Underground space construction path adaptive control system based on multi-field information fusion

Abstract

The invention provides an underground space construction path adaptive control system based on multi-field information fusion. The method comprises the following specific steps: deploying a microgravity sensor array through a space mapping module, collecting gravity anomaly data of different depths underground, and constructing a three-dimensional geological topology model; the navigation decision module is used for coupling the optical flow signal and the continuous resistance distribution function, generating a guide vector field and realizing dynamic deviation correction according to the equipment pose error and concrete sound velocity feedback; the collaborative execution module calculates the inclination angle and the energy compactness of the vibrating rod and outputs a vibrating execution evidence packet; and the closed-loop monitoring module performs error analysis on the vibration effect and executes local interpolation or global reconstruction to realize adaptive optimization of the model. Compared with an existing floor construction system which only depends on static design and experience operation, the system can achieve high-precision recognition of the density gradient of the underground structure, dynamic cooperation of a path and energy control and real-time feedback closed-loop adjustment of construction quality, and the construction precision and consistency are remarkably improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent construction, in particular to a construction path adaptive control system for underground space based on multi-field information fusion. BACKGROUND

[0002] With the continuous development of information technology and intelligent construction concept, the field of construction engineering is gradually transforming towards digitization, refinement and intelligence. In underground engineering and complex geological environment construction, BIM technology has been widely used due to its advantages in visual design, construction management and information integration. At the same time, AR technology has gradually been integrated into the construction site due to its intuitive nature in three-dimensional scene perception and human-computer interaction, which is used to improve construction precision and operation efficiency. In the floor construction, especially in the scene involving underground multi-gradient structure, the traditional construction process is gradually embedded into the intelligent construction unit.

[0003] However, the existing underground multi-gradient floor intelligent construction system generally lacks the ability to perceive micro-gravity anomalies in complex underground structure environments, making it difficult to build a geological model that conforms to the real density distribution, resulting in difficulties in adjusting the construction scheme according to local conditions. Secondly, the navigation control system embedded in the existing underground multi-gradient floor intelligent construction system is mostly based on preset paths or inertial guidance, making it difficult to respond to equipment pose deviation and concrete rheological state in real time, and path deviation and energy mismatch are likely to occur in areas with uneven density requirements. Finally, the existing underground multi-gradient floor intelligent construction system lacks a feedback mechanism based on actual execution data during the vibrating operation process, making it difficult to effectively judge the consistency of the vibrating effect and the geological environment, and lacking support for model correction and adaptive adjustment of construction parameters, which is likely to cause local construction quality fluctuations.

[0004] Therefore, the present application proposes a construction path adaptive control system for underground space based on multi-field information fusion. SUMMARY

[0005] In order to solve the technical problems of geological modeling distortion, navigation response lag and vibrating feedback deficiency of the existing underground multi-gradient floor intelligent construction system mentioned in the background art, the purpose of the present application is to provide a construction path adaptive control system for underground space based on multi-field information fusion.

[0006] In order to achieve the above-mentioned purpose, the present application adopts the following technical solutions:

[0007] The construction path adaptive control system for underground space based on multi-field information fusion comprises:

[0008] M1, a space mapping module, acquires geological micro-gravity field data to construct a geological topology model, calculates a three-dimensional gradient vector field of the geological topology model, and outputs an optical flow signal;

[0009] M2, the navigation decision module, performs resistance field correction on the optical flow signal and couples the optical flow signal with the continuous resistance distribution function to obtain the guiding vector field. At the same time, it monitors the rheological state of concrete. When the equipment posture deviation exceeds the limit, it activates the dynamic resistance field correction and outputs navigation flow control commands.

[0010] M3, the collaborative execution module, updates the navigation flow control instructions and sequentially executes the calculation of the vibratory rod inclination angle and the calculation of the vibration energy density to output the vibration execution evidence package;

[0011] M4, the closed-loop monitoring module, takes the vibration execution evidence package as input, compares the vibration energy density at the same coordinate point with the expected vibration energy density to calculate the error value, and performs dual-channel optimization of the geological topology model based on the error result.

[0012] Furthermore, gravity anomaly values ​​at different underground depths and planar coordinates were collected by deploying a microgravity sensor array in the construction area;

[0013] Suppose that n discrete points are collected, the dataset of the geological microgravity field data is stored in the form of discrete points. ,in, For the first The planar coordinates of a discrete point; For the first The underground depth of a discrete point; For the first Gravity anomaly values ​​at discrete points;

[0014] The difference The calculation formula is: ,in, For the first The actual observed values ​​at discrete points; For the first Normal gravity values ​​at discrete points;

[0015] The normal gravity value Geographic latitude needs to be considered in actual calculations. and altitude : ;in, latitude is Altitude is Normal gravity at that location; This represents the normal gravity value at the equator; A coefficient related to the flattening of the Earth's ellipsoid; This is the first eccentricity of the Earth's ellipsoid; The radius of the Earth at the equator;

[0016] The dataset of the geological microgravity field data The three-dimensional gravity anomaly field function is constructed and defined as follows: ;in, This represents the three-dimensional gravity anomaly field function at point [point]. The function value at that location; The total number of discrete points; For the target point and the first Spatial Euclidean distance between discrete points; The distance weight is a power of the distance.

[0017] The spatial Euclidean distance The calculation formula is: The three-dimensional gravity anomaly field function The data is input into the 3D voxel modeling module to construct the geological topology model. The specific steps are as follows:

[0018] The spatial region is divided into a regular three-dimensional voxel grid. The coordinates of the center point of each voxel are Subsequently, based on the constructed three-dimensional gravity anomaly field function... At the center point of each voxel The above is evaluated to obtain outliers. ,and .

[0019] Furthermore, the threshold is set as follows: To structurally partition the voxels: High-density area This is a low-density anomaly area. For the conventional area, the Marching Cubes value surface extraction algorithm is used to generate the boundary mesh to output the geological topology model. ;

[0020] Based on the geological topology model As input, calculate its three-dimensional gradient vector field: ;in, This is a three-dimensional gradient vector field, representing the direction and rate of density change; For gradient operators; Representing the geological topological model along Rate of change in the axial direction; Representing the geological topological model along Rate of change in the axial direction; Representing the geological topological model along Rate of change in the axial direction;

[0021] Through the three-dimensional gradient vector field , generate optical flow signal ,and ,in, is the scaling factor for the optical flow signal.

[0022] Furthermore, based on the aforementioned three-dimensional gravity anomaly field function Define the continuous resistance distribution function ,and ,in, This is the resistance adjustment coefficient;

[0023] The optical flow signal With the continuous resistance distribution function Coupling yields the guiding vector field. ,and ,in, This is the drag coupling coefficient;

[0024] The current end position of the construction equipment is Therefore, the target pose predicted based on the previous position and the guiding vector field is: ;

[0025] in, Indicates the target pose; This is the end position recorded in the previous control cycle; To control the cycle time length;

[0026] Based on the target pose Calculate the pose error vector ,and ;

[0027] If satisfied If the system determines that the current device is still within the allowed range of the boot path, no adjustment is needed; if the conditions are met... If the system determines that the current device has a navigation deviation, it will enter a correction state; among which, This is the error tolerance threshold;

[0028] The system uses an embedded acoustic velocity meter integrated into the end of the vibrating equipment to collect the propagation velocity of sound inside the concrete in real time. ,and ,in, Fixed propagation path length between sensors; This refers to the time difference in sound wave propagation.

[0029] Based on the speed of sound propagation Calculation of concrete rheological state parameters ;in, These are the standardized coefficients; This refers to the theoretical density of concrete. For reference shear modulus;

[0030] like If so, it means that the concrete rheological state index is normal; if This indicates that the concrete rheological state index is abnormal.

[0031] Furthermore, if the following conditions are met or Under any condition, dynamic drag field correction is initiated. Specifically, the steps of dynamic drag field correction include navigation attitude correction, energy excitation adjustment, and navigation path micro-reconstruction.

[0032] The navigation attitude correction is used to adjust the navigation guidance direction: ;

[0033] in, The guide vector after attitude correction; Adjust the weighting factor for the attitude;

[0034] The energy excitation regulation is used to dynamically control the energy density per unit time: ;in, Energy density per unit time; The system vibration efficiency coefficient; This refers to the vibration frequency; This refers to the vibration amplitude.

[0035] like Increase the vibration frequency and vibration amplitude, and refer to the table for values. ;like Keep the default value and retrieve the value from the table. ;like Reduce the vibration frequency and vibration amplitude, and refer to the table for values. ;

[0036] The navigation path micro-reconstruction is used to re-search for the minimum-cost path in the current network domain, so that the guiding vector field is updated as follows: ;in, The candidate path offset vector;

[0037] This represents the local resistance value. For offset cost; The resistance is a finite weighting coefficient;

[0038] This refers to the distance-first weighting coefficient.

[0039] The resistance finite weight coefficient The value is 0.7, and the distance priority weight coefficient is... The value is 0.3;

[0040] Based on the updated guiding vector field The navigation flow control command is generated and defined as follows: ,and ,in, This represents the theoretical maximum propulsion speed.

[0041] Furthermore, the guidance vector field corresponding to the navigation flow control command is updated. For, among which, The guiding vector field is respectively in , , Components along the axial direction;

[0042] Based on the components The angle between the vibrator and the vertical direction is defined as the angle between the projection of the guiding vector onto the horizontal plane and the vertical direction. The calculation formula is: ,in, It is the arctangent function;

[0043] The duration of vibration by the vibrator at a certain coordinate point is: The calculation formula is: ,in Let be the effective radius of the vibrator, and record the time of vibration completion as . ;

[0044] Based on the energy density per unit time and the duration of vibration The vibration energy density was obtained. ,and ;

[0045] Based on the spatial coordinates The time when vibration is completed The inclination angle of the vibrator The vibration energy density Generate the vibration execution evidence package : ;

[0046] in, This marks the completion of construction work.

[0047] Furthermore, the predicted density Based on the geological topology model and the compaction density of the vibration energy The combined calculation reflects the expected density of this coordinate point under theoretical construction conditions. The calculation formula is as follows: ;

[0048] in, This is a density correction factor used to correlate the linear relationship between the geological topology model and the density. This represents the density characteristic value of the geological topology model at this coordinate point. The basic density constant;

[0049] Based on the vibration energy density and the expected vibration energy density Calculate coordinate points error value ,and ;

[0050] Based on the error value The numerical value corresponds to dividing it into three levels: when When it is judged as a level 1 error, it means that the error is within the allowable range and no optimization is needed; when When it is judged as a level 2 error, it indicates that the error is moderate and local optimization is required; when If the error is classified as Level 3, it indicates a significant error that requires global optimization.

[0051] Furthermore, when the error is at the second level, the density eigenvalues ​​of local voxels are corrected by adjusting the interpolation weights of the three-dimensional gravity anomaly field function:

[0052] Based on the error value Calculate the correction factor ,and ,in, This is the error sensitivity coefficient, used to control the correction strength; simultaneously, based on the correction factor... Correct coordinates The corresponding voxel density eigenvalues ​​are used to generate the locally optimized geological topology model. , and;

[0053] When the error is at the third level, based on the error value Regenerate threshold : ;in, The original threshold; This is the threshold adjustment coefficient; and then based on the new threshold... The voxel structure partitioning and Marching Cubes value surface extraction algorithm steps are re-executed to generate the globally optimized geological topology model.

[0054] Compared with the prior art, the advantages of the present invention are as follows:

[0055] 1. To address the problem that existing technologies cannot accurately reflect changes in the density of underground structures, this invention introduces a three-dimensional inverse distance weighted interpolation algorithm driven by microgravity data. This algorithm transforms gravity anomaly values ​​at different depths and planar coordinates into a spatially continuous gravity anomaly field. By dividing the data into voxel structures and threshold density partitions, a geological topological model that reflects the hierarchical nature of underground structures is constructed. This enables spatial identification and boundary extraction of complex areas such as cavities and high-density areas, significantly improving modeling accuracy and regional resolution.

[0056] 2. To address the problem that traditional path control is difficult to dynamically respond to geological disturbances and equipment deviations, this invention constructs a guiding vector field based on the joint construction of optical flow field and resistance field. Combined with the real-time pose of the construction equipment end and the sound velocity feedback inside the concrete, it adaptively determines whether to enter the correction state and activates mechanisms such as attitude correction, energy adjustment and navigation path micro-reconstruction. It dynamically updates navigation flow control commands at three levels: guidance direction, propulsion intensity and path structure, to achieve precise crossing and attitude stability control of high resistance areas, boundary zones or areas of abrupt state change.

[0057] 3. To address the lack of data closed-loop control in existing vibration compaction operations, this invention uses the pose and energy parameters output by navigation commands as a basis, combined with construction time, vibration inclination angle, and energy density, to generate a multi-dimensional vibration execution evidence package. This package is then compared with the expected compaction degree derived from the topology model to perform point-by-point error analysis. By classifying the error levels, local interpolation correction or global partition reconstruction is triggered, effectively avoiding the compaction degree drift problem caused by geological changes. This achieves a high-frequency closed loop of construction-perception-inversion-correction, ensuring that the compaction quality is consistent with the model in real time. Attached Figure Description

[0058] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0059] Fig. 1 This is a schematic diagram of the system workflow of the present invention;

[0060] Fig. 2 This is a schematic diagram of the geological topology model construction process of the present invention;

[0061] Fig. 3 This is a schematic diagram of the navigation flow control command generation process of the present invention. Detailed Implementation

[0062] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0063] To achieve the above objectives, the present invention provides an adaptive control system for underground space construction paths based on multi-field information fusion, such as... Figs. 1-3 As shown, the system includes:

[0064] M1, the spatial mapping module, collects geological microgravity field data to construct a geological topology model, calculates the three-dimensional gradient vector field of the geological topology model, and outputs optical flow signals.

[0065] Gravity anomaly values ​​at different depths and planar coordinates underground are collected by a microgravity sensor array deployed in the construction area. The measurement accuracy of the microgravity sensors needs to reach [a certain level]. ,and To capture gravity changes caused by differences in the density of geological bodies;

[0066] In this embodiment, assuming n discrete points are collected, the dataset of the geological microgravity field data is stored in the form of discrete points. ,in, For the first The planar coordinates of a discrete point (unit: m); For the first The underground depth of each discrete point (unit: m); For the first The gravity anomaly value at the nth discrete point, i.e. the th The difference between the actual gravity value at a discrete point and the theoretical value of the normal gravity field is used to reflect the abnormal characteristics of the density distribution of underground geological bodies, such as high-density rock layers leading to... If positive, voids will lead to Negative; The total number of discrete points;

[0067] The difference The calculation formula is:

[0068] ,in, For the first The actual observed values ​​at discrete points; For the first Normal gravity values ​​at discrete points;

[0069] The normal gravity value Geographic latitude needs to be considered in actual calculations. and altitude : ;

[0070] in, latitude is Altitude is The normal gravity value at that location, in units of ; This represents the normal gravity value at the equator; A coefficient related to the flattening of the Earth's ellipsoid; This is the first eccentricity of the Earth's ellipsoid; The radius of the Earth at the equator;

[0071] In this embodiment, the coefficient The value is 0.001931851353; the first eccentricity The square value is Earth's radius The value is ;

[0072] The dataset of the geological microgravity field data Construct a three-dimensional gravity anomaly field function, and use a three-dimensional inverse distance weighted interpolation algorithm for any spatial point. For the estimation, the three-dimensional gravity anomaly field function is defined as follows: ;

[0073] in, This represents the three-dimensional gravity anomaly field function at point [point]. The function value at that location; The total number of discrete points; For the target point and the first Spatial Euclidean distance between discrete points; The distance weight is a power of the distance.

[0074] In this embodiment, the distance weight power The constant value is 2;

[0075] The spatial Euclidean distance The calculation formula is: ;

[0076] The three-dimensional gravity anomaly field function The data is input into the 3D voxel modeling module to construct the geological topology model. The specific steps are as follows:

[0077] The spatial region is divided into a regular three-dimensional voxel grid. The coordinates of the center point of each voxel are Subsequently, based on the constructed three-dimensional gravity anomaly field function... At the center point of each voxel The above is evaluated to obtain outliers. ,and ;

[0078] Subsequently, the threshold was set as To structurally partition the voxels: For high-density areas (such as rock strata), This is a low-density anomaly area (such as a cavity). For the conventional region, the MarchingCubes value surface extraction algorithm is used to generate the boundary mesh, thus outputting the geological topology model. ;

[0079] Based on the geological topology model As input, calculate its three-dimensional gradient vector field: ;

[0080] in , which is a three-dimensional gradient vector field, representing the direction and rate of density change, and is used to construct spatial optical flow; This is a gradient operator used to find the spatial directional derivative of a three-dimensional scalar function. Representing the geological topological model along Rate of change in the axial direction; Representing the geological topological model along Rate of change in the axial direction ; indicates that the geological topological model is along Rate of change in the axial direction;

[0081] Through the three-dimensional gradient vector field Generate optical flow signal ,and ,in, This is a scaling factor for the optical flow signal, used to adjust the gradient intensity;

[0082] In this embodiment, the scaling factor The example value is 0.5.

[0083] M2, the navigation decision module, performs resistance field correction on the optical flow signal and couples the optical flow signal with the continuous resistance distribution function to obtain the guiding vector field. At the same time, it monitors the rheological state of the concrete. When the equipment posture deviation exceeds the limit, it activates the dynamic resistance field correction and outputs navigation flow control commands.

[0084] Based on the three-dimensional gravity anomaly field function Define the continuous resistance distribution function ,and ,in, This is the resistance adjustment coefficient;

[0085] In this embodiment, the drag adjustment coefficient The empirical value is 1.2~2.5;

[0086] The optical flow signal With the continuous resistance distribution function

[0087] Coupling yields the guiding vector field. ,in, This is the resistance coupling coefficient, used to adjust the correlation strength between the navigation path and the geological resistance distribution;

[0088] In this embodiment, the drag coupling coefficient The value is set to 0.6 to ensure the responsiveness of the navigation path while fully considering the impact of geological resistance on path selection; for areas with drastic changes in geological structure or dense underground cavities, it can be adjusted to 0.8; while in continuous homogeneous areas, it can be adjusted to 0.4 to improve propulsion efficiency.

[0089] In this embodiment, the current end position of the construction equipment is

[0090] Therefore, the target pose predicted based on the previous position and the guiding vector field is: ;

[0091] in, Indicates the target pose; This is the end position recorded in the previous control cycle; To control the cycle time length;

[0092] Based on the target pose Calculate the pose error vector ,and ;

[0093] If full If the conditions are met, the system determines that the current device is still within the allowed range of the boot path and no adjustment is needed; if the conditions are met... If the system determines that the current device has a navigation deviation, it will enter a correction state; among which, This is the error tolerance threshold;

[0094] In this embodiment, the error tolerance threshold The default value is 0.05m;

[0095] The system uses an embedded acoustic velocity meter integrated into the end of the vibrating equipment to collect the propagation velocity of sound inside the concrete in real time. ,and ,in, Fixed propagation path length between sensors; This refers to the time difference in sound wave propagation.

[0096] Based on the speed of sound propagation Calculation of concrete rheological state parameters ;

[0097] in, These are the standardized coefficients; This refers to the theoretical density of concrete. For reference shear modulus;

[0098] like If so, it means that the concrete rheological state index is normal; if This indicates that the concrete rheological state index is abnormal.

[0099] In this embodiment, the standardized coefficient The recommended value is 2.5; The value is 1.8; The value is 3.2;

[0100] If satisfied Under any condition, dynamic drag field correction is initiated. Specifically, the steps of dynamic drag field correction include navigation attitude correction, energy excitation adjustment, and navigation path micro-reconstruction.

[0101] The navigation attitude correction is used to adjust the navigation guidance direction: ;

[0102] in, The guide vector after attitude correction; Adjust the weighting factor for the attitude;

[0103] In this embodiment, the attitude correction weight factor The range of values ​​is as follows ;

[0104] The energy excitation regulation is used to dynamically control the energy density per unit time: ;

[0105] in, Energy density per unit time; The system vibration efficiency coefficient; This refers to the vibration frequency; This refers to the vibration amplitude.

[0106] In this embodiment, if Increase the vibration frequency and vibration amplitude, and refer to the table for values. ;like Keep the default value and retrieve the value from the table. ;like Reduce the vibration frequency and vibration amplitude, and refer to the table for values. ;

[0107] The navigation path micro-reconstruction is used to re-search for the minimum-cost path in the current network domain, so that the guiding vector field is updated as follows: ;

[0108] in, Candidate path offset vector This represents the local resistance value. Offset cost; This is a drag-limited weighting coefficient used to enhance obstacle avoidance capabilities; This is the distance priority weighting coefficient, used to ensure the shortest possible guiding path;

[0109] In this example, the resistance finite weighting coefficient The value is 0.7, and the distance priority weight coefficient is... The value is 0.3;

[0110] Based on the updated guiding vector field The navigation flow control command is generated and defined as follows: ,and ,in, This is the theoretical maximum propulsion speed;

[0111] In this embodiment, the theoretical maximum propulsion speed The value is 12 mm / s.

[0112] M3, the collaborative execution module, updates the navigation flow control instructions and sequentially executes the calculation of the vibratory rod inclination angle and the calculation of the vibration energy density to output the vibration execution evidence package.

[0113] In this embodiment, the guidance vector field corresponding to the navigation flow control command is updated to... ,in, The guiding vector field is respectively in , , Components along the axial direction;

[0114] Based on the components The angle between the vibrator and the vertical direction is defined as the angle between the projection of the guiding vector onto the horizontal plane and the vertical direction. The calculation formula is: ,in, It is the arctangent function;

[0115] The duration of vibration by the vibrator at a certain coordinate point is: The calculation formula is: ,in Let be the effective radius of the vibrator, and record the time of vibration completion as . ;

[0116] In this embodiment, the effective radius of action The default value is 0.3m;

[0117] Based on the energy density per unit time and the duration of vibration The vibration energy density was obtained. ,and ;

[0118] Based on the spatial coordinates The time when vibration is completed The inclination angle of the vibrator The vibration energy density Generate the vibration execution evidence package ;

[0119] in, This marks the completion of construction work.

[0120] M4, the closed-loop monitoring module, takes the vibration execution evidence package as input, compares the vibration energy density at the same coordinate point with the expected vibration energy density to calculate the error value, and performs dual-channel optimization of the geological topology model based on the error result.

[0121] The predicted density Based on the geological topology model and the compaction density of the vibration energy The combined calculation reflects the expected density of this coordinate point under theoretical construction conditions. The calculation formula is as follows:

[0122] in, This is a density correction factor used to correlate the linear relationship between the geological topology model and the density. This represents the density characteristic value of the geological topology model at this coordinate point. The basic density constant;

[0123] In this embodiment, the density correction coefficient The basic density constant The default value is ;

[0124] Based on the vibration energy density and the expected vibration energy density Calculate coordinate points error value ,and ;

[0125] Based on the error value The numerical value corresponds to dividing it into three levels: when When it is judged as a level 1 error, it means that the error is within the allowable range and no optimization is needed; when When it is judged as a level 2 error, it indicates that the error is moderate and local optimization is required; when When the error is classified as Level 3, it indicates a significant error that requires global optimization.

[0126] When the error is at the second level, the density eigenvalues ​​of local voxels are corrected by adjusting the interpolation weights of the three-dimensional gravity anomaly field function:

[0127] Based on the error value Calculate the correction factor ,and ,in, This is the error sensitivity coefficient, used to control the correction strength; simultaneously, based on the correction factor... Correct coordinates The corresponding voxel density eigenvalues ​​are used to generate the locally optimized geological topology model. ,and ;

[0128] In this embodiment, the error sensitivity coefficient The default value is 0.6;

[0129] When the error is at the third level, based on the error value Regenerate threshold : ;

[0130] in, The original threshold; This is the threshold adjustment coefficient;

[0131] In this embodiment, the threshold adjustment coefficient The default value is 0.3;

[0132] Furthermore, based on the new threshold The voxel structure partitioning and Marching Cubes value surface extraction algorithm steps are re-executed to generate the globally optimized geological topology model.

[0133] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0134] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A system for adaptive control of underground space construction path based on multi-field information fusion, characterized in that, Comprise: M1, a spatial mapping module, collects geological micro-gravity field data to construct a geological topology model, calculates a three-dimensional gradient vector field of the geological topology model to output an optical flow signal; M2, a navigation decision module, corrects the optical flow signal with a resistance field, couples the optical flow signal with a continuous resistance distribution function to obtain a guide vector field, simultaneously monitors the rheological state of concrete, activates a dynamic resistance field when the device pose deviation exceeds the limit, and outputs a navigation flow control instruction; M3, a collaborative execution module, updates the navigation flow control instruction, and sequentially executes the vibration rod inclination calculation and the vibration energy density calculation to output a vibration execution evidence package; M4, a closed-loop monitoring module, takes the vibration execution evidence package as input, compares the vibration energy density at the same coordinate point with the expected vibration energy density to calculate an error value, and optimizes the geological topology model based on the error result through a double-channel. 2.The multi-field information fusion based underground space construction path adaptive control system according to claim 1, characterized in that, Through the micro-gravity sensor array deployed in the construction area, the gravity anomaly values at different depths and plane coordinates are collected; Suppose n discrete points are collected, the data set of the geological micro-gravity field data is stored in the form of discrete points as wherein, is the plane coordinate of the first discrete point; is the underground depth of the first discrete point; is the gravity anomaly value of the first discrete point; The difference The formula for calculating the difference is: where, is the actual observed value of the gravity at the discrete point; is the normal gravity value at the discrete point. the normal gravity value The geographical latitude needs to be considered in the actual calculation and the altitude : ; wherein, is the normal gravity at a latitude of and an altitude of ; is the normal gravity value at the equator; is a coefficient related to the Earth's ellipsoid flattening; is the first eccentricity of the Earth's ellipsoid; is the Earth's radius at the equator; The dataset of the geological microgravity field data The three-dimensional gravity anomaly field function is constructed and defined as follows: ;in, This represents the three-dimensional gravity anomaly field function at point [point]. The function value at that location; The total number of discrete points; For the target point and the first Spatial Euclidean distance between discrete points; The distance weight is a power of the distance. The spatial Euclidean distance The calculation formula is: ; input the three-dimensional gravity anomaly field function to a three-dimensional voxel modeling module to construct the geological topological model, and the specific steps are: Dividing a spatial region into a regular three-dimensional voxel grid , each voxel center point coordinate is ; then, based on the constructed three-dimensional gravity anomaly field function , the evaluation is performed at the center point of each voxel , and the anomaly value is obtained, and . 3.The multi-field information fusion based underground space construction path adaptive control system according to claim 2, characterized in that, Setting the threshold value as Structural partitioning of the voxel is performed as follows: For high-density areas, For low-density abnormal areas, For normal areas; finally, a Marching Cubes value surface extraction algorithm is used to generate a boundary grid to output the geological topological model ; Based on the geological topology model As input, calculate its three-dimensional gradient vector field: ;in, This is a three-dimensional gradient vector field, representing the direction and rate of density change; For gradient operators; Representing the geological topological model along Rate of change in the axial direction; Representing the geological topological model along Rate of change in the axial direction; Representing the geological topological model along Rate of change in the axial direction; by the three-dimensional gradient vector field generating an optical flow signal and wherein is a scaling factor for the optical flow signal. 4.The underground space construction path self-adaptive control system based on multi-field information fusion of claim 2, wherein, based on the three-dimensional gravity anomaly field function , defining a continuous resistance distribution function , and wherein, is a resistance adjustment coefficient; coupling the optical flow signal with the continuous resistance distribution function results in a guidance vector field , and wherein, is a resistance coupling coefficient; Let the current end position of the construction equipment be , and the target pose predicted based on the last time position and the guide vector field be: ; wherein, represents a target pose; is an end position recorded for the previous control cycle; is a control cycle time length; based on the target pose calculating a pose error vector , and ; If the following condition is met , the system determines that the current device is still within the allowable range of the guiding path and no adjustment is needed. If the following condition is met , the system determines that there is a deviation in navigation of the current device and enters the deviation correction state. Wherein, is an error tolerance threshold. The system collects the propagation speed of sound in the concrete in real time through the embedded sound wave speed measuring instrument integrated at the end of the vibrating device , and wherein, is the fixed propagation path length between sensors; is the sound wave propagation time difference; based on the propagation speed of sound calculating a concrete rheological state indicator ; wherein, is a normalization coefficient; is a theoretical density of the concrete; is a reference shear modulus; If , it represents that the concrete rheological state index is normal; if , it represents that the concrete rheological state index is abnormal.

5. The underground space construction path self-adaptive control system based on multi-field information fusion according to claim 4, characterized in that, If satisfied or Under any condition, dynamic drag field correction is initiated. Specifically, the steps of dynamic drag field correction include navigation attitude correction, energy excitation adjustment, and navigation path micro-reconstruction. The navigation attitude correction is used to adjust the navigation guidance direction: ; wherein, is the attitude corrected guidance vector; is the attitude correction weight factor; The energy excitation regulation is used to dynamically control the energy density per unit time: ; wherein, is the energy density per unit time; is the system vibration energy efficiency coefficient; is the vibration frequency; is the vibration amplitude; If increase the frequency and amplitude of the vibration, look up the values ; if keep the default values, look up the values ; if decrease the frequency and amplitude of the vibration, look up the values ; The navigation path micro-reconstruction is used to re-search the minimum cost path in the current network field, so that the guide vector field is updated as: ; wherein, is a candidate path offset vector; is a local resistance value; is an offset cost; is a resistance limited weight coefficient; is a distance priority weight coefficient; In the present example, the limited resistance weight coefficient has a value of 0.7, and the distance priority weight coefficient has a value of 0.

3. based on the updated guidance vector field generating the navigation flow control instructions, defined as , and wherein, is the theoretical maximum propulsion speed. 6.The multi-field information fusion based underground space construction path adaptive control system according to claim 1, wherein, updating the guidance vector field corresponding to the navigation flow control instruction wherein, are respectively the components of the guiding vector field in the 、 、 axis direction; based on the components The inclination of the vibrator with respect to the vertical is defined as the angle between the projection of the guide vector on the horizontal plane and the vertical The calculation formula is wherein is the arctangent function; The duration of vibration of the vibrating rod at a coordinate point is , The calculation formula is wherein is the effective action radius of the vibrating rod, and the time when the vibrating is completed is recorded as ; based on the energy density per unit time and the duration of the vibration deriving the vibration energy density , and ; based on the spatial coordinates , the time of completion of the vibration , the inclination of the vibrator , the vibration energy density generating the vibration execution evidence package : ; wherein, is a construction execution completion confirmation mark.

7. The underground space construction path adaptive control system based on multi-field information fusion according to claim 6, characterized in that, The pre-density Based on the geological topological model And the vibration energy density Joint calculation to reflect the expected density of the coordinate point under the theoretical construction condition, the calculation formula is: ; wherein, is a density correction factor for correlating the geological topology model with the linear relationship of density; is the density eigenvalue of the geological topology model at the coordinate point; is a base density constant; based on the vibrated energy density and the expected vibrated energy density calculating an error value of the coordinate point , and ; based on the error value The numerical value of the error value corresponds to dividing it into three levels: when it is determined as a first-level error, indicating that the error is within the allowable range and does not need to be optimized; when it is determined as a second-level error, indicating that the error is moderate and needs to be locally optimized; and when it is determined as a third-level error, indicating that the error is significant and needs to be globally optimized. 8.The multi-field information fusion based underground space construction path adaptive control system according to claim 7, characterized in that, When in the secondary error, the interpolation weight of the three-dimensional gravity anomaly field function is adjusted to correct the density characteristic value of the local voxel: based on the error value computing a correction factor , and wherein, is an error sensitivity coefficient for controlling the correction strength; and based on the correction factor correcting the coordinate point corresponding voxel density feature value to generate a locally optimized geological topology model , and; when in the third error, based on the error value re-generate threshold value : ; wherein, is the original threshold value; is the threshold adjustment coefficient; further based on the new threshold re-performing the voxel structure partitioning and the Marching Cubes value surface extraction algorithm steps to generate the globally optimized geological topology model.

Images