A standard field-based measurement field dynamic monitoring and correction method

Abstract

The application discloses a kind of based on standard field measurement field dynamic monitoring and correction method, belong to large size space measurement and precision metrology technical field.The method includes: in measurement field, standard field with known geometric relation and multiple laser transmitters are arranged, and initial measurement field model is established using nonlinear least square algorithm;Periodically scan standard field during system operation, calculate monitoring index and carry out hierarchical determination to transmitter state;For slightly distorted transmitter, construct target function containing orthogonal constraint and regular term, update attitude parameter on line;For severely failed transmitter, after replacing new transmitter, realize self-organizing network orientation using mark target mutual sighting relationship, and quickly complete attitude calibration;Combined with omnidirectional photoelectric receiver and adaptive monitoring strategy, long-term stable maintenance to measurement field is realized.The application significantly reduces the frequency of re-calibration, improves the measurement accuracy and engineering application efficiency under complex environment.

Description

Technical Field

[0001] This invention belongs to the field of large-size spatial measurement and precision metrology technology, specifically relating to a method for dynamic monitoring and online correction of the measurement field suitable for ALPS distributed laser scanning measurement systems. It can be used for high-precision coordinate measurement and error control in large equipment manufacturing, tooling assembly and long-term online measurement scenarios. Background Technology

[0002] Currently, laser trackers, optical photogrammetry, and indoor GPS (such as wMPS, iGPS, ALPS, etc.) have become important means of large-scale measurement. Among them, the ALPS system, based on multiple transmitters and multiple photoelectric receivers, can achieve parallel positioning and attitude calculation of multiple targets through fan-shaped laser scanning and intersection measurement. It has advantages such as high precision, large range, and scalability, and is an important technical direction for the assembly and online measurement of large equipment.

[0003] However, in complex industrial environments, factors such as mechanical vibration, ambient temperature changes, and foundation structural deformation can continuously affect transmitters, causing slow or abrupt drifts in the relative pose between transmitters. Existing methods typically involve: 1) Accuracy can be restored by single or phased recalibration, but the recalibration process is time-consuming and labor-intensive, requires manual setting of control points, and is difficult to perform frequently; 2) The transmitter attitude is monitored using additional hardware such as tilt sensors, but the sensors themselves have errors such as zero drift, non-orthogonality, and temperature drift, and cannot directly reflect the overall accuracy changes in the measurement field.

[0004] On the other hand, existing research on abnormal transmitter identification and online correction mainly focuses on data post-processing, such as using kurtosis, 3σ criteria or principal component analysis to remove gross errors or common-mode noise, but lacks a real-time monitoring and correction mechanism based on spatial geometric constraints. When the transmitter is severely distorted, the system accuracy can often only be restored by shutting down the machine, manually checking and rearranging the field, which seriously affects the production cycle.

[0005] Therefore, there is an urgent need for a method that can operate for extended periods in complex environments, sense changes in the measurement field in real time, correct minor distortions online, and enable rapid replacement and self-organizing network orientation of severely faulty transmitters, thereby maintaining the overall measurement accuracy and reliability of the ALPS system without interrupting the measurement task. Summary of the Invention

[0006] The purpose of this invention is to address the problems of existing ALPS large-space measurement systems, such as measurement field degradation due to vibration and other factors during long-term operation, high recalibration costs, and lack of online correction capabilities. This invention proposes a method for dynamic monitoring and online correction of the measurement field based on a standard field, achieving: real-time / periodic monitoring and hierarchical judgment of transmitter status; online parameter correction for transmitters with slight distortion; rapid replacement and self-organizing network orientation for severely failed transmitters; and maintaining the continuity of measurement tasks while stably maintaining the overall accuracy of the system.

[0007] To achieve the above objectives, this invention provides a method for dynamic monitoring and online correction of a measurement field based on a standard field, the method comprising the following steps: (1) Set up a standard field containing several calibration points in the ALPS large space measurement field, place multiple laser transmitters on the outer edge or above the measurement area, fix the target point on each transmitter, and arrange an omnidirectional photoelectric receiver in the area to be measured. (2) Under low interference conditions, the measurement coordinates of each calibration point in the standard field in each transmitter coordinate system are obtained by using the ALPS intersection measurement principle. Combined with the real geometric relationship of the standard field, the nonlinear least squares algorithm is used to jointly solve the attitude parameters of each transmitter and the coordinates of the standard field to establish the initial measurement field model. (3) During system operation, standard field scanning is periodically or as needed. The coordinates of the calibration points measured in real time are compared with the theoretical coordinates calculated by the initial measurement field model to obtain monitoring indicators such as coordinate residuals, direction deviations and root mean square of residuals. The status of each transmitter is graded based on preset thresholds. (4) When the transmitter is in a slightly distorted state, under the premise of maintaining the standard field geometric constraints, the objective function is constructed with the transmitter attitude parameters and offset as the variables to be estimated, and the rotation matrix and translation vector of the corresponding transmitter are updated online by using an iterative optimization algorithm to realize the online correction of the transmitter coordinate system; (5) When the transmitter is in a state of severe distortion or failure, shut down the transmitter and replace it with a new transmitter. Using the correspondence between the target markers of each transmitter and the standard field calibration points, the self-organizing network coordinate orientation method is used to quickly solve the transformation relationship between the new transmitter and the reference coordinate system and restore the measurement field. (6) By analyzing the remeasurement error between the standard field and the measured point, the current measurement field performance is evaluated, and the monitoring cycle, threshold and standard field layout are adaptively adjusted to achieve continuous optimization of the measurement field.

[0008] Preferably, in step (4), to achieve online correction of the transmitter attitude parameters, the following constrained nonlinear least squares objective function is constructed: ; in, It is a 3×3 identity matrix, ||·|| 2 Denotes the Frobenius norm; Let be the real-time measured coordinates of the i-th standard field calibration point, where i = 1, 2, ..., N, and N is the total number of calibration points participating in online correction; These are the theoretical coordinates of the calibration point in the initial measurement field model; Here are the current transmitter attitude parameters, where Let be a rotation matrix. These are translation vectors, both derived from the initial measurement field model or the estimation results of the previous monitoring cycle; These are the increments of the attitude rotation parameters and the translation corrections, respectively. The corresponding rotation correction matrix is ​​denoted as ; The second term is the rotation matrix orthogonality constraint penalty term, used to constrain the approximation of the identity matrix. ; The third term is the attitude parameter offset regularization term, which is used to suppress overfitting and limit the magnitude of a single online correction. Let be the weighting coefficient, satisfying Preferred , ; When the transmitter is determined to be in a slightly distorted state, the system only performs local online correction on the corresponding attitude parameters of the transmitter. The specific process includes: 41) Keep the coordinates of the fixed standard field calibration point unchanged, and only increment the transmitter attitude. Translation correction As the variable to be estimated, among which Used to update the rotation correction matrix ; 42) Using the attitude parameters of the previous cycle as initial values, construct an objective function with orthogonal constraints and regularization terms; 43) The Levenberg-Marquardt algorithm is used for a small number of iterations to solve the problem. The iterations are stopped when the objective function converges or the preset number of iterations is reached. 44) The updated attitude parameters are immediately used for subsequent coordinate calculation of the measured points to achieve real-time recovery of the measurement field accuracy; 45) The correction process does not interrupt the measurement task and does not affect the calculation of other normal transmitters; In step (5), the coordinates of two or more target markers are pre-calibrated in the new transmitter's own coordinate system: ; The coordinates of the corresponding target in the reference coordinate system are obtained by scanning with the reference transmitter: ; Then the two coordinate systems satisfy the rigid body transformation relationship: ; The initial rotation matrix can be obtained by solving the above equations using least squares. With translation vector To improve accuracy, this solution can be used as an initial value, and standard field constraints can be introduced. Further fine-tuning can be achieved through nonlinear optimization, thereby enabling rapid orientation and access of the new transmitter. In step (6), the standard field retest error is obtained by repeatedly scanning the standard field calibration point, and is defined as: ; The repeatability error of the measured point is obtained by locating the measured target multiple times under the same working conditions and calculating its coordinate variance or RMS value. When the above error exceeds the preset threshold, the system will first trigger the online correction of the transmitter with slight distortion. If the error still cannot be reduced to the allowable range after correction, it is determined that there is a transmitter with severe distortion and the replacement process is initiated. Monitoring cycle The state determination threshold is based on the measured environmental vibration intensity. Task accuracy requirements Adaptive adjustment can be performed, and the relationship can be expressed as: ; in The basic monitoring period is measured in seconds (s); V is the normalized vibration intensity index, dimensionless, which can be obtained from the root mean square value of vibration velocity within the monitoring window. Compared with reference vibration value The ratio is obtained, i.e., V=v A represents the normalized accuracy requirement, which is dimensionless and can be determined from the reference allowable error. Coordinate error relative to the current task The ratio is obtained, that is ; These are empirical weighting coefficients, all greater than 0, with a preferred value of 0.2 to 2.0; When V or A increases, the system automatically shortens the monitoring cycle and tightens the judgment threshold; conversely, it reduces the computational load.

[0009] Preferably, in step (1), the standard field includes at least six calibration points, of which at least four calibration points are in the same plane and at least two calibration points are not coplanar, in order to improve the stability and disease resistance of coordinate solution.

[0010] Preferably, in step (2), the nonlinear least squares algorithm adopts the Levenberg-Marquardt algorithm with a penalty factor, and a penalty term constraining the orthogonality of each transmitter rotation matrix is ​​added to the objective function to improve the stability and physical rationality of attitude parameter calculation.

[0011] Preferably, in step (3), the dynamic monitoring indicators include at least: (1) Three-dimensional coordinate residuals of each calibration point in the standard field; (2) The angle error between the ray direction vector measured by each transmitter and the theoretical direction vector in the initial model; (3) The root mean square of the coordinate residuals and the maximum residuals for each transmitter; Based on the above indicators, a comprehensive health evaluation function is constructed to classify the transmitter status as normal, slightly distorted, or severely distorted; the comprehensive health evaluation function for the transmitter is defined. : ; in: RMS is the root mean square of the standard field coordinate residuals; E_max is the maximum coordinate residual; Average deviation angle of ray direction; RMS_ref, E_max_ref, and α_ref are reference thresholds; w_1, w_2, and w_3 are weight coefficients that satisfy w_1>0, w_2>0, w_3>0, and w_1+w_2+w_3=1; The transmitter status determination rules are as follows: when : Determined to be in a normal state; when The distortion was determined to be mild. when The condition is determined to be severely distorted or in a state of failure. Under healthy, low-interference operating conditions, a series of health status samples are obtained by continuously collecting standard field scans for a period of time (e.g., 30-60 minutes). (k=1,2,...,K), where K is the number of samples. This represents the overall health value corresponding to the k-th standard field scan. ; set up ; ; in, This represents the sample mean of overall health under healthy working conditions. The corresponding standard deviation; This corresponds to a very low false alarm rate; This can be considered an abnormal / mutation level; threshold Used to distinguish between normal system fluctuations and mild distortions, it can be adaptively determined based on the statistical distribution of health status under healthy operating conditions, such as taking... The threshold can also be set by reverse calculation based on the upper limit of the allowable error of the task, so as to realize the adaptive adjustment of the monitoring threshold according to the environment and task. threshold Used to determine severe distortion or failure status, can be taken .

[0012] Preferably, in step (4), the objective function for online correction consists of the following three parts: (1) The sum of squares of the residuals between the measured coordinates and the theoretical coordinates of the standard field calibration points; (2) Constraints on the deviation between the transmitter rotation matrix and the orthogonal matrix; (3) Regularization term for the offset between the transmitter's current attitude parameters and its initial attitude parameters; By adjusting the weights of the above three parts, a trade-off between measurement accuracy and parameter stability can be achieved.

[0013] Preferably, in step (5), the self-organizing network coordinate orientation method includes the following steps: (1) Pre-calibrate the coordinates of at least two self-marked targets in the coordinate system of each transmitter; (2) When a new transmitter is connected, at least two normal transmitters scan the new transmitter's target to obtain the target's coordinates in each transmitter's coordinate system; (3) Based on the correspondence of the target in different coordinate systems, establish constraint equations and solve the rotation matrix and translation vector of the new transmitter relative to the reference coordinate system to achieve rapid orientation without redeploying the field control points.

[0014] Preferably, in step (1), the omnidirectional photodetector uses a polyhedral structure to arrange multiple high-speed photodetectors, and each photodetector is connected in sequence to a transimpedance amplifier circuit, an amplifier circuit and a buffer circuit to convert laser signals incident from any direction into high signal-to-noise ratio voltage signals.

[0015] Preferably, in step (6), the monitoring cycle and the state judgment threshold can be adaptively adjusted according to the vibration intensity at the measurement site, the measurement task duration and the accuracy requirements. When the vibration intensity is greater or the accuracy requirements are higher, the monitoring cycle is shorter and the threshold setting is more stringent.

[0016] The present invention also discloses a measurement field dynamic monitoring and correction system for implementing the above method, comprising: The standard field module is used to provide a spatial arrangement of multiple calibration points with known geometric relationships; Multiple transmitter modules are used to emit fan-shaped lasers and form a scanning measurement network; An omnidirectional optoelectronic receiver module is used to receive reference pulses and sector laser signals from different transmitters; The data acquisition and processing module is used to acquire time signals between each transmitter and calibration point and the measured point, and to perform coordinate calculations. The dynamic monitoring and online correction module is used to perform standard field residual calculation, transmitter status assessment, online optimization of attitude parameters, replacement of failed transmitters, and orientation of self-organizing networks. The dynamic monitoring and online correction module is configured to dynamically maintain the measurement field according to the above method.

[0017] Advantages of this invention: Compared with the prior art, the present invention has the following beneficial effects: 1) Based on the unified geometric constraints of the standard field, the monitoring results directly correspond to the measurement accuracy. Unlike relying solely on tilt or vibration sensors to monitor transmitter attitude, this invention uses standard field calibration points as a unified geometric benchmark. It directly reflects the true accuracy changes of the measurement results through spatial coordinate residuals, ensuring that the monitoring indicators are highly consistent with the measurement tasks and thus have higher engineering relevance.

[0018] 2) Mild distortion is corrected online to avoid frequent recalibration. By constructing an objective function that includes orthogonal constraints and regularization terms, and using algorithms such as LM for small-scale parameter updates, this invention can perform online correction of the attitude of a transmitter with slight distortion without interrupting the measurement task, significantly reducing the number of times and the cost of full-field recalibration.

[0019] 3) Severely failed transmitters can be quickly replaced, and self-organizing networks improve operational efficiency. By utilizing the marker target and standard field constraints built into the transmitter, this invention can achieve self-organizing network orientation through multi-transmitter mutual aiming when introducing a new transmitter, without the need to rearrange all control points, achieving "plug and play" rapid access and significantly shortening downtime.

[0020] 4) Omnidirectional optoelectronic receivers improve signal availability and robustness. The omnidirectional photoelectric receiver with a polyhedral arrangement and its matching analog front-end circuit can increase the number of visible transmitters and signal quality in complex obstructed environments, making standard field monitoring and measurement tasks more robust.

[0021] 5) Adaptive maintenance of the measurement field to meet the needs of long-term operation in complex environments. By linking vibration intensity, task accuracy requirements, and monitoring strategies, this invention enables adaptive adjustment of monitoring cycles and thresholds, allowing the system to automatically increase monitoring frequency in high-vibration, high-measurement scenarios and reduce computational load in relatively stable scenarios, making it more suitable for long-term continuous operation.

[0022] In summary, this invention provides a measurement field maintenance method for distributed large-space measurement systems such as ALPS that balances accuracy, efficiency, and engineering feasibility through an integrated design of "standard field + dynamic monitoring + online correction + rapid station addition". Attached Figure Description

[0023] Figure 1 This is a flowchart of the method for dynamic monitoring and online correction of a measurement field based on a standard field according to the present invention; Figure 2 This is a schematic diagram of the mutual aiming points between the calibration plate and the transmitter in the rotating laser positioning system of the present invention. Detailed Implementation

[0024] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, but the present invention is not limited to the following embodiments.

[0025] Example 1

[0026] Standard field construction and initial measurement field establishment.

[0027] like Figure 1 As shown, the standard field-based measurement field dynamic monitoring and online correction method of the present invention mainly includes standard field construction, initial model establishment, dynamic monitoring, hierarchical judgment, online correction, and rapid replacement. Specifically, in this embodiment, in the ALPS measurement field of a large assembly workshop, four laser transmitters are arranged using a combination of ceiling-mounted and wall-fixed installation to reduce the direct impact of on-site obstruction and ground vibration.

[0028] To meet the requirements for stability and disease resistance of the standard field arrangement, a standard field consisting of 6 high-precision reflective targets is arranged in the area to be measured. The specific layout is as follows: 4 calibration points are located on the same horizontal reference plane, and the other 2 calibration points are arranged at different heights to form a spatially non-coplanar set of points. This geometric relationship is pre-calibrated by a high-precision standard ruler.

[0029] Under low-interference operating conditions with no significant external vibration, each transmitter is started sequentially. Utilizing the ALPS scanning intersection principle, the measurement coordinates of each calibration point in the standard field are acquired in the coordinate system of each transmitter. To establish a high-precision initial measurement field model, the system employs the Levenberg-Marquardt (LM) algorithm with a penalty factor to construct an initial optimization objective function containing an orthogonality penalty term. : ; The first term is the sum of squared residuals between the measured coordinates and the theoretical coordinates, and the second term is the constrained transmitter rotation matrix. Orthogonality penalty term, The penalty factor (preferably set to [value] in this embodiment) ), The identity matrix is ​​used. The LM algorithm is employed to select appropriate initial values ​​and iteratively solve the problem. When the objective function converges, the initial high-precision attitude parameters and true coordinates of the standard field for each transmitter are obtained, thus establishing the initial measurement field model.

[0030] Example 2

[0031] Dynamic monitoring and online correction based on standard field residuals.

[0032] During the long-term assembly process, the system needs to periodically evaluate the measurement field status. In this embodiment, the data acquisition module obtains the measurement coordinates of each transmitter to the standard field target point, compares them with the theoretical values ​​in the initial measurement field model, and calculates the single-point three-dimensional coordinate residual, the ray direction deviation angle of each transmitter, the root mean square of the coordinate residual, and the maximum residual.

[0033] To quantitatively assess the transmitter's operational status, the system introduces a comprehensive health evaluation function. In the specific configuration of this embodiment, the overall health level is considered. The calculation formula is as follows: ; The specific definitions of each parameter and the preferred settings in this embodiment are as follows: RMS, , These are the root mean square of the standard field coordinate residuals, the maximum coordinate residuals, and the average deviation angle of the ray direction, measured in real time, respectively. , , This is a preset reference threshold. Based on the accuracy requirements of this assembly workshop, it is set... , , ; Let be the weighting coefficient, satisfying Since the root mean square residual best reflects the overall deviation, this embodiment preferably sets it to... .

[0034] The system calculates the real-time health status. Then, based on the set threshold (e.g., set to 1.2) and (If set to 2.5) Perform state determination: when When this occurs, it is considered a normal state; when If the distortion is determined to be mild, the system will automatically trigger an online correction process in the background. when If the condition is determined to be severely distorted or in a failure state, the rapid replacement process in Example 3 is triggered.

[0035] When a transmitter is determined to have slight distortion, in order to restore accuracy without interrupting the measurement task, the system constructs a nonlinear least-squares objective function with orthogonal constraints and regularization terms. Online correction is performed. In this embodiment, the objective function is specifically expressed as: ; The execution logic and parameter definitions of the above objective function in the embodiment are as follows: The first item represents the measurement coordinates of the standard field calibration point. With theoretical coordinates The sum of squared residuals is used to constrain the accuracy after correction; The second term is the orthogonality constraint penalty term for the rotation matrix, ensuring that the updated rotation matrix... Satisfies physical plausibility (approximates the identity matrix) ); The third term is a regularization term for attitude parameter offset, used to suppress system oscillations caused by excessively large single corrections. and These are the rotation increment and translation increment to be optimized, respectively. As the corresponding penalty weight, in this embodiment, to balance accuracy and stability, a certain value is set. .

[0036] The system uses the LM algorithm to solve the objective function in a small number of iterations (e.g., 5-10 times). The updated attitude parameters are immediately used for the coordinate calculation of the subsequent measured points, thus achieving "soft maintenance".

[0037] In addition, to balance the monitoring needs of computational load and extreme operating conditions, this system also introduces an adaptive adjustment mechanism for the monitoring cycle. The calculation rules are as follows: ; The specific parameter is defined as follows: Set the basic monitoring cycle (e.g., 600 seconds). The normalized vibration intensity index (calculated from data from on-site vibration sensors); This is a normalized accuracy requirement indicator; These are empirical weighting coefficients (e.g., all are taken as 1.0).

[0038] When the vibration at the scene intensifies ( (Increase) or the current process has extremely high precision requirements ( When the period is increased, the calculated period is... It will automatically shorten the cycle to achieve high-frequency monitoring; conversely, it will extend the cycle to reduce the system's computational load.

[0039] Example 3

[0040] Rapid replacement of failed transmitters and orientation of self-organizing networks.

[0041] like Figure 2 As shown, each laser transmitter has a fixed target point on its casing for mutual aiming and coordinate system transfer. When a transmitter is determined to be severely distorted or malfunctioning according to the monitoring mechanism in Example 2, the system sends an alarm to the operator and automatically shuts down the transmitter. At this point, the self-organizing network orientation and rapid replacement process begins.

[0042] After replacing the transmitter, at least two self-marked targets are pre-calibrated and fixed on its outer casing (three marker targets are used in this embodiment to improve the redundancy of the solution). At this time, at least two normally functioning reference transmitters in the measurement field scan the marker targets on the new transmitter to obtain the measurement coordinates $P_{ref\_j}$ ($j=1,2,3$) of these three marker targets in the reference coordinate system.

[0043] Meanwhile, the local coordinates $P_{new\_j}$ of these three marker targets in the new transmitter's own coordinate system are known quantities. The following rigid body transformation equations are satisfied between the two coordinate systems: ; in, Let be the rotation matrix of the new transmitter relative to the reference coordinate system. Let be the translation vector to be determined. The system collects the corresponding coordinate data of the three marked targets, constructs an overdetermined system of equations, and firstly obtains the initial rotation matrix quickly using the linear least squares method. With translation vector .

[0044] To further improve orientation accuracy, the system uses this solution as an initial value, introduces known geometric constraints of the standard field, and again uses the nonlinear optimization algorithm from Example 1 for local fine-tuning. Through the above steps, the new transmitter achieves "plug-and-play" rapid access without the need to redeploy global control points, ensuring the continuity of production cycles.

[0045] Example 4

[0046] Design and application of omnidirectional optoelectronic receivers.

[0047] To accommodate the need for standard field calibration points to be arranged at different heights and orientations, and to ensure the continuity of dynamic monitoring signals, this system design adopts an omnidirectional photoelectric receiver structure.

[0048] The receiver housing employs a regular polyhedral (e.g., dodecahedral) structure, with high-speed photodetectors (PIN photodiodes) distributed across multiple faces. To convert weak laser signals incident from any direction into high signal-to-noise ratio signals, signal processing circuitry is sequentially connected to the back end of each photodetector. Transimpedance amplifier (TIA): This circuit converts the weak photocurrent generated by the photodiode (such as...) The level) is linearly converted into a voltage signal. In this embodiment, the transimpedance gain is preferably set to [value missing]. ; Amplifier circuit: performs secondary amplification on the primary voltage signal and filters out high-frequency environmental noise; Buffer circuit: performs impedance matching and outputs a stable high signal-to-noise ratio voltage signal for subsequent high-precision time measurement (TDC) and coordinate calculation.

[0049] Experiments show that, compared with traditional single-sided photoelectric receivers, this multi-faceted omnidirectional structure and its matching amplification circuit significantly increase the number of visible transmitters in complex obstructed environments. Even when the transmitter attitude is distorted to a certain extent, the standard field can still be scanned stably, providing a reliable hardware data source for dynamic monitoring and online correction in Example 2.

Claims

1. A method for dynamic monitoring and correction of a measurement field based on a standard field, characterized in that, Includes the following steps (1) Set up a standard field containing several calibration points in the ALPS large space measurement field, place multiple laser transmitters on the outer edge or above the measurement area, fix the target point on each transmitter, and arrange an omnidirectional photoelectric receiver in the area to be measured. (2) Under low interference conditions, the measurement coordinates of each calibration point in the standard field in each transmitter coordinate system are obtained by using the ALPS intersection measurement principle. Combined with the real geometric relationship of the standard field, the nonlinear least squares algorithm is used to jointly solve the attitude parameters of each transmitter and the coordinates of the standard field to establish the initial measurement field model. (3) During system operation, standard field scanning is periodically or as needed. The coordinates of the calibration points measured in real time are compared with the theoretical coordinates calculated by the initial measurement field model to obtain monitoring indicators such as coordinate residuals, direction deviations and root mean square of residuals. The status of each transmitter is graded based on preset thresholds. (4) When the transmitter is in a slightly distorted state, under the premise of maintaining the standard field geometric constraints, the objective function is constructed with the transmitter attitude parameters and offset as the variables to be estimated, and the rotation matrix and translation vector of the corresponding transmitter are updated online by using an iterative optimization algorithm to realize the online correction of the transmitter coordinate system; (5) When the transmitter is in a state of severe distortion or failure, shut down the transmitter and replace it with a new transmitter. Using the correspondence between the target markers of each transmitter and the standard field calibration points, the self-organizing network coordinate orientation method is used to quickly solve the transformation relationship between the new transmitter and the reference coordinate system and restore the measurement field. (6) By analyzing the remeasurement error between the standard field and the measured point, the current measurement field performance is evaluated, and the monitoring cycle, threshold and standard field layout are adaptively adjusted to achieve continuous optimization of the measurement field.

2. The method according to claim 1, characterized in that, In step (4), to achieve online correction of the transmitter attitude parameters, the following constrained nonlinear least squares objective function is constructed: ； in, It is a 3×3 identity matrix, ||·|| 2 Denotes the Frobenius norm; Let be the real-time measured coordinates of the i-th standard field calibration point, where i = 1, 2, ..., N, and N is the total number of calibration points participating in online correction; These are the theoretical coordinates of the calibration point in the initial measurement field model; Here are the current transmitter attitude parameters, where Let be a rotation matrix. These are translation vectors, both derived from the initial measurement field model or the estimation results of the previous monitoring cycle; These are the increments of the attitude rotation parameters and the translation corrections, respectively. The corresponding rotation correction matrix is ​​denoted as ; The second term is the rotation matrix orthogonality constraint penalty term, used to constrain the approximation of the identity matrix. ; The third term is the attitude parameter offset regularization term, which is used to suppress overfitting and limit the magnitude of a single online correction. Let be the weighting coefficient, satisfying Preferred , ; When the transmitter is determined to be in a slightly distorted state, the system only performs local online correction on the corresponding attitude parameters of the transmitter. The specific process includes: 41) Keep the coordinates of the fixed standard field calibration point unchanged, and only increment the transmitter attitude. Translation correction As the variable to be estimated, among which Used to update the rotation correction matrix ; 42) Using the attitude parameters of the previous cycle as initial values, construct an objective function with orthogonal constraints and regularization terms; 43) The Levenberg-Marquardt algorithm is used for a small number of iterations to solve the problem. The iterations are stopped when the objective function converges or the preset number of iterations is reached. 44) The updated attitude parameters are immediately used for subsequent coordinate calculation of the measured points to achieve real-time recovery of the measurement field accuracy; 45) The correction process does not interrupt the measurement task and does not affect the calculation of other normal transmitters; In step (5), the coordinates of two or more target markers are pre-calibrated in the new transmitter's own coordinate system: ； The coordinates of the corresponding target in the reference coordinate system are obtained by scanning with the reference transmitter: ; Then the two coordinate systems satisfy the rigid body transformation relationship: ； The initial rotation matrix can be obtained by solving the above equations using least squares. With translation vector To improve accuracy, this solution can be used as an initial value, and standard field constraints can be introduced. Further fine-tuning can be achieved through nonlinear optimization, thereby enabling rapid orientation and access of the new transmitter. In step (6), the standard field retest error is obtained by repeatedly scanning the standard field calibration point, and is defined as: ； The repeatability error of the measured point is obtained by locating the measured target multiple times under the same working conditions and calculating its coordinate variance or RMS value. When the above error exceeds the preset threshold, the system will first trigger the online correction of the transmitter with slight distortion. If the error still cannot be reduced to the allowable range after correction, it is determined that there is a transmitter with severe distortion and the replacement process is initiated. Monitoring cycle The state determination threshold is based on the measured environmental vibration intensity. Task accuracy requirements Adaptive adjustment can be performed, and the relationship can be expressed as: ； in The basic monitoring period is measured in seconds (s); V is the normalized vibration intensity index, dimensionless, which can be obtained from the root mean square value of vibration velocity within the monitoring window. Compared with reference vibration value The ratio is obtained, i.e., V=v A represents the normalized accuracy requirement, which is dimensionless and can be determined from the reference allowable error. Coordinate error relative to the current task The ratio is obtained, that is ; These are empirical weighting coefficients, all greater than 0, with a preferred value of 0.2 to 2.0; When V or A increases, the system automatically shortens the monitoring cycle and tightens the judgment threshold; conversely, it reduces the computational load.

3. The method according to claim 1, characterized in that, In step (1), the standard field includes at least six calibration points, of which at least four calibration points are in the same plane and at least two calibration points are not coplanar, in order to improve the stability and disease resistance of coordinate solution.

4. The method according to claim 1, characterized in that, In step (2), the nonlinear least squares algorithm adopts the Levenberg-Marquardt algorithm with a penalty factor, and adds a penalty term to the objective function to constrain the orthogonality of each transmitter rotation matrix, so as to improve the stability and physical rationality of attitude parameter solution.

5. The method according to claim 1, characterized in that, In step (3), the dynamic monitoring indicators shall include at least: (1) Three-dimensional coordinate residuals of each calibration point in the standard field; (2) The angle error between the ray direction vector measured by each transmitter and the theoretical direction vector in the initial model; (3) The root mean square of the coordinate residuals and the maximum residuals for each transmitter; Based on the above indicators, a comprehensive health evaluation function is constructed to classify the transmitter status as normal, slightly distorted, or severely distorted; the comprehensive health evaluation function for the transmitter is defined. : ； in: RMS is the root mean square of the standard field coordinate residuals; E_max is the maximum coordinate residual; Average deviation angle of ray direction; RMS_ref, E_max_ref, and α_ref are reference thresholds; w_1, w_2, and w_3 are weight coefficients that satisfy w_1>0, w_2>0, w_3>0, and w_1+w_2+w_3=1; The transmitter status determination rules are as follows: when : Determined to be in a normal state; when The distortion was determined to be mild. when The condition is determined to be severely distorted or in a state of failure. Under healthy, low-interference operating conditions, a series of health status samples are obtained by continuously collecting standard field scans for a period of time (e.g., 30-60 minutes). (k=1,2,...,K), where K is the number of samples. This represents the overall health value corresponding to the k-th standard field scan. ； set up ； ； in, This represents the sample mean of overall health under healthy working conditions. The corresponding standard deviation; This corresponds to a very low false alarm rate; This can be considered an abnormal / mutation level; threshold Used to distinguish between normal system fluctuations and mild distortions, it can be adaptively determined based on the statistical distribution of health status under healthy operating conditions, such as taking... The threshold can also be set by reverse calculation based on the upper limit of the allowable error of the task, so as to realize the adaptive adjustment of the monitoring threshold according to the environment and task. threshold Used to determine severe distortion or failure status, can be taken .

6. The method according to claim 1, characterized in that, In step (4), the objective function for online correction consists of the following three parts: (1) The sum of squares of the residuals between the measured coordinates and the theoretical coordinates of the standard field calibration points; (2) Constraints on the deviation between the transmitter rotation matrix and the orthogonal matrix; (3) Regularization term for the offset between the transmitter's current attitude parameters and its initial attitude parameters; By adjusting the weights of the above three parts, a trade-off between measurement accuracy and parameter stability can be achieved.

7. The method according to claim 1, characterized in that, In step (5), the self-organizing network coordinate orientation method includes the following steps: (1) Pre-calibrate the coordinates of at least two self-marked targets in the coordinate system of each transmitter; (2) When a new transmitter is connected, at least two normal transmitters scan the new transmitter's target to obtain the target's coordinates in each transmitter's coordinate system; (3) Based on the correspondence of the target in different coordinate systems, establish constraint equations and solve the rotation matrix and translation vector of the new transmitter relative to the reference coordinate system to achieve rapid orientation without redeploying the field control points.

8. The method according to claim 1, characterized in that, In step (1), the omnidirectional photodetector uses a polyhedral structure to arrange multiple high-speed photodetectors. Each photodetector is connected in sequence to a transimpedance amplifier circuit, an amplifier circuit, and a buffer circuit to convert laser signals incident from any direction into high signal-to-noise ratio voltage signals.

9. The method according to claim 1, characterized in that, In step (6), the monitoring cycle and state judgment threshold can be adaptively adjusted according to the vibration intensity at the measurement site, the duration of the measurement task and the accuracy requirements. When the vibration intensity is greater or the accuracy requirements are higher, the monitoring cycle is shorter and the threshold setting is more stringent.

10. A measurement field dynamic monitoring and correction system for implementing the method according to any one of claims 1 to 9, characterized in that, include: The standard field module is used to provide a spatial arrangement of multiple calibration points with known geometric relationships; Multiple transmitter modules are used to emit fan-shaped lasers and form a scanning measurement network; An omnidirectional optoelectronic receiver module is used to receive reference pulses and sector laser signals from different transmitters; The data acquisition and processing module is used to acquire time signals between each transmitter and calibration point and the measured point, and to perform coordinate calculations. The dynamic monitoring and online correction module is used to perform standard field residual calculation, transmitter status assessment, online optimization of attitude parameters, replacement of failed transmitters, and orientation of self-organizing networks. The dynamic monitoring and online correction module is configured to dynamically maintain the measurement field according to the method described in any one of claims 1 to 9.

Images