A Method for Reconstructing and Measuring the 3D Trajectory of Upward-Facing Holes Based on Dual-Path Constraints

By using the method of bidirectional same track constraint, common events are identified, monotonic depth mapping and common hole axis state are established, and the model is solved iteratively. This solves the error accumulation problem in the upward borehole trajectory measurement and realizes high-precision three-dimensional trajectory reconstruction and skew measurement.

CN123109635APending Publication Date: 2026-11-17CENT SOUTH UNIV +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611208070.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-08-11
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

In the existing technology, the upward borehole trajectory measurement method suffers from the accumulation of single-pass inertial drift error, inconsistency of two-pass measurement data, and the influence of local abnormal observations, which makes it difficult to guarantee measurement accuracy and cannot accurately reconstruct the three-dimensional trajectory of the borehole.

Method used

A method based on dual-path constraints is adopted. Common events are identified through borehole entry and withdrawal measurement data, a monotonic depth mapping is established, a common borehole axis state is set, a dual-path observation model is established, and the direction error and depth correspondence error are corrected by iterative solution of the joint objective function, and the three-dimensional center trajectory of the upward borehole is reconstructed.

Benefits of technology

It effectively reduces the impact of single-journey inertial drift and inconsistency between two-journey mileage on trajectory reconstruction, and achieves high-precision reconstruction of the three-dimensional center trajectory and deviation measurement of the upward blast hole, providing a basis for construction quality evaluation and blasting design optimization.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123109635A_ABST
    Figure CN123109635A_ABST
Patent Text Reader

Abstract

A method for reconstructing the 3D trajectory and measuring the skewness of an upward-facing borehole based on a two-way track constraint is proposed. The method acquires parameters by collecting inertial data, mileage data, and state data from the probe's entry and withdrawal journeys. Through time synchronization, error compensation, attitude calculation, anomaly identification, and orientation unification, the observation sequences of the entry and withdrawal borehole axes are obtained. A monotonic depth mapping between the entry and withdrawal mileages is established using common feature events at the borehole opening, bottom, and interior. A unique common borehole axis state is set in the common physical depth domain, and prediction sequences for the entry and withdrawal borehole axes are established separately. A joint objective function is constructed using the two-way prediction and observation residuals to simultaneously solve for the common borehole axis state, two-way measurement error parameters, and depth mapping parameters. The 3D center trajectory of the borehole is reconstructed segment by segment from the common borehole axis state, and compared with the designed trajectory to obtain the deviation along the path, the bottom deviation, and the local borehole axis direction change rate. This method can accurately and efficiently obtain the 3D center trajectory and skewness measurement results of an upward-facing borehole.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention belongs to the field of intelligent measurement and spatial trajectory reconstruction technology of underground mine blast holes, specifically involving a method for reconstructing the three-dimensional trajectory of upward blast holes and measuring their deviation based on dual-path parallel constraints. Background Technology

[0002] In underground metal mining operations, upward blast holes are the core engineering structure for medium-deep hole blasting, sector hole blasting, and ore cutting operations, with a wide range of applications. Upward blast holes are generally drilled from the bottom of the roadway or stope upwards, and their key parameters such as azimuth and elevation are determined according to the mining design plan. However, during actual drilling, multiple factors, including drilling rig positioning deviation, drill rod stiffness characteristics, rock mass structure distribution, ore lithology variations, drilling thrust, and drill bit wear, can all cause varying degrees of deviation between the actual drilling trajectory and the designed trajectory.

[0003] Borehole deviation directly alters the actual spacing between adjacent boreholes and the blasting resistance line, leading to uneven charge density. This can result in localized over- or under-charge, creating blasting blind zones and causing a series of production problems, including increased ore block ratios, mine dilution, and higher loss indicators. Severe borehole deviation can also cause boreholes to penetrate adjacent stopes, goafs, and non-blasting control areas, significantly increasing the safety risks of blasting operations. Therefore, accurately obtaining the three-dimensional spatial trajectory of upward-facing boreholes before blasting operations is of significant engineering and production value for optimizing borehole charge structure, verifying blasting design parameters, quantitatively evaluating drilling quality, and ensuring the safe and efficient conduct of mine blasting operations.

[0004] Currently, the mainstream method for measuring borehole trajectories in the industry mainly relies on a measuring probe equipped with an inertial measurement unit (IMU) to reconstruct the trajectory by moving it back and forth along the borehole. This method calculates the probe's real-time attitude by collecting angular velocity and acceleration data during its movement, and then integrates the borehole axis direction using depth measurement data to finally fit the borehole spatial trajectory. Low-cost microelectromechanical inertial sensors (MEMS) are widely used in portable borehole measurement equipment due to their small size, low cost, and high integration. However, these sensors have inherent defects: their zero-bias error, proportional error, installation error, and integral drift error accumulate with the measurement distance, causing a significant positional offset at the bottom of the borehole in a single-pass measurement, making it difficult to guarantee measurement accuracy. In practical engineering measurement scenarios, the probe can complete both the entry measurement from the borehole opening to the bottom and the retraction measurement from the bottom to the opening. Both measurements are performed along the same borehole central axis, theoretically representing completely identical borehole spatial trajectories.

[0005] However, during the borehole entry and retraction measurement processes, the equipment's stress state and motion posture differ, resulting in differential directional deviations and integral drift errors. At the same time, factors such as the extension and retraction of the traction rope, the slippage of the encoder wheel, the friction interference of the borehole wall, and the sudden changes in the probe's motion state can further cause misalignment of the mileage data in the two-way measurement, resulting in the same mileage value not corresponding to the same physical location within the borehole.

[0006] To address the shortcomings of existing technologies, there is an urgent need to provide a novel method for reconstructing upward borehole trajectories. This method relies on the geometric constraints of the shared track between the borehole entry and retraction measurements. Through an integrated optimization solution mechanism, the directional and depth errors of the two-way measurements are simultaneously corrected, resulting in a unique and accurate borehole axis state and three-dimensional center trajectory. This enables high-precision reconstruction of the upward borehole trajectory. Summary of the Invention

[0007] To address the problems existing in the prior art, this invention provides a method for reconstructing the three-dimensional trajectory and measuring the skewness of an upward borehole based on dual-path co-track constraints. This method can effectively reduce the impact of single-path inertial drift, inconsistency in dual-path mileage, and local abnormal observations on the trajectory reconstruction results, and can accurately and efficiently obtain the three-dimensional center trajectory and skewness measurement results of the upward borehole.

[0008] To achieve the above objectives, this invention provides a method for reconstructing and measuring the three-dimensional trajectory of an upward borehole based on dual-path confinement constraints, comprising the following steps: Step 1: Acquisition of measurement parameters; acquire the three-dimensional coordinates of the borehole opening, the initial axial direction of the borehole opening, the designed borehole parameters, and the calibration parameters of the measurement system; Step 2: Two-way measurement data acquisition and preprocessing; Complete the acquisition of measurement data during the two strokes of the probe entering and retracting in the borehole, and perform time synchronization, error compensation, attitude calculation, direction unification and unitization on the raw measurement data, and finally obtain the borehole axis observation sequence and the borehole axis observation sequence arranged in the direction from the borehole opening to the bottom of the borehole. Step 3: Common event identification and monotonic depth mapping establishment; Identify common events corresponding to the same physical location of the borehole during the drilling and retraction processes, establish a monotonic depth mapping from the retraction mileage to the common physical depth using the common events, and determine the initial parameters of the mapping. Step 4: Setting up common hole axis status and establishing two-way observation model; Set up a set of common hole axis statuses in the common physical depth domain, which are jointly corresponding to the hole entry hole axis observation sequence and the hole withdrawal hole axis observation sequence, and establish the hole entry observation model between the common hole axis status and the hole entry hole axis observation sequence, as well as the hole withdrawal observation model between the common hole axis status and the hole withdrawal hole axis observation sequence. Step 5: Generation of two-way borehole shaft prediction sequence and calculation of observation residual; using the borehole entry observation model and withdrawal observation model established in Step 4, generate the borehole entry prediction sequence and withdrawal prediction sequence according to the current common borehole shaft status and measurement error parameters, and calculate the observation residual between the prediction sequence and the corresponding borehole shaft observation sequence; Step 6: Establishment and Iterative Solution of Joint Objective Function; Establish a joint objective function based on observation residuals, common borehole axis state constraints, common event depth correspondence constraints, and measurement error parameter constraints, and iteratively solve the joint objective function; In each iteration, redetermine the common physical depth corresponding to the withdrawn borehole axis observation based on the current monotonic depth mapping parameters, and synchronously update the common borehole axis state, borehole entry measurement error parameters, withdrawn measurement error parameters, and monotonic depth mapping parameters until the preset convergence condition is met; Step 7: Reconstruction of the three-dimensional center trajectory; Based on the common hole axis state, common physical depth and three-dimensional coordinates of the hole opening obtained by joint solution, calculate the coordinates of the center point of the blast hole segment by segment along the direction of increasing hole depth to obtain the three-dimensional center trajectory of the upward blast hole to be measured in the mine coordinate system. Step 8: Calculate the deviation parameters; compare the reconstructed three-dimensional center trajectory with the designed borehole trajectory, calculate the deviation along the borehole, the bottom deviation, and the local borehole axis direction change rate of the borehole to be measured, and quantitatively evaluate the degree of borehole deviation.

[0009] As a preferred embodiment, in step 1, the three-dimensional coordinates of the borehole are used to determine the starting point for reconstructing the three-dimensional center trajectory; the initial axial direction of the borehole serves as a reference for borehole direction constraints, used to constrain the value of the common borehole axis state at the borehole; the designed borehole parameters include the designed borehole depth, designed azimuth angle, and inclination angle, used to establish the designed borehole trajectory and perform skew comparison; the calibration parameters of the measurement system include the zero bias, scale factor, temperature coefficient, and installation error parameters of the inertial sensor, used for compensation of subsequent measurement data.

[0010] As a preferred option, the process of acquiring and preprocessing the two-way measurement data in step 2 is as follows: S21: Static initial calibration; Before the probe enters the upward borehole to be measured, keep the probe stationary, collect the first static measurement data, and determine the inertial sensor zero bias and the probe initial attitude at the start of the measurement based on the first static measurement data; S22: Data acquisition during borehole entry; The probe is advanced from the borehole opening to the bottom of the borehole, and angular velocity data, acceleration data, mileage data, temperature data and measurement status data are acquired simultaneously during the borehole entry process; S23: Retraction Data Acquisition; After the probe reaches the bottom of the hole, the probe is retracted from the bottom of the hole to the opening, and angular velocity data, acceleration data, mileage data, temperature data and measurement status data are collected simultaneously during the retraction process; S24: Time synchronization and error compensation; synchronize the in-hole measurement data and the retraction measurement data according to a unified timestamp, and perform zero offset compensation, scale factor compensation, temperature compensation and installation error compensation according to the measurement system calibration parameters; S25: Attitude calculation and direction conversion; calculate the probe attitude based on the compensated inertial data of the borehole and the inertial data of the retraction, and convert the probe attitude into a three-dimensional orientation observation of the probe mechanical axis in the mine coordinate system; S26: Retraction data rearrangement and orientation unification; rearrange the retraction measurement data according to the direction from the borehole opening to the bottom of the borehole, and unify the orientation of the three-dimensional orientation observation during the retraction process according to the relationship between the direction of the probe's mechanical axis and the direction of the probe's movement. S27: Normalization; The three-dimensional directional observations during the drilling process and the retraction process are normalized to obtain the drilling axis observation sequence and the retraction axis observation sequence, respectively.

[0011] As a preferred embodiment, the process of common event identification and monotonic depth mapping establishment in step 3 is as follows: S31: Determination of orifice and bottom events; determine orifice and bottom events from the entry measurement data and the withdrawal measurement data respectively; S32: Extraction of candidate common events in the borehole; Based on the changes in acceleration, angular velocity, and odometer, extract candidate common events in the borehole during the entry and exit processes, respectively. S33: Initial correspondence establishment; Based on the similarity of characteristics of candidate common events, the order of arrangement along the hole, and the mileage interval between adjacent candidate common events, establish the initial correspondence between candidate common events for entering the hole and candidate common events for withdrawing; S34: Sequence consistency check; Perform a sequence consistency check on the initial correspondence, delete candidate common events whose corresponding order overlaps, feature similarity is lower than a preset threshold, and the difference between adjacent events exceeds a preset range, and obtain a set of common event corresponding points; S35: Establishing a monotonic depth mapping; Based on the set of points corresponding to the common events, establish a monotonic depth mapping from the pullback mileage to the common physical depth; Utilize multiple pairs of mileage-depth correspondences in the set of points corresponding to the common events, and preliminarily determine the proportional parameter and zero-point offset parameter through least squares fitting. S36: Verification and determination of initial mapping parameters; based on the depth correspondence error after the monotonic depth mapping of the orifice event, the proportional parameters and zero-point offset parameters obtained by the preliminary fitting in S35 are verified and corrected to determine the initial parameters of the monotonic depth mapping.

[0012] As a preferred option, the process of setting the common hole shaft state and establishing the two-way observation model in step 4 is as follows: S41: Common Depth Node Setting; Set up multiple common depth nodes in a common physical depth domain according to preset depth intervals; S42: Common hole axis state setting; A common hole axis state is set at each common depth node. The common hole axis state is used to represent the three-dimensional hole axis direction of the upward borehole to be measured at the corresponding common physical depth. S43: Determine the common hole axis direction at the hole entry observation position; based on the physical depth corresponding to the hole entry axis observation, interpolate the common hole axis state at adjacent common depth nodes to obtain the common hole axis direction corresponding to the hole entry axis observation position; S44: Determine the common aperture axis direction of the retreat observation position; Based on the current monotonic depth mapping, convert the retreat mileage corresponding to the retreat aperture axis observation into a common physical depth, and interpolate the common aperture axis state at adjacent common depth nodes to obtain the common aperture axis direction corresponding to the retreat aperture axis observation position. S45: Establishment of the inlet observation model; Establish the inlet observation model based on the common hole axis direction, the inlet direction constant deviation parameter, and the inlet direction drift parameter corresponding to the inlet hole axis observation position; S46: Establishment of the retreat observation model; establish the retreat observation model based on the common hole axis direction corresponding to the retreat hole axis observation position, the retreat direction constant deviation parameter, and the retreat direction drift parameter.

[0013] As a preferred embodiment, the process of generating the two-way borehole shaft prediction sequence and calculating the observation residual in step 5 is as follows: S51: Prediction calculation of the borehole axis; using the borehole observation model, calculate the predicted value of the borehole axis corresponding to each borehole axis observation position based on the common borehole axis status, borehole direction constant deviation parameter and borehole direction drift parameter; S52: Retreat mileage-depth conversion; using the current monotonic depth mapping, convert the retreat mileage corresponding to each retreat hole axis observation into a common physical depth; S53: Calculation of retraction hole axis prediction; Using the retraction observation model, calculate the predicted value of the retraction hole axis corresponding to each retraction hole axis observation position based on the mapped common physical depth, common hole axis state, retraction direction constant deviation parameter and retraction direction drift parameter. S54: Calculation of the residual of the hole entry observation; Calculate the residual of the hole entry observation between the predicted value of the hole shaft and the observed value of the hole shaft. S55: Calculation of the residual of the pullback observation; Calculate the residual of the pullback observation between the predicted value of the pullback hole axis and the observed value of the pullback hole axis.

[0014] As a preferred option, the process of establishing and iteratively solving the joint objective function in step 6 is as follows: S61: Establishment of two-way observation residuals; establish two-way observation residuals based on the borehole entry observation residuals and the withdrawal observation residuals; S62: Establishment of common hole-axis continuity constraint terms; establish common hole-axis continuity constraint terms based on the directional changes between the common hole-axis states at adjacent common depth nodes; S63: Establish unit length and orifice direction constraint terms; establish unit length constraint terms based on the modulus of the common hole shaft state, and establish orifice direction constraint terms based on the initial axial direction of the orifice; S64: Establishment of common event depth correspondence constraint terms; Establish common event depth correspondence constraint terms based on the depth difference after mapping common events through the current monotonic depth; S65: Establish parameter constraint items; establish parameter constraint items based on the allowable variation range of the hole inlet direction constant deviation parameter, hole inlet direction drift parameter, retraction direction constant deviation parameter, retraction direction drift parameter, proportional parameter and zero point offset parameter; S66: Combination of joint objective functions; combining the two-way observation residual terms, common borehole axis continuity constraints, unit length constraints, borehole direction constraints, common event depth corresponding constraints, and parameter constraints into a joint objective function; S67: Joint variable setting; setting joint variables including common hole axis status, hole inlet direction constant deviation parameter, hole inlet direction drift parameter, retraction direction constant deviation parameter, retraction direction drift parameter, proportional parameter and zero point offset parameter; S68: Nonlinear Least Squares Iterative Solution; The joint objective function is solved iteratively using a nonlinear least squares method. In each iteration, the common physical depth corresponding to all backtracking observations is first redefined based on the current scale parameter and zero offset parameter. Then, the common borehole axis state at adjacent common depth nodes is interpolated based on the redefined common physical depth, and the predicted backtracking borehole axis values ​​corresponding to each backtracking borehole axis observation position are recalculated. Subsequently, the backtracking observation residuals are recalculated. Simultaneously, the predicted borehole axis values ​​and the borehole observation residuals corresponding to each borehole axis observation position are recalculated based on the updated common borehole axis state. Finally, the updated joint objective function is minimized using a nonlinear least squares method, and all joint variables to be solved are corrected simultaneously.

[0015] S69: Convergence Judgment and Result Output; Determine whether the change in the joint objective function and the change in the joint variable to be solved between two adjacent iterations meet the preset convergence conditions; If not, continue iterating; If so, output the common hole axis state, hole inlet direction constant deviation parameter, hole inlet direction drift parameter, pullback direction constant deviation parameter, pullback direction drift parameter, scale parameter and zero point offset parameter. As a preferred embodiment, the three-dimensional center trajectory reconstruction process in step 7 is as follows: S71: Obtain the common hole-axis state; Obtain the common hole-axis state after joint solution at each common depth node; S72: Determination of average hole axis direction; Determine the average hole axis direction of the corresponding depth interval based on the common hole axis status at two adjacent common depth nodes; S73: Node coordinates are calculated segment by segment; taking the three-dimensional coordinates of the borehole as the starting point of the three-dimensional center trajectory, the three-dimensional coordinates of the next common depth node are calculated segment by segment according to the length of each depth interval and the average borehole axis direction. S74: Trajectory traversal output; sequentially traverse all common depth nodes to obtain the three-dimensional center trajectory of the upward borehole to be tested in the mine coordinate system. As a preferred embodiment, the skew parameter calculation process in step 8 is as follows: S81: Design trajectory establishment; Establish the design borehole trajectory based on the design borehole parameters, and determine the design trajectory point at the same common physical depth as the reconstructed 3D center trajectory; S82: Spatial Deviation Vector Calculation; Calculates the spatial deviation vector between the reconstructed trajectory point and the design trajectory point at each common physical depth; S83: Calculation of friction deviation; Project the spatial deviation vector onto a plane perpendicular to the direction of the corresponding design hole axis to obtain the lateral deviation at each common physical depth, and use the lateral deviation as the friction deviation; S84: Hole bottom deviation calculation; The total hole bottom deviation is determined based on the spatial distance between the reconstructed trajectory point and the design trajectory point at the hole bottom, and the lateral deviation of the hole bottom is determined based on the projection of the hole bottom spatial deviation vector onto the plane perpendicular to the design hole axis. S85: Calculation of local hole axis direction change rate; Calculate the included angle between the common hole axis directions at adjacent common depth nodes, and determine the local hole axis direction change rate based on the ratio of the included angle to the hole distance between adjacent common depth nodes.

[0016] To address the problems in existing upward borehole trajectory measurement methods, such as the easy accumulation of single-journey inertial errors with borehole depth, the difficulty in accurately corresponding the depth positions of borehole entry and withdrawal data, and the ineffectiveness of simple averaging of two-journey data in eliminating measurement errors, this invention provides a method for reconstructing and measuring the three-dimensional trajectory and deviation of upward boreholes based on two-journey parallel track constraints. This method utilizes the geometric condition that the probe passes through the same borehole centerline during entry and withdrawal, and jointly solves for the common borehole axis state, two-journey measurement error parameters, and monotonic depth mapping parameters within a common physical depth domain, achieving accurate measurement of the three-dimensional center trajectory and deviation of the upward borehole. Compared with existing technologies, this invention has the following technical advantages: 1. This invention identifies common events corresponding to the same physical location of the borehole from the borehole entry and withdrawal measurement data, establishes an affine monotonic depth mapping including proportional parameters and zero-point offset parameters, and corrects the withdrawal mileage to a common physical depth reference consistent with the borehole entry side, thus solving the problem of inconsistent depth caused by factors such as odometer slippage and borehole wall friction in two-way data.

[0017] 2. This invention utilizes the geometric condition that the borehole measurement and the retraction measurement pass through the same borehole centerline, sets common depth nodes at preset depth intervals within a common physical depth domain, and establishes a unique set of common borehole axis states at each node. By interpolation, the common borehole axis direction corresponding to each borehole and retraction observation position is obtained, avoiding the error accumulation caused by simple averaging after reconstructing the two-way trajectory separately.

[0018] 3. This invention establishes an entry observation model and a withdrawal observation model, respectively. The entry measurement error parameters include the constant deviation parameter of the three-dimensional direction of the entry hole and the drift parameter of the three-dimensional direction of the entry hole that varies linearly with depth. The withdrawal measurement error parameters include the constant deviation parameter of the three-dimensional direction of the withdrawal hole and the drift parameter of the three-dimensional direction of the withdrawal hole that varies linearly with depth. The common hole axis state, the entry measurement error parameters, the withdrawal measurement error parameters, and the monotonic depth mapping parameters are incorporated into the same joint solution process, so that the direction error, depth misalignment, and the true hole axis state can be mutually corrected.

[0019] 4. In each iteration, this invention redetermines the common physical depth corresponding to all backtracking observations based on the current scaling parameters and zero-point offset parameters. Based on this, it interpolates the common borehole axis state and recalculates the predicted value of the backtracking borehole axis and the residual of the backtracking observation. At the same time, it recalculates the predicted value of the borehole axis and the residual of the borehole observation based on the updated common borehole axis state, so that the depth mapping and borehole axis state reconstruction form a dynamic coupling, avoiding erroneous observation fusion caused by fixed depth registration results.

[0020] 5. This invention combines the two-way observation residual term, the common hole axis continuity constraint term, the unit length constraint term, the hole opening direction constraint term, the common event depth correspondence constraint term, and the parameter constraint term into a joint objective function, and solves iteratively through the nonlinear least squares method. The stability and accuracy of the trajectory reconstruction results are improved through the synergistic effect of multiple constraints.

[0021] 6. This invention determines the observation weights of each entry hole axis observation and withdrawal hole axis observation based on the local variance of angular velocity, the degree of deviation of acceleration modulus from gravitational acceleration, and the degree of deviation of mileage increment from the average mileage increment of adjacent sampling points. At the same time, the Huber loss function is used to replace the squared loss in the two-way observation residual term, which effectively reduces the impact of abnormal observations on the trajectory reconstruction results.

[0022] 7. This invention can obtain the three-dimensional center trajectory of the upward borehole, as well as the deviation along the path, the deviation at the bottom of the hole, and the rate of change of the local hole axis direction, providing a measurement basis for evaluating the construction quality of boreholes, verifying charging parameters, and optimizing blasting design.

[0023] This method utilizes the geometric redundancy of the probe repeatedly passing through the same borehole centerline during entry and exit, through common event-driven depth mapping, two-way separation error modeling, dynamic iterative joint optimization, and multi-constraint collaborative solution. It effectively reduces the impact of single-way inertial drift, two-way mileage inconsistency, and local anomaly observations on the trajectory reconstruction results, and can accurately and efficiently obtain the three-dimensional center trajectory and skew measurement results of the upward borehole. Attached Figure Description

[0024] Figure 1 This is a flowchart of the measurement method in this invention; Figure 2 This is a schematic diagram of the two-way measurement common event matching and depth mapping of the present invention; Figure 3 This is a schematic diagram of the dual-track joint trajectory reconstruction solution of the present invention; Figure 4 This is a schematic diagram illustrating the double-pass hole shaft direction consistency correction of the present invention; Figure 5 This is a comparison diagram of the three-dimensional trajectory reconstruction results of the boreholes in this invention. Detailed Implementation

[0025] The present invention will be further described below with reference to the accompanying drawings and embodiments. For example... Figure 1 As shown, this invention sequentially performs the following steps: borehole information acquisition, two-way measurement data acquisition and preprocessing, common event identification, monotonic depth mapping initialization, common borehole axis state setting, two-way observation model establishment, joint objective function construction and iterative solution, borehole three-dimensional center trajectory reconstruction, and borehole deviation result calculation. The following embodiments are used to illustrate the specific implementation process, parameter settings, joint solution method, and result output of the method described in this invention, and are not intended to limit the scope of protection of this invention.

[0026] like Figure 1 As shown, this invention provides a method for reconstructing and measuring the three-dimensional trajectory of an upward borehole based on a two-way parallel track constraint, comprising the following steps: Step 1: Acquisition of measurement parameters; acquire the three-dimensional coordinates of the borehole opening, the initial axial direction of the borehole opening, the designed borehole parameters, and the calibration parameters of the measurement system; The three-dimensional coordinates of the borehole are used to determine the starting point for reconstructing the three-dimensional center trajectory; the initial axial direction of the borehole serves as a reference for borehole direction constraints, used to constrain the value of the common borehole axis state at the borehole; the designed borehole parameters include the designed borehole depth, designed azimuth angle, and dip angle, used to establish the designed borehole trajectory and perform skew comparison; the calibration parameters of the measurement system include the zero bias, scale factor, temperature coefficient, and installation error parameters of the inertial sensor, used for compensation of subsequent measurement data; In this technical solution, by acquiring the three-dimensional coordinates of the borehole and the initial axial direction of the borehole, a priori reference for the spatial starting point and borehole direction constraint is provided for the reconstruction of the three-dimensional center trajectory. At the same time, the design borehole parameters and the measurement system calibration parameters such as the zero bias, scale factor, temperature coefficient and installation error of the inertial sensor are acquired, laying the foundation for subsequent design trajectory establishment, skew comparison and systematic error compensation of measurement data.

[0027] Step 2: Two-way measurement data acquisition and preprocessing; Complete the acquisition of measurement data during the two strokes of the probe entering and retracting in the borehole, and perform time synchronization, error compensation, attitude calculation, direction unification and unitization on the raw measurement data, and finally obtain the borehole axis observation sequence and the borehole axis observation sequence arranged in the direction from the borehole opening to the bottom of the borehole. As a preferred option, the process of acquiring and preprocessing the two-way measurement data in step 2 is as follows: S21: Static initial calibration; Before the probe enters the upward borehole to be measured, the probe is kept stationary, the first static measurement data is collected, and the inertial sensor zero bias and the probe initial attitude at the start of the measurement are determined based on the first static measurement data; During static acquisition, the probe is stationary, the accelerometer only senses gravity, and the gyroscope angular velocity is zero. Based on this, the initial zero bias of the accelerometer and gyroscope can be estimated, and the initial attitude of the probe coordinate system relative to the mine coordinate system can be determined using the gravity vector and the geomagnetic vector, providing the initial integral value for subsequent attitude calculation; S22: Hole entry data acquisition; the probe is advanced from the borehole opening to the bottom of the borehole, and angular velocity data, acceleration data, mileage data, temperature data and measurement status data are acquired simultaneously during the hole entry process; during the advancement process, the probe moves along the borehole axis, the inertial sensor records the angular motion and linear motion information of the probe in real time, the odometer records the advancement distance of the probe, the temperature sensor records the ambient temperature for temperature compensation, and the measurement status data is used to identify abnormal states in the acquisition process; S23: Retraction data acquisition; After the probe reaches the bottom of the hole, the probe is retracted from the bottom of the hole to the opening of the hole, and angular velocity data, acceleration data, mileage data, temperature data and measurement status data are collected simultaneously during the retraction process; The retraction stroke and the hole entry stroke move in opposite directions along the same borehole path, forming a "double-track" measurement condition, which provides redundant observation information for subsequent double-track constraints; S24: Time Synchronization and Error Compensation; Time synchronization of in-hole measurement data and retraction measurement data is performed according to a unified timestamp, and zero bias compensation, scaling factor compensation, temperature compensation, and installation error compensation are performed according to the measurement system calibration parameters; Time synchronization aligns inertial data, mileage data, and temperature data under a unified time reference; Zero bias compensation eliminates sensor constant drift, scaling factor compensation eliminates scaling factor error, temperature compensation eliminates drift changes caused by temperature, and installation error compensation eliminates the deviation between the sensor mounting axis and the probe mechanical axis; S25: Attitude calculation and direction conversion; The probe attitude is calculated based on the compensated inertial data of the entry and retraction, and the probe attitude is converted into a three-dimensional orientation observation of the probe mechanical axis in the mine coordinate system; Using the compensated angular velocity and acceleration data, the attitude matrix of the probe coordinate system relative to the mine coordinate system is recursively calculated through the attitude update algorithm, and then combined with the fixed direction of the probe mechanical axis in the probe coordinate system, it is converted to the mine coordinate system to obtain the three-dimensional orientation observation of the probe mechanical axis; S26: Retraction Data Rearrangement and Orientation Unification; The retraction measurement data is rearranged according to the direction from the borehole opening to the bottom of the borehole, and the three-dimensional orientation observations during the retraction process are unified according to the relationship between the probe mechanical axis direction and the probe movement direction; When entering the borehole, the probe movement direction is from the borehole opening to the bottom of the borehole, and the probe mechanical axis points to the bottom of the borehole, so its three-dimensional orientation observation represents the borehole axis direction from the borehole opening to the bottom of the borehole; When retracting, the probe movement direction is from the bottom of the borehole to the borehole opening, and the probe mechanical axis points to the borehole opening, so its three-dimensional orientation observation is opposite to the borehole axis direction from the borehole opening to the bottom of the borehole; Therefore, after rearranging the retraction data in the order from the borehole opening to the bottom of the borehole, the three-dimensional orientation observations of the retraction are reversed to unify them with the borehole entry observations as the borehole axis direction from the borehole opening to the bottom of the borehole. Simultaneously, based on the difference between the mileage value corresponding to the start of the retraction (i.e., when the probe leaves the bottom of the hole) and the mileage value corresponding to the end of the retraction (i.e., when the probe returns to the hole opening), the original retraction mileage is converted in reverse to obtain an approximate retraction mileage measured along the arc length of the hole axis with the hole opening as the origin, so that the retraction mileage increases monotonically from the hole opening to the bottom of the hole; the approximate retraction mileage will be further corrected in the monotonic depth mapping established in step 3 to eliminate the mileage error introduced by factors such as odometer slippage and hole wall friction during the retraction process.

[0028] S27: Normalization; The three-dimensional observations during the drilling and retraction processes are normalized to obtain the drilling axis observation sequences for drilling and retraction, respectively; each three-dimensional observation is normalized to a unit vector to eliminate observation amplitude errors, thus obtaining the drilling axis observation sequence. and the observation sequence of the retraction hole axis Both are arranged from the orifice opening to the bottom of the orifice, and are associated with their respective mileage information; In this technical solution, the zero bias and initial attitude of the inertial sensor are determined by static calibration before the probe enters the borehole, providing a reliable initial integral value for attitude calculation. Utilizing the redundant observation condition of "dual-track" movement along the same borehole path during entry and withdrawal, the probe attitude is calculated and converted into a three-dimensional directional observation in the mine coordinate system based on system compensation for multiple aspects such as time synchronization, zero bias, scale factor, temperature, and installation errors. Then, through rearrangement of withdrawal data, unification of directions, and reverse conversion of withdrawal mileage, the dual-track data is unified into an entry borehole axis observation sequence and a withdrawal borehole axis observation sequence arranged in the direction from borehole opening to bottom, with the borehole opening as the origin. This provides standardized input with consistent direction and unified depth benchmark for subsequent dual-track joint optimization.

[0029] Step 3: Common event identification and monotonic depth mapping establishment; Identify common events corresponding to the same physical location of the borehole during the borehole entry and withdrawal processes, establish a monotonic depth mapping from the withdrawal mileage to the common physical depth using the common events, and determine the initial parameters of the mapping to provide a depth alignment basis for the joint processing of subsequent two-way data in the common physical depth domain. As a preferred embodiment, the process of common event identification and monotonic depth mapping establishment in step 3 is as follows: S31: Determination of orifice and bottom events; orifice and bottom events are determined from the orifice entry measurement data and the orifice retraction measurement data, respectively; the orifice event corresponds to the moment when the probe begins to enter the orifice and the moment when the retraction ends and returns to the orifice, and the orifice bottom event corresponds to the moment when the probe reaches the bottom of the orifice and the moment when the retraction begins to leave the bottom of the orifice, which can be identified by sudden changes in mileage value, acceleration impact, and motion status indicators; the orifice and bottom events constitute the boundary anchor points of the depth mapping; S32: Extraction of candidate common events in the borehole; Based on the changes in acceleration, angular velocity, and mileage, candidate common events in the borehole are extracted during the borehole entry and withdrawal processes, respectively; Physical features such as joints, diameter reduction, lithological changes, and borehole wall roughness changes in the borehole will cause local feature changes in the inertial measurement signal and mileage signal when the probe passes through, and corresponding candidate common events are extracted from the data of the borehole entry and withdrawal processes, respectively. S33: Initial Correspondence Establishment; Based on the similarity of features of candidate common events, the order of arrangement along the hole, and the mileage interval between adjacent candidate common events, the initial correspondence between candidate common events entering the hole and candidate common events withdrawing from the hole is established; the similarity of features measures the consistency of the signal morphology of events in two data segments, the arrangement order constraint ensures that the correspondence does not intersect along the hole direction, and the mileage interval constraint ensures that the relative positional relationship between events is consistent; S34: Sequence consistency check; Perform a sequence consistency check on the initial correspondence, delete candidate common events where the corresponding order overlaps, the feature similarity is lower than a preset threshold, or the difference between adjacent events exceeds a preset range, and obtain a set of common event corresponding points; This check eliminates mismatched events and ensures that each pair of events in the set of common event corresponding points reliably corresponds to the same physical location of the blast hole; S35: Establishment of Monotonic Depth Mapping; Based on the set of points corresponding to the common events, establish a monotonic depth mapping from the pullback mileage to the common physical depth; The monotonic depth mapping adopts an affine mapping that includes a scaling parameter and a zero-point offset parameter, i.e. ,in For common physical depth, For the retracement mileage, For proportional parameters, This is the zero-point offset parameter; this mapping satisfies the monotonically increasing property that the common physical depth increases synchronously with the increase of retreat mileage, i.e., the proportional parameter. Using multiple pairs of mileage-depth correspondences in the set of points corresponding to common events, the scale parameter and zero-point offset parameter are initially determined by least squares fitting. S36: Verification and Determination of Initial Mapping Parameters; Based on the depth correspondence errors of the orifice opening event, orifice bottom event, and common event within the orifice after the monotonic depth mapping, the proportional parameters and zero-point offset parameters initially fitted in S35 are verified and corrected to determine the initial parameters of the monotonic depth mapping; wherein, the orifice opening event should satisfy that the common physical depth corresponding to its retreat mileage after mapping is zero, the orifice bottom event should satisfy that the common physical depth corresponding to its retreat mileage after mapping is the orifice depth, and the common event within the orifice should satisfy that its mapped common physical depth is consistent with the physical depth corresponding to the orifice entry side; the proportional parameters and zero-point offset parameters are adjusted according to the depth correspondence errors of the above three types of events, while verifying and ensuring that the proportional parameters are greater than zero to satisfy the monotonically increasing constraint that the common physical depth increases synchronously as the retreat mileage increases, and finally determining the initial mapping parameters that satisfy the minimum depth correspondence error and the monotonically increasing requirement.

[0030] In this technical solution, candidate common events are extracted from the data of both the entry and withdrawal stages by utilizing the local changes in inertial and mileage signals caused by physical features such as joints, borehole diameter reduction, and lithological changes within the borehole as the probe passes through. A reliable correspondence between these events is established through feature similarity, borehole arrangement order, and mileage interval constraints. After a sequence consistency check to eliminate mismatches, a set of corresponding points for the common events is obtained. Based on this set of points, an affine monotonic depth mapping including scaling parameters and zero-point offset parameters is established. The monotonic constraint is that the common physical depth increases synchronously with the withdrawal mileage. The mapping parameters are verified and corrected using the depth correspondence errors of three types of anchor points: borehole opening events, borehole bottom events, and borehole common events. This achieves accurate alignment of the two-way data within the common physical depth domain, laying a unified depth benchmark for subsequent joint estimation of the common borehole axis state.

[0031] Step 4: Setting up common aperture axis states and establishing two-way observation models; In the common physical depth domain, set up a set of common aperture axis states corresponding to both the entry aperture axis observation sequence and the withdrawal aperture axis observation sequence, and establish entry observation models between common aperture axis states and entry aperture axis observation sequences, as well as withdrawal observation models between common aperture axis states and withdrawal aperture axis observation sequences, to lay the state-space foundation for joint estimation of two-way data in the unified physical depth domain; As a preferred option, the process of setting the common hole shaft state and establishing the two-way observation model in step 4 is as follows: S41: Common depth node setting; Multiple common depth nodes are set in the common physical depth domain according to a preset depth interval; The preset depth interval is determined comprehensively based on the measurement accuracy requirements and calculation efficiency, and the common depth nodes are evenly distributed along the common physical depth, covering the entire borehole range from the borehole opening to the bottom of the borehole. S42: Common Hole Axis State Setting; A common hole axis state is set at each common depth node. The common hole axis state is used to represent the three-dimensional hole axis direction of the upward borehole to be measured at the corresponding common physical depth. The common hole axis state is represented by a three-dimensional unit tangent vector in the mining coordinate system. The three-dimensional unit tangent vector includes two horizontal components and one vertical component. Its magnitude is constrained to 1 by the unit length constraint term in the joint solution and is an unknown quantity to be solved. All of them constitute the hole axis direction field to be reconstructed.

[0032] S43: Determining the common borehole axis direction at the borehole entry observation position; Based on the physical depth corresponding to the borehole axis observation, interpolate the common borehole axis state at adjacent common depth nodes to obtain the common borehole axis direction corresponding to the borehole axis observation position; The physical depth of the borehole axis observation is its borehole mileage with the borehole opening as the origin, thereby determining the position of the observation between common depth nodes, and obtaining the common borehole axis direction at this position through linear interpolation or spline interpolation; The borehole mileage is directly used as the reference benchmark for the common physical depth due to the forward advancement of the probe and the high reliability of the odometer; The retraction mileage needs to be corrected to this reference benchmark through monotonic depth mapping due to reverse movement and possible slippage; S44: Determine the common aperture axis direction of the retreat observation position; Based on the current monotonic depth mapping, convert the retreat mileage corresponding to the retreat aperture axis observation into a common physical depth, and interpolate the common aperture axis state at adjacent common depth nodes to obtain the common aperture axis direction corresponding to the retreat aperture axis observation position; During the first execution, the current monotonic depth mapping adopts the initial mapping determined in step 3; In subsequent iterations, the mapping parameters updated in step 6 are used to reconvert the parameters. S45: Establishment of the inlet observation model; An inlet observation model is established based on the common hole axis direction, the inlet direction constant deviation parameter, and the inlet direction drift parameter corresponding to the inlet hole axis observation position; the inlet measurement error parameters include the inlet three-dimensional direction constant deviation parameter and the inlet three-dimensional direction drift parameter that varies linearly with depth; the inlet observation model is expressed as the inlet hole axis observation value equal to the common hole axis direction at the inlet observation position superimposed with the inlet direction constant deviation and the inlet direction drift that varies linearly with depth, i.e. The normalized form, in which The common hole axis direction at the hole observation position. The parameters for the three-dimensional orientation of the inlet hole are the constant deviation parameters. These are the three-dimensional drift parameters for the borehole. This represents the physical depth corresponding to the observation position at the borehole. S46: Establishment of the retraction observation model; A retraction observation model is established based on the common hole axis direction corresponding to the retraction hole axis observation position, the retraction direction constant deviation parameter, and the retraction direction drift parameter; the retraction measurement error parameters include the retraction three-dimensional direction constant deviation parameter and the retraction three-dimensional direction drift parameter that varies linearly with depth; the retraction observation model is expressed as the retraction hole axis observation value equal to the common hole axis direction at the retraction observation position superimposed with the retraction direction constant deviation and the retraction direction drift that varies linearly with depth, i.e. The normalized form, in which To determine the direction of the common aperture axis at the observation location, To retract the constant offset parameters in the three-dimensional direction, To retract the three-dimensional drift parameters, The common physical depth of the retreated observation location after monotonic depth mapping; In this technical solution, common depth nodes are set at preset intervals within a common physical depth domain. At each node, a common borehole axis state represented by a three-dimensional unit tangent vector in the mine coordinate system is established. The common borehole axis direction corresponding to the entry and withdrawal observation positions is obtained through interpolation, enabling the two-way observations to be correlated under a unified spatial reference. Simultaneously, entry and withdrawal observation models are established separately, each containing a three-dimensional direction constant offset parameter and a three-dimensional direction drift parameter that varies linearly with depth. This incorporates the structural differences in the two-way measurement errors into a unified state space, laying the foundation for mutual correction of direction errors and borehole axis states in subsequent joint solutions.

[0033] Step 5: Generation of two-way borehole shaft prediction sequence and calculation of observation residual; Using the borehole entry observation model and withdrawal observation model established in Step 4, generate the borehole entry shaft prediction sequence and withdrawal shaft prediction sequence according to the current common borehole shaft status and measurement error parameters, and calculate the observation residual between the prediction sequence and the corresponding borehole shaft observation sequence to provide data fitting terms for the construction of the joint objective function; As a preferred embodiment, the process of generating the two-way borehole shaft prediction sequence and calculating the observation residual in step 5 is as follows: S51: Prediction calculation of the borehole axis; using the borehole observation model, the predicted value of the borehole axis corresponding to each borehole axis observation position is calculated based on the common borehole axis state, the borehole direction constant deviation parameter, and the borehole direction drift parameter; for each borehole axis observation, the common borehole axis direction at the observation position is first obtained by interpolation of the common borehole axis state, and then the borehole direction constant deviation and borehole direction drift are superimposed to obtain the predicted value of the borehole axis; S52: Retreat mileage-depth conversion; using the current monotonic depth mapping, the retreat mileage corresponding to each retreat hole axis observation is converted into a common physical depth; this step is re-executed in each iteration according to the current scale parameter and zero offset parameter, so that the position of the retreat observation in the depth domain is dynamically adjusted as the mapping parameter is updated; S53: Calculation of retraction hole axis prediction; Using the retraction observation model, the predicted value of the retraction hole axis corresponding to each retraction hole axis observation position is calculated based on the mapped common physical depth, common hole axis state, retraction direction constant deviation parameter, and retraction direction drift parameter; For each retraction hole axis observation, its position between common depth nodes is first determined by the mapped common physical depth and the common hole axis direction is obtained by interpolation, and then the retraction direction constant deviation and retraction direction drift are superimposed to obtain the predicted value of the retraction hole axis; S54: Calculation of the residual of the hole observation; calculate the residual of the hole observation between the predicted value of the hole shaft and the observed value of the hole shaft; the residual of the hole observation is the directional difference between the predicted value of the hole shaft and the observed value of the hole shaft, expressed in the form of vector difference or spherical distance; S55: Calculation of the withdrawal observation residual; Calculate the withdrawal observation residual between the predicted value of the withdrawal hole axis and the observed value of the withdrawal hole axis; The withdrawal observation residual is the directional difference between the predicted value of the withdrawal hole axis and the observed value of the withdrawal hole axis, and its form is consistent with the hole entry observation residual. In this technical solution, based on the entry and exit observation models, the current common borehole axis state and measurement error parameters are used to generate entry and exit borehole axis prediction sequences, respectively. By calculating the directional residuals between the predicted values ​​and the corresponding observed values, the redundant observation information is quantified into a unified data fitting term, providing a driving basis for the construction of the joint objective function. Simultaneously, the exit mileage-depth conversion is dynamically updated with the mapping parameters in each iteration, ensuring that the calculation of the exit prediction values ​​and residuals is coupled with the depth mapping, thus ensuring the consistency between depth alignment and borehole axis state estimation in the joint optimization.

[0034] Step 6: Establishment and Iterative Solution of Joint Objective Function; Establish a joint objective function based on observation residuals, common borehole axis state constraints, common event depth correspondence constraints, and measurement error parameter constraints, and iteratively solve the joint objective function; in each iteration, redetermine the common physical depth corresponding to the withdrawn borehole axis observation based on the current monotonic depth mapping parameters, and synchronously update the common borehole axis state, borehole entry measurement error parameters, withdrawn measurement error parameters, and monotonic depth mapping parameters until the preset convergence condition is met, thus achieving joint optimization estimation of two-way data; As a preferred option, the process of establishing and iteratively solving the joint objective function in step 6 is as follows: S61: Establishment of Two-Way Observation Residual Term; A two-way observation residual term is established based on the entry observation residual and the withdrawal observation residual; the two-way observation residual term is the weighted sum of squares of the entry observation residual and the withdrawal observation residual according to their respective observation weights; the observation weights of each entry hole axis observation and withdrawal hole axis observation are determined based on the local variance of angular velocity, the degree of deviation of acceleration modulus from gravitational acceleration, and the degree of deviation of mileage increment from the average mileage increment of adjacent sampling points: the larger the local variance of angular velocity, the more violent the probe movement and the less reliable the attitude calculation, and the lower the weight; the farther the acceleration modulus deviates from gravitational acceleration, the more the probe is subjected to additional impact or friction interference, and the lower the weight; the farther the mileage increment deviates from the average mileage increment of adjacent sampling points, the more the odometer slips or jumps, and the lower the weight; simultaneously, the Huber loss function is used instead of the squared loss in the two-way observation residual term to reduce the impact of abnormal observations on the solution results of the common hole axis state; S62: Establishment of common hole axis continuity constraint term; Establish a common hole axis continuity constraint term based on the directional change between the common hole axis states at adjacent common depth nodes; The borehole axis should change continuously and smoothly in space, and the hole axis direction at adjacent common depth nodes should not change abruptly. This constraint term is measured in terms of the directional angle between adjacent common hole axis states or the sum of their squares, and it penalizes drastic changes in the hole axis direction. S63: Establish unit length and orifice direction constraints; establish unit length constraints based on the modulus of the common orifice shaft state, and establish orifice direction constraints based on the initial axial direction of the orifice; the common orifice shaft state is a three-dimensional unit tangent vector, and the unit length constraints penalize the degree to which its modulus deviates from 1; the orifice direction constraints penalize the deviation between the common orifice shaft state at the orifice and the initial axial direction of the orifice obtained in step 1, and introduce prior direction information into the optimization; S64: Establishment of common event depth correspondence constraint terms; Establish common event depth correspondence constraint terms based on the depth difference of common events after the current monotonic depth mapping; For each pair of common events, the physical depth of the entry side and the common physical depth of the withdrawal side after monotonic depth mapping should be consistent. This constraint term penalizes the depth difference between the two, thereby integrating the depth correspondence of common events into joint optimization. S65: Establishment of parameter constraints; Establish parameter constraints based on the allowable variation range of the borehole direction constant deviation parameter, borehole direction drift parameter, retraction direction constant deviation parameter, retraction direction drift parameter, proportional parameter, and zero point offset parameter; This constraint imposes reasonable range restrictions on the values ​​of each measurement error parameter and monotonic depth mapping parameter to prevent the parameters from diverging to physically infeasible values ​​during iteration; S66: Combination of joint objective functions; combining the two-way observation residual terms, common borehole axis continuity constraints, unit length constraints, borehole direction constraints, common event depth corresponding constraints, and parameter constraints into a joint objective function; the constraints are weighted and summed according to preset weight coefficients to form a unified minimization objective; S67: Setting joint variables to be solved; setting joint variables to be solved including common hole shaft state, hole entry direction constant deviation parameter, hole entry direction drift parameter, retraction direction constant deviation parameter, retraction direction drift parameter, proportional parameter and zero point offset parameter; the proportional parameter and zero point offset parameter are used as joint parameters to be solved, and are updated synchronously with the common hole shaft state, hole entry measurement error parameter and retraction measurement error parameter during the iterative solution process in step 6.

[0035] S68: Nonlinear least squares iterative solution; The joint objective function is solved iteratively using a nonlinear least squares method; In each iteration, the common physical depth corresponding to all backtracking observations is first re-determined based on the current scale parameter and zero offset parameter (i.e., S52 is re-executed), then the common aperture axis state at adjacent common depth nodes is interpolated based on the re-determined common physical depth, and the predicted backtracking aperture axis value corresponding to each backtracking aperture axis observation position is recalculated (i.e., S53 is re-executed), and then the backtracking observation residuals are recalculated (i.e., S55 is re-executed); At that time, based on the updated common hole axis status, the predicted values ​​of the entry hole axis and the entry hole observation residuals corresponding to the observation positions of each entry hole axis are recalculated (i.e., S51 and S54 are re-executed); finally, the updated joint objective function is minimized by the nonlinear least squares method, and all joint variables to be determined, including the common hole axis status, the entry hole direction constant deviation parameter, the entry hole direction drift parameter, the retreat direction constant deviation parameter, the retreat direction drift parameter, the scale parameter and the zero point offset parameter, are simultaneously corrected to achieve synchronous joint update of the monotonic depth mapping parameters, the common hole axis status and the measurement error parameters.

[0036] S69: Convergence Judgment and Result Output; Determine whether the change in the joint objective function and the change in the joint variable to be solved between two adjacent iterations meet the preset convergence conditions; if not, continue iterating; if so, output the common hole axis state, hole entry direction constant deviation parameter, hole entry direction drift parameter, pullback direction constant deviation parameter, pullback direction drift parameter, scale parameter, and zero point offset parameter; the convergence conditions include the change in the joint objective function being less than the preset objective function threshold and the change in the joint variable to be solved being less than the preset variable threshold; In this technical solution, the two-way observation residual term is combined with the common aperture axis continuity constraint, unit length constraint, aperture direction constraint, common event depth correspondence constraint, and parameter constraints into a unified joint objective function. This function is then iteratively solved using a nonlinear least squares method. This achieves simultaneous joint optimization of the common aperture axis state, two-way measurement error parameters, and monotonic depth mapping parameters under the synergistic effect of multiple constraints, effectively suppressing the impact of single-way inertial drift and depth inconsistency on the reconstruction results. Simultaneously, the weights of each observation are adaptively determined based on the local variance of angular velocity, the degree of deviation of acceleration modulus from gravitational acceleration, and the degree of deviation of mileage increment. Furthermore, the Huber loss function is used instead of the squared loss to reduce the interference of abnormal observations on the solution results and improve the robustness of trajectory reconstruction.

[0037] Step 7: Reconstruction of the three-dimensional center trajectory; Based on the common hole axis state, common physical depth and three-dimensional coordinates of the hole opening obtained by joint solution, calculate the coordinates of the center point of the blast hole segment by segment along the direction of increasing hole depth to obtain the three-dimensional center trajectory of the upward blast hole to be measured in the mine coordinate system. As a preferred embodiment, the three-dimensional center trajectory reconstruction process in step 7 is as follows: S71: Obtain the common hole axis state; obtain the common hole axis state after joint solution at each common depth node; the common hole axis state is a three-dimensional unit tangent vector in the mining coordinate system, representing the direction of the borehole axis at each common depth node; S72: Determination of average hole axis direction; Determine the average hole axis direction of the corresponding depth interval based on the common hole axis state at two adjacent common depth nodes; The common hole axis direction of two adjacent nodes is averaged or weighted to obtain the representative direction of the depth interval, which is used to calculate the hole axis extension direction within the interval; S73: Node coordinates are calculated segment by segment; taking the three-dimensional coordinates of the borehole opening as the starting point of the three-dimensional center trajectory, the three-dimensional coordinates of the next common depth node are calculated segment by segment according to the length of each depth interval and the average borehole axis direction; the length of the depth interval is the common physical depth difference between adjacent common depth nodes (i.e., the difference in arc length along the borehole axis), and the three-dimensional coordinates of the next node are equal to the coordinates of the previous node plus the product of the depth interval length and the average borehole axis direction, thereby completing the segment-by-segment extension from the borehole opening to the bottom of the borehole; S74: Trajectory traversal output; sequentially traverse all common depth nodes to obtain the three-dimensional center trajectory of the upward blast hole in the mine coordinate system; connect the three-dimensional coordinates of the starting point of the borehole and each common depth node in sequence to obtain the complete three-dimensional center trajectory of the borehole. In this technical solution, based on the optimal common borehole axis state obtained through joint solution, the three-dimensional coordinates of each node are calculated segment by segment according to the average borehole axis direction and depth interval length between adjacent common depth nodes, starting from the three-dimensional coordinates of the borehole opening. This yields the continuous three-dimensional center trajectory of the borehole from the opening to the bottom. The method uses the borehole axis direction field optimized by double-track constraints and multi-constraint co-optimization as a guide for trajectory integration, effectively avoiding error accumulation caused by directly integrating the original observations, and ensuring the smoothness and accuracy of the reconstructed trajectory.

[0038] Step 8: Calculate the deviation parameters; compare the reconstructed 3D center trajectory with the designed borehole trajectory, calculate the deviation along the borehole, the bottom deviation, and the local borehole axis direction change rate of the borehole to be measured, and quantitatively evaluate the degree of borehole deviation. As a preferred embodiment, the skew parameter calculation process in step 8 is as follows: S81: Design trajectory establishment; establish the design borehole trajectory based on the design borehole parameters, and determine the design trajectory points at the same common physical depth as the reconstructed 3D center trajectory; determine the ideal straight line trajectory based on the design borehole depth, design azimuth, and dip angle, and determine the corresponding design trajectory point positions at each common depth node; S82: Spatial deviation vector calculation; calculate the spatial deviation vector between the reconstructed trajectory point and the design trajectory point at each common physical depth; the spatial deviation vector is the coordinates of the reconstructed trajectory point minus the coordinates of the design trajectory point at the same common physical depth, reflecting the deviation direction and distance of the borehole center at that depth; S83: Calculation of deviation along the borehole; Project the spatial deviation vector onto a plane perpendicular to the direction of the corresponding design borehole axis to obtain the lateral deviation at each common physical depth, and use the lateral deviation as the deviation along the borehole; The deviation along the borehole characterizes the lateral offset of the borehole relative to the design borehole axis at each depth position, and is the main indicator for measuring the degree of borehole deviation. S84: Hole bottom deviation calculation; The total hole bottom deviation is determined based on the spatial distance between the reconstructed trajectory point and the design trajectory point at the hole bottom, and the lateral hole bottom deviation is determined based on the projection of the hole bottom spatial deviation vector onto the plane perpendicular to the design hole axis; The total hole bottom deviation reflects the overall deviation of the hole bottom position, and the lateral hole bottom deviation reflects the offset of the hole bottom in the direction perpendicular to the design hole axis. S85: Calculation of local borehole axis direction variation rate; Calculate the included angle between the common borehole axis directions at adjacent common depth nodes, and determine the local borehole axis direction variation rate based on the ratio of the included angle to the distance along the borehole between adjacent common depth nodes; The local borehole axis direction variation rate characterizes the degree of bending of the borehole axis direction along the borehole depth, and is used to evaluate the severity of local bending of the borehole axis.

[0039] In this technical solution, the reconstructed three-dimensional center trajectory is compared with the designed borehole trajectory under a unified depth reference. The spatial deviation vector is calculated and decomposed into lateral deviation along the path, total borehole bottom deviation, and lateral deviation. At the same time, the local borehole axis direction change rate is determined based on the ratio of the included angle of adjacent common borehole axis states to the distance along the borehole. The degree of borehole deviation is comprehensively evaluated from three dimensions: borehole position offset, borehole bottom deviation, and borehole axis bending. This provides a quantitative basis for borehole construction quality evaluation, charge parameter verification, and blasting design optimization.

[0040] The present invention will be further described below with reference to specific embodiments: Example 1: Case study of three-dimensional trajectory reconstruction of upward borehole with two-way parallel track; This embodiment is a specific implementation of the method described in steps 1 to 7, and fully demonstrates the entire process from measurement parameter acquisition, two-way data acquisition and preprocessing, common event identification and monotonic depth mapping establishment, common hole axis state setting and two-way observation model establishment, two-way prediction and residual calculation, joint objective function establishment and iterative solution to three-dimensional center trajectory reconstruction.

[0041] (a) Acquisition of measurement parameters; A completed upward blast hole in an underground metal mine was selected for measurement. The designed borehole depth was 30m, and the three-dimensional coordinates of the borehole opening were (1000.000m, 2000.000m, -350.000m). The designed azimuth angle was 35°, and the designed elevation angle was 55°. A three-axis gyroscope, a three-axis accelerometer, a temperature sensor, and an odometer measurement unit were installed inside the probe. The sampling frequency for angular velocity and acceleration was set to 100Hz, and the temperature sampling frequency was set to 10Hz. Odometer data was recorded in increments not exceeding 0.010m.

[0042] (ii) Two-way measurement data acquisition and preprocessing; Execute S21. Before the probe enters the borehole, it remains stationary at the borehole opening for 10 seconds. The initial zero bias of the gyroscope is determined based on the average angular velocity during the stationary phase. The initial attitude of the probe is determined based on the direction of gravity. The attitude result is converted into the borehole axis direction in the mining coordinate system by using the calibration relationship between the inertial sensor coordinate system and the probe mechanical axis.

[0043] After completing static calibration by executing steps S22 to S23, continuously advance the probe from the borehole opening to the bottom of the borehole at a speed controlled between 0.10 and 0.25 m / s. After reaching the bottom of the borehole, keep the probe stationary for 3 seconds, and then retract it from the bottom of the borehole to the borehole opening at a speed of 0.10 to 0.30 m / s. During both the entry and retraction processes, collect angular velocity, acceleration, temperature, mileage, and measurement status data simultaneously.

[0044] Perform steps S24 to S27 to synchronize the two-way measurement data using a unified timestamp, zero-bias compensation, scale factor compensation, temperature compensation, and installation error compensation. Calculate the borehole entry and withdrawal attitudes separately. Reverse the order of the withdrawal measurement data and unify the borehole axis direction, ensuring that both the borehole entry and withdrawal axis observation sequences are arranged from borehole opening to bottom. Resample the two-way borehole axis observations at a uniform mileage interval of 0.10m.

[0045] The observation weight determination method described in S61 is executed. The observation weights are determined based on the local variance of angular velocity, the degree of deviation of acceleration modulus from gravitational acceleration, and the degree of change in mileage increment. The weight of normal observation points is set to 1.00, the weight of slightly abnormal observation points is set to 0.40, and the weight of obviously abnormal observation points is set to 0.12.

[0046] (III) Common event identification and monotonic depth mapping establishment; Execute S31 to S34, such as Figure 2As shown, the acceleration, angular velocity, and mileage changes in the entry and exit data are calculated separately. These three indicators are then normalized and weighted, with the acceleration change indicator weighted at 0.40, the angular velocity change indicator at 0.35, and the mileage change indicator at 0.25. The candidate common event response threshold is set to 0.60, and the minimum mileage interval between adjacent candidate events is set to 1.0m. A two-way event correspondence is established based on the similarity of the candidate event response waveforms, the similarity of local borehole axis direction changes, and the consistency of the distance between adjacent events. The comprehensive matching score threshold is set to 0.65, and incorrect matches with overlapping order or a difference in distance between adjacent events exceeding 20% ​​are deleted. This embodiment ultimately retains the borehole opening event, the borehole bottom event, and four common events within the borehole, with common physical depths of 0.0m, 5.2m, 11.4m, 18.1m, 24.8m, and 30.0m, respectively.

[0047] Execute steps S35 to S36 to establish an affine mapping from the pullback mileage to the common physical depth based on the points corresponding to the aforementioned common events. proportional parameters The initial search range is set to 0.990–1.010, with zero-point offset parameters. The initial search range was set to -0.10 to 0.10 m, and the initial parameters for depth mapping were obtained by using the corresponding points of the common events through least squares fitting and verification and correction of the anchor points at the orifice and the bottom of the orifice.

[0048] (iv) Setting up the common hole shaft status and establishing the two-way observation model; After executing steps S41 and S42 to complete the depth mapping initialization, 61 common depth nodes are set at 0.50m intervals within a common physical depth range of 0–30m. Each common depth node sets a three-dimensional unit borehole axis state in the mine coordinate system. The borehole entry observation and retreat observation do not establish independent trajectories but instead correspond to the aforementioned 61 common borehole axis states.

[0049] Execute steps S43 to S46. In the borehole observation model, set the borehole 3D direction constant deviation parameter and the borehole 3D direction drift parameter that varies linearly with the borehole depth. In the pullback observation model, set the pullback 3D direction constant deviation parameter and the pullback 3D direction drift parameter that varies linearly with the borehole depth. The initial range of each component of the direction constant deviation parameter is set to -0.010 to 0.010, and the initial range of each component of the direction drift parameter is set to -0.0005 to 0.0005m. -1 .

[0050] like Figure 3As shown, the borehole axis observation is interpolated to the adjacent common borehole axis state according to its common physical depth, while the borehole axis observation is converted to the common physical depth according to the current scale parameter and zero offset parameter, and then interpolated to the adjacent common borehole axis state. The borehole axis prediction value is jointly generated by the common borehole axis state, the borehole direction constant offset parameter, and the borehole direction drift parameter, while the borehole axis prediction value is jointly generated by the common borehole axis state, the borehole direction constant offset parameter, the borehole direction drift parameter, and the current depth mapping parameter.

[0051] (v) Two-way prediction and observation residual calculation; Execute steps S51 to S55 to generate the in-hole axis prediction sequence and the back-hole axis prediction sequence based on the current common hole axis status, measurement error parameters, and monotonic depth mapping parameters, and calculate the in-hole observation residual and back-hole observation residual between the predicted values ​​and the corresponding observed values.

[0052] (vi) Establishment and iterative solution of the joint objective function; Execute steps S61 to S67 to establish a joint objective function based on the aforementioned two-way prediction model. The joint objective function includes two-way observation residual terms, common borehole axis continuity constraint terms, unit length constraint terms, borehole orifice direction constraint terms, common event depth correspondence constraint terms, and parameter constraint terms. The weights of each constraint term are set as follows: two-way observation residual term weight 1.00, common borehole axis continuity constraint weight 0.50, unit length constraint weight 2.00, borehole orifice direction constraint weight 6.00, common event depth correspondence constraint weight 12.00, direction constant deviation parameter constraint weight 4.00, direction drift parameter constraint weight 80.00, proportional parameter constraint weight 5.00, and zero point offset parameter constraint weight 2.00.

[0053] S68 to S69 are executed, employing a nonlinear least squares method for joint solution. A Huber robust loss function with a scaling parameter of 0.0025 is applied to the two-way observation residuals, and the maximum number of function calculations is set to 220. In each iteration, the common physical depth corresponding to all pullback observations is first re-determined based on the current scaling parameter and zero-point offset parameter. Then, based on the updated depth position interpolation of the common borehole axis state, the predicted pullback borehole axis value and pullback observation residuals are recalculated. Simultaneously, the predicted borehole axis value and borehole observation residuals are also recalculated. Subsequently, the common borehole axis state, borehole direction constant bias parameter, borehole direction drift parameter, pullback direction constant bias parameter, pullback direction drift parameter, scaling parameter, and zero-point offset parameter are updated synchronously until the relative change in the joint objective function between two adjacent iterations is less than 1 × 10⁻⁶. -8 Furthermore, the relative change of the joint variables to be determined is less than 1×10. -7 .

[0054] (vii) Reconstruction of the three-dimensional center trajectory; After executing S71 to S74 and completing the joint solution, starting from the three-dimensional coordinates of the borehole (1000.000m, 2000.000m, -350.000m), the average borehole axis direction is determined based on the common borehole axis unit direction at adjacent common depth nodes. The three-dimensional coordinate increments are calculated segment by segment according to the length of each depth interval and accumulated to obtain the three-dimensional center trajectory of the upward borehole.

[0055] Example 2: Numerical simulation verification of the two-way co-track joint reconstruction method; This embodiment verifies the effectiveness of the dual-path co-track joint reconstruction method described in steps 1 to 8 of the present invention through numerical simulation, and compares it with the single-path integration method and the dual-path simple averaging method.

[0056] (a) Simulation of borehole trajectory and generation of measurement data; A 30m deep upward-facing borehole is designed with borehole coordinates (1000.000m, 2000.000m, -350.000m), a designed azimuth of 35°, and a designed elevation of 55°. The azimuth of the simulated real borehole gradually increases by 1.6° along the borehole depth, superimposed with a smooth periodic change of 0.35°. Similarly, the elevation of the simulated real borehole gradually decreases by 1.2° along the borehole depth, superimposed with a smooth change of 0.25°, creating a continuous three-dimensional curvature in the borehole.

[0057] Perform the methods described in steps 1 to 4, setting the common depth node interval to 0.50m, with a total of 61 common hole axis status nodes. 181 sampling points are set for both hole entry and pullback observations, with an average spacing of approximately 0.167m between adjacent observation points. The hole entry 3D direction constant deviation parameter is set to (0.0018, -0.0012, 0.0007), and the hole entry 3D direction drift parameter is set to (0.000055, -0.000035, 0.000020)m. -1 The retraction three-dimensional direction constant deviation parameters are set to (-0.0015, 0.0010, -0.0005), and the retraction three-dimensional direction drift parameters are set to (-0.000045, 0.000030, -0.000018) m. -1 The standard deviation of random directional noise was set to 3.5 × 10⁻⁶. -4 .

[0058] Setting a proportional error and zero-point offset between the pullback mileage and the common physical depth, the actual proportional parameter. Set to 1.0045, the true zero-point offset parameter. The value was set to -0.045m. Large-amplitude directional disturbances were superimposed at the 42nd, 105th, and 146th observation positions in the borehole observation sequence and the 31st, 92nd, and 151st observation positions in the retreat observation sequence to simulate probe collisions, jamming, and local mileage anomalies.

[0059] (ii) Common event identification and depth mapping initialization; Perform the method described in step 3, and fit the initial depth mapping using six common events (including the orifice event and the orifice bottom event) at 0.0m, 5.2m, 11.4m, 18.1m, 24.8m and 30.0m, and continue to use the scale parameter and zero offset parameter as variables to be solved in step 6.

[0060] (III) Joint optimization and effect verification of dual-track operation; After completing the joint optimization of the two-way same track using the method described in Example 1, the difference in the observation direction of the two-way borehole axis before and after the joint correction is calculated. For example... Figure 4 As shown, before joint correction, there was a significant directional difference between the entry and exit observations at the same common physical depth, and local fluctuations occurred at the location of abnormal disturbances. After joint optimization using the method of this invention, by synchronously updating the common borehole axis state, the two-way direction error parameter, and the depth mapping parameter, the two-way borehole axis direction difference at the same common physical depth was significantly reduced. This indicates that the method of this invention can effectively eliminate the two-way measurement inconsistency caused by mileage mapping error, sensor direction deviation, and error accumulation. The proportional parameter obtained by joint solution is 1.004298, and the zero-point offset parameter is -0.039599m, which is close to the set true parameters.

[0061] (iv) Comparison of 3D trajectory reconstruction accuracy; To further verify the accuracy of the 3D trajectory reconstruction, the following three methods were used for trajectory reconstruction: Single-pass integration method for entering the hole: integration is performed segment by segment using only the observation of the hole axis at the entrance; Two-way simple averaging method: After performing depth transformation on the retreat observations using the initial depth mapping, the entry and retreat observations at the same depths are averaged. The present invention provides a dual-path co-tracking method for simultaneously solving the common hole axis state, dual-path direction error parameters, and depth mapping parameters.

[0062] Since the borehole length reaches 30m, the deviation between different reconstruction methods is on the order of centimeters. If drawn directly in the mine's global coordinate system according to the actual scale, visual overlap will occur between different trajectories. Therefore, this embodiment establishes a local orthogonal coordinate system based on the design borehole axis direction, and transforms the 3D trajectory into this local coordinate system for display. The coordinates along the design borehole axis direction maintain the true scale, while the two lateral coordinates perpendicular to the design borehole axis direction are uniformly enlarged by 20 times, solely for highlighting the centimeter-level deviations between different methods.

[0063] like Figure 5As shown, the solid black line represents the simulated real trajectory, the orange dotted line represents the single-pass reconstructed trajectory at the borehole, the blue dashed line represents the simple average trajectory of two passes, and the solid red line marked with a hollow circle represents the combined reconstructed trajectory of two passes along the same track according to the present invention. Each trajectory starts from the same borehole position and displays the spatial deviation from the simulated real trajectory at multiple representative depth positions. Figure 5 It is evident that the single-pass reconstructed trajectory at the borehole gradually deviates from the true trajectory as the borehole depth increases due to the cumulative effects of zero bias and drift from the unidirectional sensor. The simple average trajectory of the two-pass test can partially offset the two-pass measurement error, but due to the lack of synchronous optimization of depth mapping parameters and direction error parameters, there is still a certain residual deviation. The two-pass co-track joint reconstructed trajectory of this invention maintains a high degree of consistency with the simulated true trajectory throughout the entire borehole depth range.

[0064] The error statistics for the three methods are as follows: The average position error of the single-pass reconstructed trajectory is 0.039435m, the maximum position error is 0.079109m, and the bottom position error is 0.079109m; the average position error of the two-pass simple average trajectory is 0.034289m, the maximum position error is 0.069798m, and the bottom position error is 0.069798m; the average position error of the two-pass co-track joint reconstructed trajectory of this invention is 0.008692m, the maximum position error is 0.017317m, and the bottom position error is 0.017317m. Compared to the single-pass integration method, the method of this invention reduces the average position error by approximately 78.0% and the bottom position error by approximately 78.1%; compared to the two-pass simple averaging method, the method of this invention reduces the average position error by approximately 74.7% and the bottom position error by approximately 75.2%.

[0065] (v) Calculation of skew parameters; Following the method described in step 8, after obtaining the three-dimensional center trajectory, the spatial deviation vector between the reconstructed trajectory points and the design trajectory points at each common physical depth is used. This spatial deviation vector is then projected onto a plane perpendicular to the design borehole axis, with the projection length serving as the friction deviation. The local borehole axis direction variation rate is calculated based on the angle between the unit directions of the common borehole axis at adjacent common depth nodes and the node spacing. This embodiment ultimately outputs the common physical depth, three-dimensional center coordinates, common borehole axis direction components, design trajectory coordinates, friction deviation, borehole bottom deviation, and local borehole axis direction variation rate. These are used for evaluating the construction quality of upward blast holes in underground metal mines, verifying charge structure, and optimizing blasting parameters.

[0066] The above description is merely a preferred embodiment of the present invention. Equivalent substitutions or transformations made by those skilled in the art to the sensor combination, phase space reconstruction algorithm, integrated disturbance observation method, fuzzy membership function, or control execution method without departing from the concept of the present invention shall all fall within the scope defined by the corresponding claims.

Claims

1. A method for reconstructing and measuring the three-dimensional trajectory of an upward borehole based on dual-path co-trajectory constraints, characterized in that, Includes the following steps: Step 1: Acquisition of measurement parameters; acquire the three-dimensional coordinates of the borehole opening, the initial axial direction of the borehole opening, the designed borehole parameters, and the calibration parameters of the measurement system; Step 2: Two-way measurement data acquisition and preprocessing; Complete the acquisition of measurement data during the two strokes of the probe entering and retracting in the borehole, and perform time synchronization, error compensation, attitude calculation, direction unification and unitization on the raw measurement data, and finally obtain the borehole axis observation sequence and the borehole axis observation sequence arranged in the direction from the borehole opening to the bottom of the borehole. Step 3: Common event identification and monotonic depth mapping establishment; Identify common events corresponding to the same physical location of the borehole during the drilling and retraction processes, establish a monotonic depth mapping from the retraction mileage to the common physical depth using the common events, and determine the initial parameters of the mapping. Step 4: Setting up common hole axis status and establishing two-way observation model; Set up a set of common hole axis statuses in the common physical depth domain, which are jointly corresponding to the hole entry hole axis observation sequence and the hole withdrawal hole axis observation sequence, and establish the hole entry observation model between the common hole axis status and the hole entry hole axis observation sequence, as well as the hole withdrawal observation model between the common hole axis status and the hole withdrawal hole axis observation sequence. Step 5: Generation of two-way borehole shaft prediction sequence and calculation of observation residual; using the borehole entry observation model and withdrawal observation model established in Step 4, generate the borehole entry prediction sequence and withdrawal prediction sequence according to the current common borehole shaft status and measurement error parameters, and calculate the observation residual between the prediction sequence and the corresponding borehole shaft observation sequence; Step 6: Establishment and iterative solution of joint objective function; establish a joint objective function based on observation residuals, common hole axis state constraints, common event depth correspondence constraints, and measurement error parameter constraints, and iteratively solve the joint objective function; In each iteration, the common physical depth corresponding to the retraction hole axis observation is re-determined based on the current monotonic depth mapping parameters, and the common hole axis status, hole entry measurement error parameters, retraction measurement error parameters and monotonic depth mapping parameters are updated synchronously until the preset convergence condition is met; Step 7: Reconstruction of the three-dimensional center trajectory; Based on the common hole axis state, common physical depth and three-dimensional coordinates of the hole opening obtained by joint solution, calculate the coordinates of the center point of the blast hole segment by segment along the direction of increasing hole depth to obtain the three-dimensional center trajectory of the upward blast hole to be measured in the mine coordinate system. Step 8: Calculate the deviation parameters; compare the reconstructed three-dimensional center trajectory with the designed borehole trajectory, calculate the deviation along the borehole, the bottom deviation, and the local borehole axis direction change rate of the borehole to be measured, and quantitatively evaluate the degree of borehole deviation.

2. The method for reconstructing and measuring the three-dimensional trajectory of an upward borehole based on dual-path co-track constraint according to claim 1, characterized in that, In step 1, the three-dimensional coordinates of the borehole are used to determine the starting point for reconstructing the three-dimensional center trajectory; the initial axial direction of the borehole serves as a reference for borehole direction constraints, used to constrain the value of the common borehole axis state at the borehole; the designed borehole parameters include the designed borehole depth, designed azimuth angle, and inclination angle, used to establish the designed borehole trajectory and perform skew comparison; the calibration parameters of the measurement system include the zero bias, scale factor, temperature coefficient, and installation error parameters of the inertial sensor, used for compensation of subsequent measurement data.

3. The method for reconstructing and measuring the three-dimensional trajectory of an upward borehole based on dual-path co-track constraint according to claim 1, characterized in that, In step 2, the process of acquiring and preprocessing two-way measurement data is as follows: S21: Static initial calibration; Before the probe enters the upward borehole to be measured, keep the probe stationary, collect the first static measurement data, and determine the inertial sensor zero bias and the probe initial attitude at the start of the measurement based on the first static measurement data; S22: Data acquisition during borehole entry; The probe is advanced from the borehole opening to the bottom of the borehole, and angular velocity data, acceleration data, mileage data, temperature data and measurement status data are acquired simultaneously during the borehole entry process; S23: Retraction Data Acquisition; After the probe reaches the bottom of the hole, the probe is retracted from the bottom of the hole to the opening, and angular velocity data, acceleration data, mileage data, temperature data and measurement status data are collected simultaneously during the retraction process; S24: Time synchronization and error compensation; synchronize the in-hole measurement data and the retraction measurement data according to a unified timestamp, and perform zero offset compensation, scale factor compensation, temperature compensation and installation error compensation according to the measurement system calibration parameters; S25: Attitude calculation and direction conversion; calculate the probe attitude based on the compensated inertial data of the borehole and the inertial data of the retraction, and convert the probe attitude into a three-dimensional orientation observation of the probe mechanical axis in the mine coordinate system; S26: Retraction data rearrangement and orientation unification; rearrange the retraction measurement data according to the direction from the borehole opening to the bottom of the borehole, and unify the orientation of the three-dimensional orientation observation during the retraction process according to the relationship between the direction of the probe's mechanical axis and the direction of the probe's movement. S27: Normalization; The three-dimensional directional observations during the drilling process and the retraction process are normalized to obtain the drilling axis observation sequence and the retraction axis observation sequence, respectively.

4. The method for reconstructing and measuring the three-dimensional trajectory of an upward borehole based on dual-path co-trajectory constraints according to claim 1, characterized in that, In step 3, the process of common event identification and monotonic depth mapping establishment is as follows: S31: Determination of orifice and bottom events; determine orifice and bottom events from the entry measurement data and the withdrawal measurement data respectively; S32: Extraction of candidate common events in the borehole; Based on the changes in acceleration, angular velocity, and odometer, extract candidate common events in the borehole during the entry and exit processes, respectively. S33: Initial correspondence establishment; Based on the similarity of characteristics of candidate common events, the order of arrangement along the hole, and the mileage interval between adjacent candidate common events, establish the initial correspondence between candidate common events for entering the hole and candidate common events for withdrawing; S34: Sequence consistency check; Perform a sequence consistency check on the initial correspondence, delete candidate common events whose corresponding order overlaps, feature similarity is lower than a preset threshold, and the difference between adjacent events exceeds a preset range, and obtain a set of common event corresponding points; S35: Establishment of monotonic depth mapping; Based on the set of points corresponding to the common events, a monotonic depth mapping from the pullback mileage to the common physical depth is established; using multiple pairs of mileage-depth correspondences in the set of points corresponding to the common events, the proportional parameters and zero-point offset parameters are initially determined by least squares fitting. S36: Verification and determination of initial mapping parameters; based on the depth correspondence error after the monotonic depth mapping of the orifice event, the proportional parameters and zero-point offset parameters obtained by the preliminary fitting in S35 are verified and corrected to determine the initial parameters of the monotonic depth mapping.

5. The method for reconstructing and measuring the three-dimensional trajectory of an upward borehole based on dual-path co-trajectory constraints according to claim 1, characterized in that, In step 4, the process of setting the common hole axis state and establishing the two-way observation model is as follows: S41: Common Depth Node Setting; Set up multiple common depth nodes in a common physical depth domain according to preset depth intervals; S42: Common hole axis state setting; A common hole axis state is set at each common depth node. The common hole axis state is used to represent the three-dimensional hole axis direction of the upward borehole to be measured at the corresponding common physical depth. S43: Determine the common hole axis direction at the hole entry observation position; based on the physical depth corresponding to the hole entry axis observation, interpolate the common hole axis state at adjacent common depth nodes to obtain the common hole axis direction corresponding to the hole entry axis observation position; S44: Determine the common aperture axis direction of the retreat observation position; Based on the current monotonic depth mapping, convert the retreat mileage corresponding to the retreat aperture axis observation into a common physical depth, and interpolate the common aperture axis state at adjacent common depth nodes to obtain the common aperture axis direction corresponding to the retreat aperture axis observation position. S45: Establishment of the inlet observation model; Establish the inlet observation model based on the common hole axis direction, the inlet direction constant deviation parameter, and the inlet direction drift parameter corresponding to the inlet hole axis observation position; S46: Establishment of the retreat observation model; establish the retreat observation model based on the common hole axis direction corresponding to the retreat hole axis observation position, the normal deviation parameter of the retreat direction, and the drift parameter of the retreat direction.

6. The method for reconstructing and measuring the three-dimensional trajectory of an upward borehole based on dual-path co-track constraint according to claim 1, characterized in that, In step 5, the process of generating the two-way aperture shaft prediction sequence and calculating the observation residuals is as follows: S51: Prediction calculation of the borehole axis; using the borehole observation model, calculate the predicted value of the borehole axis corresponding to each borehole axis observation position based on the common borehole axis status, borehole direction constant deviation parameter and borehole direction drift parameter; S52: Retreat mileage-depth conversion; using the current monotonic depth mapping, convert the retreat mileage corresponding to each retreat hole axis observation into a common physical depth; S53: Calculation of retraction hole axis prediction; Using the retraction observation model, calculate the predicted value of the retraction hole axis corresponding to each retraction hole axis observation position based on the mapped common physical depth, common hole axis state, retraction direction constant deviation parameter and retraction direction drift parameter. S54: Calculation of the residual of the hole entry observation; Calculate the residual of the hole entry observation between the predicted value of the hole shaft and the observed value of the hole shaft. S55: Calculation of the residual of the pullback observation; Calculate the residual of the pullback observation between the predicted value of the pullback hole axis and the observed value of the pullback hole axis.

7. The method for reconstructing and measuring the three-dimensional trajectory of an upward borehole based on dual-path co-track constraint according to claim 1, characterized in that, In step 6, the process of establishing and iteratively solving the joint objective function is as follows: S61: Establishment of two-way observation residuals; establish two-way observation residuals based on the borehole entry observation residuals and the withdrawal observation residuals; S62: Establishment of common hole-axis continuity constraint terms; establish common hole-axis continuity constraint terms based on the directional changes between the common hole-axis states at adjacent common depth nodes; S63: Establish unit length and orifice direction constraint terms; establish unit length constraint terms based on the modulus of the common hole shaft state, and establish orifice direction constraint terms based on the initial axial direction of the orifice; S64: Establishment of common event depth correspondence constraint terms; Establish common event depth correspondence constraint terms based on the depth difference after mapping common events through the current monotonic depth; S65: Establish parameter constraint items; establish parameter constraint items based on the allowable variation range of the hole inlet direction constant deviation parameter, hole inlet direction drift parameter, retraction direction constant deviation parameter, retraction direction drift parameter, proportional parameter and zero point offset parameter; S66: Combination of joint objective functions; combining the two-way observation residual terms, common borehole axis continuity constraints, unit length constraints, borehole direction constraints, common event depth corresponding constraints, and parameter constraints into a joint objective function; S67: Joint variable setting; setting joint variables including common hole axis status, hole inlet direction constant deviation parameter, hole inlet direction drift parameter, retraction direction constant deviation parameter, retraction direction drift parameter, proportional parameter and zero point offset parameter; S68: Nonlinear least squares iterative solution; The joint objective function is solved iteratively using a nonlinear least squares method; In each iteration, the common physical depth corresponding to all backtracking observations is first redefined based on the current scale parameter and zero offset parameter. Then, the common hole axis state at adjacent common depth nodes is interpolated based on the redefined common physical depth, and the predicted backtracking hole axis values ​​corresponding to each backtracking hole axis observation position are recalculated. Subsequently, the backtracking observation residuals are recalculated. At the same time, the predicted entry hole axis values ​​and entry hole observation residuals corresponding to each entry hole axis observation position are recalculated based on the updated common hole axis state. Finally, the updated joint objective function is minimized using a nonlinear least squares method, and all joint variables to be determined are corrected simultaneously. S69: Convergence Judgment and Result Output; Determine whether the change in the joint objective function and the change in the joint variable to be solved between two adjacent iterations meet the preset convergence conditions; If not, continue iterating; If so, output the common hole axis state, hole inlet direction constant deviation parameter, hole inlet direction drift parameter, pullback direction constant deviation parameter, pullback direction drift parameter, scale parameter, and zero point offset parameter.

8. The method for reconstructing and measuring the three-dimensional trajectory of an upward borehole based on dual-path co-track constraint according to claim 1, characterized in that, In step 7, the process of reconstructing the three-dimensional center trajectory is as follows: S71: Obtain the common hole-axis state; Obtain the common hole-axis state after joint solution at each common depth node; S72: Determination of average hole axis direction; Determine the average hole axis direction of the corresponding depth interval based on the common hole axis status at two adjacent common depth nodes; S73: Node coordinates are calculated segment by segment; taking the three-dimensional coordinates of the borehole as the starting point of the three-dimensional center trajectory, the three-dimensional coordinates of the next common depth node are calculated segment by segment according to the length of each depth interval and the average borehole axis direction. S74: Trajectory traversal output; sequentially traverse all common depth nodes to obtain the three-dimensional center trajectory of the upward borehole to be tested in the mine coordinate system.

9. The method for reconstructing and measuring the three-dimensional trajectory of an upward borehole based on dual-path co-track constraint according to claim 1, characterized in that, In step 8, the skew parameter calculation process is as follows: S81: Design trajectory establishment; Establish the design borehole trajectory based on the design borehole parameters, and determine the design trajectory point at the same common physical depth as the reconstructed 3D center trajectory; S82: Spatial Deviation Vector Calculation; Calculates the spatial deviation vector between the reconstructed trajectory point and the design trajectory point at each common physical depth; S83: Calculation of friction deviation; Project the spatial deviation vector onto a plane perpendicular to the direction of the corresponding design hole axis to obtain the lateral deviation at each common physical depth, and use the lateral deviation as the friction deviation; S84: Hole bottom deviation calculation; The total hole bottom deviation is determined based on the spatial distance between the reconstructed trajectory point and the design trajectory point at the hole bottom, and the lateral deviation of the hole bottom is determined based on the projection of the hole bottom spatial deviation vector onto the plane perpendicular to the design hole axis. S85: Calculation of local hole axis direction change rate; Calculate the included angle between the common hole axis directions at adjacent common depth nodes, and determine the local hole axis direction change rate based on the ratio of the included angle to the hole distance between adjacent common depth nodes.