuits (ASIC). Of course, the processor and the readable storage medium can also exist as discrete components in a device.

[0180] The division of units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Additionally, the couplings or direct couplings or communication connections shown or discussed between each other can be indirect couplings or communication connections through some interfaces, devices or units, and can be in electrical, mechanical or other forms.

[0181] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0182] Furthermore, in each embodiment of the present invention, the functional units can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit.

[0183] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in each embodiment of the present invention. The aforementioned storage medium includes various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs.

[0184] Those of ordinary skill in the art will understand that all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions. The foregoing program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments; and the foregoing storage medium includes: various media such as ROM, RAM, magnetic disk, or optical disk that can store program code.

[0185] Finally, it should be noted that those skilled in the art will readily conceive of other embodiments of the present invention after considering the specification and practicing the invention disclosed herein. The present invention is intended to cover any variations, uses, or adaptations of the present invention, which follow the general principles of the present invention and include common general knowledge or conventional technical means in the technical field not disclosed in the present invention. It is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present invention is only limited by the appended claims.

Claims

1. A method for controlling the transportation of soil and slag by a tunnel boring machine, characterized in that: Applied to a slag transportation control system, the slag transportation control system includes an image acquisition device and a slag conveyor control system; the method includes: Acquire video data of the slag discharge belt control system through the image acquisition device; Dividing the video data into training set data, validation set data and test set data; Annotating the bounding boxes and target categories of the training set data, the validation set data, and the test set data by using an image annotation tool to obtain annotated training set data, annotated validation set data, and annotated test set data; The labeled training set data, the labeled validation set data, and the labeled test set data are converted into corresponding training set heat maps, validation set heat maps, and test set heat maps through a conversion function; Acquire image data of the slag transport vehicle through the image acquisition device, and generate a slag receiving image of the carriage of the slag transport vehicle according to the image data of the slag transport vehicle; Training a pre-created image processing model according to the training set heat map, the validation set heat map, and the carriage slag connection image to generate a trained image processing model; Verifying the model accuracy of the trained image processing model according to the test set heat map; If the trained image processing model meets the preset model accuracy, the trained image processing model is configured in the slag discharge belt control system to obtain the slag discharge belt control system configured with the image processing model; Acquire image information of the work site through the image acquisition device, and input the image information of the work site into the slag conveyor control system configured with the image processing model; The slag discharge belt control system configured with the image processing model controls the operation of the slag discharge belt according to the image information of the work site.

2. The method according to claim 1, characterized in that The step of training a pre-created image processing model according to the training set heat map, the validation set heat map and the carriage slag connection image to generate a trained image processing model includes: Performing image preprocessing on the training set heat map, the validation set heat map, and the carriage slag connection image to generate a processed training set heat map, a processed validation set heat map, and a processed carriage slag connection image; The pre-created image processing model is trained according to a set number of iterations using the processed training set heat map and the processed carriage slag connection image to obtain model weights of multiple groups of trained image processing models; Verifying the model weights of the plurality of groups of trained image processing models according to the processed validation set heat map to obtain optimal weights; A trained image processing model is generated according to the optimal weights.

3. The method according to claim 2, characterized in that The pre-created image processing model includes: a space module, a context module and a detection head module; the loss function of the pre-created image processing model is: Loss=λ1L k +λ2L size +λ3L off +λ4L cls Where, L k represents the key point loss; L size represents the bounding box loss; L off Indicates the center point offset loss; L cls represents the target category classification loss; λ1 represents the key point loss weight; λ2 represents the bounding box loss weight; λ3 represents the center point offset loss weight; λ4 represents the target category classification loss weight; λ1+λ2+λ3+λ4=1.

4. The method according to claim 1, characterized in that: The method of acquiring the image data of the slag truck by the image acquisition device and generating the slag receiving image of the carriage of the slag truck according to the image data of the slag truck comprises: Acquire, by the image acquisition device, first image data when the front baffle of the carriage of the slag transport vehicle is aligned with the material drop opening; Annotating a boundary box of the first image data by an image annotation tool to obtain first image position information; Acquire, by the image acquisition device, second image data when the rear baffle of the compartment of the slag transport vehicle is aligned with the material drop opening; annotating a boundary box of the second image data by an image annotation tool to obtain second image position information; A slag receiving image of the carriage of the slag transport vehicle is generated according to the first image position information and the second image position information.

5. The method according to claim 1, characterized in that The slag discharge belt control system configured with the image processing model controls the operation of the slag discharge belt according to the image information of the work site, including: Identify the position information of the slag truck compartment according to the image information of the work site through an image processing model; If the position information of the carriage of the slag truck is within the data range of the carriage slag receiving image, it is determined whether the carriage of the slag truck is fully loaded with slag; If the slag in the carriage of the slag truck is not fully loaded, the operation of the slag conveyor belt is controlled by the slag conveyor belt control system configured with the image processing model.

6. The method according to claim 5, characterized in that If the slag in the slag truck compartment is not fully loaded, after the slag conveyor belt control system configured with the image processing model controls the operation of the slag conveyor belt, the method further includes: If the slag in the carriage of the slag truck is fully loaded, an early warning message is sent to the slag truck to make the slag truck move forward by the distance of one carriage and record the number of fully loaded carriages; If the slag in all carriages of the slag transport vehicle is fully loaded, the slag transport belt control system is used to control the slag transport belt to stop running, and a departure instruction is sent to the slag transport vehicle; If the position information of the slag transport vehicle compartment is not within the data range of the compartment slag receiving image, the collected image information is sent to the industrial computer, and the industrial computer sends a slag discharging stop instruction to the slag discharging belt control system, so that the slag discharging belt control system controls the slag discharging belt to stop running.

7. A roadheader soil transportation control device, characterized in that: Applied to a slag transportation control system, the slag transportation control system includes an image acquisition device and a slag conveyor control system; the device includes: A first acquisition module, used for acquiring video data of the slag discharge belt control system through the image acquisition device; A division module, used for dividing the video data into training set data, verification set data and test set data; An annotation module is used to annotate the bounding boxes and target categories of the training set data, the validation set data, and the test set data by using an image annotation tool to obtain the annotated training set data, the annotated validation set data, and the annotated test set data; A conversion module, used to convert the labeled training set data, the labeled validation set data and the labeled test set data into corresponding training set heat map, validation set heat map and test set heat map through a conversion function; A second acquisition module is used to acquire image data of the slag transport vehicle through the image acquisition device, and generate a slag receiving image of the carriage of the slag transport vehicle according to the image data of the slag transport vehicle; A training module, used for training a pre-created image processing model according to the training set heat map, the validation set heat map and the carriage slag connection image, so as to generate a trained image processing model; A verification module, used to verify the model accuracy of the trained image processing model according to the test set heat map; A configuration module, configured to configure the trained image processing model to the slag conveyor control system if the trained image processing model meets the preset model accuracy, so as to obtain the slag conveyor control system configured with the image processing model; A third acquisition module is used to acquire image information of the work site through the image acquisition device, and input the image information of the work site into the slag conveyor control system configured with the image processing model; A control module is used to control the operation of the slag discharge belt according to the image information of the work site through the slag discharge belt control system configured with the image processing model.

8. A roadheader soil transportation control device, characterized in that: include: at least one processor and memory; The memory stores computer-executable instructions; The at least one processor executes the computer-executable instructions stored in the memory, so that the at least one processor executes the tunnel boring machine slag transportation control method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the tunnel boring machine slag transportation control method as described in any one of claims 1 to 6.

10. A computer program product, characterized in that It comprises a computer program, which, when executed by a processor, implements the method for controlling the transportation of slag from a tunnel boring machine as claimed in any one of claims 1 to 6.

Images