Tunneling equipment position calibration method and system adopting cursor frame

Abstract

The invention discloses a tunneling equipment position calibration method and system adopting a cursor frame, and relates to the technical field of coal mine tunnel construction tagging. Geologic feature data in a coal mine tunnel are collected in real time through a dynamic environment data set, an environment adaptability index Szp is calculated, and the influence of the environment on equipment calibration is evaluated; a calibration error evaluation coefficient Gz and an error growth trend value Wczs are combined, a calibration strategy is dynamically optimized, adaptive filtering and a multi-modal fusion algorithm are adopted to improve positioning precision, and stable work in an environment with large geological change is ensured; meanwhile, calibration parameters are adjusted in real time by using a deep learning model and a self-adaptive path calibration algorithm, and the adaptability to complex geological conditions is enhanced; the interference compensation module processes electromagnetic interference through Kalman filtering, particle filtering and other algorithms, optimizes the error coefficient Ins of the inertial navigation system, corrects the deviation of the sensor, reduces the influence of external interference on the calibration precision, and guarantees the high precision and stability of the system.

Description

Technical Field

[0001] The invention relates to the technical field of coal mine tunnel construction marking, and in particular to a method and system for calibrating the position of tunneling equipment using a cursor frame. Background Art

[0002] The excavation equipment position calibration system originated from the excavation operations of underground projects such as mines and coal mine tunnels, and is used to accurately determine the position of excavation equipment in complex underground environments. This type of system initially relied on simple ground measurement equipment for positioning, but with technological advances, it has gradually developed into a high-precision automation system that combines sensors, lidar, inertial navigation and other technologies. With the increase in the scale of coal mine tunnel projects and the increase in automation needs, the accuracy and real-time performance of the excavation equipment position calibration system have become a research focus. Modern systems can not only improve the positioning accuracy of equipment, but also monitor the status of equipment in real time, support automatic adjustment and control, and significantly improve operating efficiency and safety.

[0003] The application fields of tunneling equipment are mainly concentrated in mining, coal mine tunnel construction, underground pipeline laying and other fields; however, when used in coal mine tunnel construction, there are often the following technical disadvantages: Insufficient calibration accuracy: Under complex geological conditions, such as soft soil or rock faults, the accuracy of the tunneling equipment position calibration system may not meet the requirements, resulting in large marking errors and affecting the accuracy of the tunneling path; Unable to cope with complex geological conditions: The geological conditions in coal mines are usually complex and changeable. Differences in stratum hardness and structure may make it difficult for the marking system of tunneling equipment to accurately cope with them. The marking system of the equipment may fail or become inaccurate when encountering abnormal geological conditions; Sensor interference: There are a large number of electromagnetic interference sources in coal mine tunnels, such as construction equipment and power facilities, which may affect the normal operation of sensors and thus affect the accuracy of position calibration. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention provides a method and system for calibrating the position of tunneling equipment using a cursor frame, which solves the technical shortcomings of insufficient calibration accuracy, signal loss and sensor interference mentioned in the background technology.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: a tunneling equipment position calibration system using a cursor frame, including an environment acquisition module, a calibration optimization module, a cursor positioning module, an environment perception module, an interference compensation module and a feedback optimization module; The environment collection module is used to collect coal mine tunnel environmental data and build a dynamic environment data set; The calibration optimization module calculates the calibration error evaluation coefficient Gz based on the dynamic environment data set, uses an adaptive path calibration algorithm to predict the path deviation and calculate the error trend value Wczs, and optimizes the error compensation parameters by combining multi-objective optimization and machine learning models; The cursor positioning module is used to calibrate the position of the tunneling equipment using a multimodal fusion algorithm and adjust the cursor frame weight; The environmental perception module uses multi-sensor data fusion technology to perceive the environment in the coal mine tunnel and calculate and evaluate the environmental adaptability index Gjs; optimize the fusion weight of the inertial navigation and lidar systems; The interference compensation module calculates the error coefficient Ins of the inertial navigation system, uses a deep learning model to analyze the electromagnetic interference characteristics, combines time series prediction to optimize the compensation strategy, and adjusts the weights of inertial navigation and lidar data through adaptive filtering; The feedback optimization module uses time series anomaly detection to analyze and calibrate historical data, combines it with reinforcement learning to optimize calibration parameters, and proposes correction suggestions.

[0006] Preferably, the environment acquisition module collects the internal illumination data of the coal mine tunnel in real time by setting high-precision photosensitive sensors around the tunneling equipment, and performs data sampling according to the set time interval, and uses the time series filtering algorithm to eliminate abnormal illumination data to form illumination intensity time series data; secondly, by using a broadband electromagnetic field sensor, the electromagnetic field intensity change in the coal mine tunnel environment is detected, and the construction equipment interference signal and external electromagnetic interference are distinguished based on the multi-band signal analysis algorithm, and the electromagnetic field related data is output in real time; then a high-precision temperature and humidity sensor is used to collect the temperature and humidity in the coal mine tunnel in real time and perform multi-layer data fusion to obtain temperature and humidity related data; finally, based on the ultrasonic geological radar and the multi-point sensor array, the layered scanning technology is used to obtain the rock structure information of the coal mine tunnel, wherein the ultrasonic signal is emitted through the pulse excitation mode, and the rock density, water content and fault zone position are calculated using the echo signal, thereby generating the coal mine tunnel geological characteristics related data; After the coal mine tunnel environmental data is collected, the adaptive filtering method is used to denoise the light intensity time series data, electromagnetic field related data, temperature and humidity related data, and coal mine tunnel geological characteristics related data, and a dynamic environmental data set is constructed through hierarchical modeling technology.

[0007] Preferably, the calibration optimization module includes an error evaluation unit, a path prediction and compensation unit, and a machine learning optimization unit; The error evaluation unit is used to calculate the calibration error evaluation coefficient Gz based on the dynamic environment data set; the calibration error evaluation coefficient Gz is obtained by extracting the ambient light intensity change Ggl, electromagnetic field interference intensity Gdq, temperature change Gwb and humidity change Gsb in the dynamic environment data set and performing dimensionless processing, and then combining the following formula to calculate: ; Then, the calibration error is evaluated; by comparing and evaluating the preset calibration error evaluation threshold Q with the calibration error evaluation coefficient Gz, it is determined whether the calibration error is qualified. The specific evaluation contents are as follows: If the calibration error evaluation coefficient Gz ≤ the calibration error evaluation threshold Q, it means that the calibration error is qualified, indicating that the current calibration accuracy meets the requirements, and the equipment continues to work normally at this time; If the calibration error evaluation coefficient Gz> calibration error evaluation threshold Q, it means that the calibration error is unqualified, indicating that the position calibration error of the equipment exceeds the preset range, and adjustment or correction is performed at this time; When the evaluation results show that the calibration error is unqualified, the future error growth trend value Wczs of the tunneling equipment is calculated. First, according to the motion trajectory of the tunneling equipment and the dynamic environment data set, the path error model is constructed to obtain the equipment and path position error value Wwz, tunneling equipment position deviation value Wpc, equipment sensor error Wsc and coal mine tunnel soil hardness index Wtc. Then, after the amount-free processing, the error growth trend value Wczs is calculated by the following formula: ; In the formula, y o They represent the equipment and path position error value Wwz, the tunneling equipment position deviation value Wpc, the equipment sensor error Wsc and the coal mine tunnel soil layer hardness index Wtc respectively, and m represents the number of variables involved in the calculation; By presetting the error growth trend threshold W, the error growth trend value Wczs of the tunneling equipment is compared and evaluated, and the calibration strategy and parameter settings of the tunneling equipment are adjusted in real time. The specific evaluation contents are as follows: If the error growth trend value Wczs ≤ the error growth trend threshold W, it means that the error growth trend of the tunneling equipment does not have the risk of divergence, the equipment operates normally, and the calibration strategy remains unchanged; If the error growth trend value Wczs>the error growth trend threshold W, it means that the error growth trend of the tunneling equipment has a risk of divergence. At this time, the calibration strategy and parameter settings are adjusted in real time, including sending a compensation activation instruction to the path prediction and compensation unit.

[0008] Preferably, the path prediction and compensation unit adopts an adaptive path calibration algorithm to predict the deviation of the excavation path of the next 5-10 meters based on the calibration error evaluation coefficient Gz and the error growth trend value Wczs; dynamically adjusts the calibration strategy through a multi-objective optimization algorithm to generate a compensation parameter set including propulsion speed, cutter head torque and hydraulic cylinder pressure.

[0009] Preferably, the machine learning optimization unit is used to build and deploy a random forest and neural network fusion model, taking the error growth trend value Wczs, the dynamic environment data set and the compensation parameter set as input; outputs the optimal compensation parameter weight coefficient, and updates the compensation parameter mapping relationship in real time through an online learning mechanism.

[0010] Preferably, the environment perception module includes a geological condition quantification unit, an environment analysis unit and a deep learning adjustment unit; The geological condition quantification unit is used to collect geological characteristic data including stratum hardness, rock type, groundwater level, and structural differences in coal mine tunnels in real time through multi-sensor data fusion technology to construct a geological data set; extract the geological data set and perform dimensionless processing, including rock hardness value Geo, rock type number Pha, groundwater level depth Mul, and fault displacement value Ion, and calculate the environmental adaptability index Szp through the following formula: ; In the formula, They represent the rock hardness value Geo, the number of rock types Pha, the groundwater level depth Mul and the fault displacement value Ion respectively; n represents the number of variables involved in the calculation.

[0011] Preferably, the environment analysis unit compares and evaluates the preset environment adaptability threshold E with the environment adaptability index Gjs to determine the environment adaptability. The specific evaluation content is as follows: If the environmental adaptability index Gjs ≥ environmental adaptability threshold E, the environmental adaptability is judged to be "qualified", indicating that the geological conditions in the coal mine tunnel have little impact on the equipment calibration system, and the equipment continues to use the current calibration parameters for positioning; If the environmental adaptability index Gjs is less than the environmental adaptability threshold E, the environmental adaptability is judged as "unqualified", indicating that the geological conditions in the coal mine tunnel may cause the calibration system to be inaccurate, and the deep learning model needs to be started for dynamic adjustment.

[0012] Preferably, the deep learning adjustment unit analyzes the impact of complex geological conditions in the coal mine tunnel on the calibration system and dynamically adjusts the calibration parameters based on the real-time collected environmental data; the specific process is as follows: When the environmental adaptability assessment is "qualified", maintain the existing calibration parameters; When the environmental adaptability assessment is "unqualified", the deep learning model is started to learn the relationship between the geological characteristics in the coal mine tunnel and the calibration error, adjust the calibration parameters in real time, and optimize the fusion weight of the inertial navigation and lidar system.

[0013] Preferably, the interference compensation module is used to receive the output data of the inertial navigation system, including acceleration-related data, angular velocity-related data and attitude information-related data, and then extract the acceleration Ijs, angular velocity Isd, device attitude angle Izt, magnetic field intensity Icc and working environment temperature Iwd and perform dimensionless processing to calculate the inertial navigation system error coefficient Ins. The specific calculation formula is as follows: .

[0014] Preferably, the interference compensation module presets an inertial navigation system error threshold R and an inertial navigation system error coefficient Ins for comparative evaluation, and the specific contents are as follows: When the inertial navigation system error coefficient Ins ≤ the inertial navigation system error threshold R, it means that the system error is within an acceptable range and the inertial navigation system is in normal working condition; When the inertial navigation system error coefficient Ins> the inertial navigation system error threshold R, it means that the system error exceeds the acceptable range, the system error is abnormal, and the navigation accuracy is inaccurate. At this time, compensation processing is performed. The specific compensation processing is as follows: When the error coefficient Ins of the inertial navigation system exceeds the error threshold R of the inertial navigation system, the electromagnetic interference signal is first processed through filtering algorithms including Kalman filtering and particle filtering, especially the electromagnetic wave interference that affects the performance of the inertial navigation system; during the filtering process, the electromagnetic interference signal is identified and filtered out in the real-time collected data, and dynamic correction is performed according to the relevant data of the inertial sensor and the external environment information; secondly, the system parameters are adjusted through the compensation algorithm according to the error of the inertial navigation system; then the error of the inertial navigation system is corrected by adjusting the calibration parameters, angle deviation and position error of the sensor; finally, the electromagnetic interference is compensated.

[0015] Preferably, the feedback optimization module includes a data preprocessing subunit and an intelligent optimization subunit; The data preprocessing subunit is used to receive the time series data stream of the calibration error evaluation index Gz, the signal quality optimization index Szp and the inertial navigation system error coefficient Ins, and extract the latest 20 sets of historical data through the sliding window mechanism; then build and use the Prophet time series model to detect abnormal fluctuation points, and mark them as key abnormal events when the following conditions are met: The calibration error evaluation index Gz exceeds the threshold Gz_threshold for three consecutive sampling values ​​and the gradient change rate is greater than 5% / s; The absolute value of the correlation coefficient between the signal quality optimization index Szp and the inertial navigation system error coefficient Ins is less than 0.3 and lasts for more than 10 seconds; Finally, a weighted error trend analysis report is generated, in which the weight coefficients of the calibration error evaluation index Gz, the signal quality optimization index Szp and the inertial navigation system error coefficient Ins are dynamically allocated according to the current geological type.

[0016] Preferably, the intelligent optimization subunit is used to construct a parameter optimization model based on a multi-agent reinforcement learning framework, and use the parameter optimization model to generate a set of correction suggestions through offline training and online fine-tuning, including: When the calibration error evaluation index Gz is detected as the dominant anomaly, the calibration strategy is adjusted first: shorten the laser target sampling interval to 200ms, and increase the gyroscope data weight by 40%; when the signal quality optimization index Szp abnormality persists, switch to the multi-band signal fusion mode, and inject spread spectrum coding check bits into the UWB channel; trigger composite compensation for the inertial navigation system error coefficient Ins abnormality: enable CEEMDAN filtering and inertial navigation zero bias online calibration at the same time.

[0017] A method for calibrating the position of a tunneling device using a cursor frame comprises the following steps: Step 1: Collect coal mine tunnel environmental data based on multi-sensor fusion technology, including illumination, electromagnetic field, temperature and humidity, and geological structure characteristics, and use adaptive filtering and hierarchical modeling to construct a dynamic environmental data set; Step 2: Calculate the calibration error evaluation coefficient Gz based on the dynamic environment data set, use the adaptive path calibration algorithm to predict the path deviation and calculate the error trend value Wczs, adjust the calibration strategy based on multi-objective optimization, and optimize the error compensation parameters in combination with the machine learning model; Step 3: Combining inertial navigation, lidar SLAM and visual recognition technology, a multimodal fusion algorithm is used to achieve position calibration, and the cursor frame weight is adjusted based on the calibration error evaluation index Gz; Step 4: Use multi-sensor data fusion technology to perceive the environment in the coal mine tunnel, further perceive the geological characteristics including stratum hardness, rock type, groundwater level, and structural differences, and calculate and evaluate the environmental adaptability index Szp based on real-time geological data; use a deep learning model to analyze the impact of complex geological conditions in coal mine tunnels on the calibration system, dynamically adjust calibration parameters, and optimize the fusion weight of inertial navigation and lidar systems; Step 5: Calculate the error coefficient Ins of the inertial navigation system, use a deep learning model to analyze the electromagnetic interference characteristics, combine time series prediction to optimize the compensation strategy, and adjust the weights of inertial navigation and lidar data through adaptive filtering; Step 6. Based on the evaluation contents of the calibration error evaluation index Gz, the signal quality optimization index Szp and the inertial navigation system error coefficient Ins, the time series anomaly detection is used to analyze the calibration historical data, automatically generate error trend analysis, and combine reinforcement learning to optimize the calibration parameters, and put forward correction suggestions, including adjusting the calibration strategy, optimizing the signal enhancement scheme and the interference compensation strategy.

[0018] The present invention provides a method and system for calibrating the position of tunneling equipment using a cursor frame. The method has the following beneficial effects: (1) This method and system for calibrating the position of tunneling equipment using a cursor frame solves the problem of insufficient calibration accuracy of tunneling equipment under complex geological conditions, especially in soft soil layers or rock faults. The system adopts a dynamic environmental data set construction technology, which can obtain geological characteristic data in coal mine tunnels in real time, such as rock hardness value Geo, rock type number Pha, groundwater level depth Mul and fault displacement value Ion, and calculates the environmental adaptability index Szp after dimensionless processing, so as to evaluate the impact of the environment on the equipment calibration system. Combined with the calibration error evaluation coefficient Gz and the error growth trend value Wczs, the calibration strategy is dynamically optimized to predict and compensate for the deviation of the equipment position, thereby improving the calibration accuracy in complex geological environments. In addition, the positioning accuracy of the system is optimized by using adaptive filtering and multimodal fusion algorithms, so that it can work continuously and stably in coal mine tunnels with large geological changes.

[0019] (2) This method and system for calibrating the position of tunneling equipment using a cursor frame effectively solves the problem of the complexity of geological conditions in coal mines, especially in areas where the hardness of the strata varies greatly and the structural differences are more significant. Traditional calibration systems are prone to inaccuracy or failure due to environmental factors. The system collects environmental data in coal mine tunnels in real time, such as light intensity, electromagnetic field intensity, temperature and humidity data, and geological structure data, and uses deep learning models and adaptive path calibration algorithms for dynamic adjustment, thereby enhancing the system's adaptability to complex geological conditions and being able to adjust calibration parameters in real time, so that the calibration system can still maintain high-precision positioning in uncertain or changing environments. At the same time, combined with the multi-sensor data fusion technology of the inertial navigation system and the lidar system, the system can obtain more accurate tunneling equipment position data in complex coal mine tunnel environments.

[0020] (3) This method and system for calibrating the position of tunneling equipment using a cursor box solves the problem of sensor interference. There are a large number of electromagnetic interference sources in coal mine tunnels, such as construction equipment and power facilities, which can easily affect the normal operation of sensors, thereby reducing the accuracy of position calibration. The system adopts an interference compensation module, combined with electromagnetic interference feature analysis and deep learning models, and processes interference signals in real time through algorithms such as Kalman filtering and particle filtering. It also compensates for the errors of the inertial navigation system, optimizes the error coefficient Ins of the inertial navigation system, and corrects the deviation of the sensor in real time, ensuring the stable operation of the system and reducing the negative impact of external interference on the calibration accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 It is a schematic diagram of the framework structure of a tunneling equipment position calibration system using a cursor frame according to the present invention; Figure 2 The present invention is a schematic flow chart of the steps of a method for calibrating the position of a tunneling device using a cursor frame. DETAILED DESCRIPTION

[0022] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0023] Example 1 See also Figure 1 , the present invention provides a tunneling equipment position calibration system using a cursor frame, including an environment acquisition module, a calibration optimization module, a cursor positioning module, an environment perception module, an interference compensation module and a feedback optimization module; The environment collection module is used to collect coal mine tunnel environmental data and build a dynamic environment data set; The calibration optimization module calculates the calibration error evaluation coefficient Gz based on the dynamic environment data set, uses an adaptive path calibration algorithm to predict the path deviation and calculate the error trend value Wczs, and optimizes the error compensation parameters by combining multi-objective optimization and machine learning models; The cursor positioning module is used to calibrate the position of the tunneling equipment using a multimodal fusion algorithm and adjust the cursor frame weight; The environmental perception module uses multi-sensor data fusion technology to perceive the environment in the coal mine tunnel and calculate and evaluate the environmental adaptability index Szp; optimize the fusion weight of the inertial navigation and lidar systems; The interference compensation module calculates the error coefficient Ins of the inertial navigation system, uses a deep learning model to analyze the electromagnetic interference characteristics, combines time series prediction to optimize the compensation strategy, and adjusts the weights of inertial navigation and lidar data through adaptive filtering; The feedback optimization module uses time series anomaly detection to analyze and calibrate historical data, combines it with reinforcement learning to optimize calibration parameters, and proposes correction suggestions.

[0024] In this embodiment, the environment acquisition module can collect coal mine tunnel environmental data in real time, generate a dynamic environment data set, and improve the timeliness and accuracy of the environmental data; the calibration optimization module calculates the calibration error evaluation coefficient Gz based on the dynamic environment data set, and uses an adaptive path calibration algorithm to predict the path deviation and calculate the error trend value Wczs, combining multi-objective optimization with machine learning models to optimize error compensation parameters, significantly improving the accuracy and stability of position calibration; the cursor positioning module calibrates the position of the tunneling equipment through a multimodal fusion algorithm, and dynamically adjusts the cursor frame weight according to environmental conditions to improve the positioning accuracy and adaptability of the equipment; the environmental perception module uses multi-sensor data to Fusion technology can sense the environment in the coal mine tunnel and calculate and evaluate the environmental adaptability index Szp, optimize the fusion weight of the inertial navigation and lidar systems, and enhance the system's adaptability to complex geological environments; the interference compensation module can calculate the error coefficient Ins of the inertial navigation system, and use the deep learning model to analyze the electromagnetic interference characteristics, combine the time series prediction to optimize the compensation strategy, and adjust the weights of inertial navigation and lidar data through adaptive filtering to reduce the impact of interference on calibration accuracy; the feedback optimization module analyzes the calibration historical data through time series anomaly detection, optimizes the calibration parameters through reinforcement learning, and puts forward correction suggestions, thereby continuously improving the calibration accuracy and automatically adjusting the calibration strategy.

[0025] Example 2 The environment acquisition module sets high-precision photosensitive sensors around the tunneling equipment to collect real-time data related to the illumination inside the coal mine tunnel, and samples data according to the set time interval, and uses the time series filtering algorithm to eliminate abnormal illumination related data to form illumination intensity time series data; secondly, by using a broadband electromagnetic field sensor, the electromagnetic field intensity changes in the coal mine tunnel environment are detected, and the construction equipment interference signal and external electromagnetic interference are distinguished based on the multi-band signal analysis algorithm, and the electromagnetic field related data are output in real time; then a high-precision temperature and humidity sensor is used to collect the temperature and humidity in the coal mine tunnel in real time and perform multi-layer data fusion to obtain temperature and humidity related data; finally, based on ultrasonic geological radar and multi-point sensor array, the layered scanning technology is used to obtain the rock structure information of the coal mine tunnel, in which the ultrasonic signal is emitted through the pulse excitation mode, and the echo signal is used to calculate the rock density, water content and fault zone position, thereby generating data related to the geological characteristics of the coal mine tunnel; After the coal mine tunnel environmental data is collected, the adaptive filtering method is used to denoise the light intensity time series data, electromagnetic field related data, temperature and humidity related data, and coal mine tunnel geological characteristics related data, and a dynamic environmental data set is constructed through hierarchical modeling technology.

[0026] In this embodiment, the environment acquisition module realizes the accurate acquisition of various environmental data in the coal mine tunnel by combining multiple high-precision sensors, including light intensity time series data, electromagnetic field related data, temperature and humidity related data and geological characteristics related data, providing comprehensive real-time environmental basic data for subsequent calibration and optimization; the light-related data in the coal mine tunnel is collected in real time by high-precision photosensitive sensors, and the abnormal light data is eliminated by the time series filtering algorithm to ensure data accuracy; the wide-band electromagnetic field sensor and multi-band signal analysis algorithm are used to effectively distinguish the interference signal of construction equipment from the external electromagnetic interference, thereby enhancing the system's resistance to interference signals. Resistance; through high-precision temperature and humidity sensors and multi-layer data fusion technology, real-time acquisition of temperature and humidity data in coal mine tunnels improves the accuracy of environmental adaptability assessment; the layered scanning technology of ultrasonic geological radar and multi-point sensor array is used to obtain the rock density, water content and fault zone location of coal mine tunnels, helping to build more accurate coal mine geological characteristic data; through adaptive filtering and layered modeling technology to denoise and build a dynamic environmental data set, it provides high-quality, real-time updated environmental data support for calibration optimization, ensuring that the system can maintain stable and accurate operation in complex environments, and improving the adaptability and intelligence level of the overall system.

[0027] Example 3 The calibration optimization module includes an error evaluation unit, a path prediction and compensation unit, and a machine learning optimization unit; The error evaluation unit is used to calculate the calibration error evaluation coefficient Gz based on the dynamic environment data set; the calibration error evaluation coefficient Gz is obtained by extracting the ambient light intensity change Ggl, electromagnetic field interference intensity Gdq, temperature change Gwb and humidity change Gsb in the dynamic environment data set and performing dimensionless processing, and then combining the following formula to calculate: ; Then, the calibration error is evaluated; by comparing and evaluating the preset calibration error evaluation threshold Q with the calibration error evaluation coefficient Gz, it is determined whether the calibration error is qualified. The specific evaluation contents are as follows: If the calibration error evaluation coefficient Gz ≤ the calibration error evaluation threshold Q, it means that the calibration error is qualified, indicating that the current calibration accuracy meets the requirements, and the equipment continues to work normally at this time; If the calibration error evaluation coefficient Gz>calibration error evaluation threshold Q, it means that the calibration error is unqualified, indicating that the error of the position calibration of the equipment exceeds the preset range, and adjustment or correction is performed at this time.

[0028] When the evaluation results show that the calibration error is unqualified, the future error growth trend value Wczs of the tunneling equipment is calculated. First, according to the motion trajectory of the tunneling equipment and the dynamic environment data set, the path error model is constructed to obtain the equipment and path position error value Wwz, tunneling equipment position deviation value Wpc, equipment sensor error Wsc and coal mine tunnel soil hardness index Wtc. Then, after the amount-free processing, the error growth trend value Wczs is calculated by the following formula: ; In the formula, y o They represent the equipment and path position error value Wwz, the tunneling equipment position deviation value Wpc, the equipment sensor error Wsc and the coal mine tunnel soil layer hardness index Wtc respectively, and m represents the number of variables involved in the calculation; By presetting the error growth trend threshold W, the error growth trend value Wczs of the tunneling equipment is compared and evaluated, and the calibration strategy and parameter settings of the tunneling equipment are adjusted in real time. The specific evaluation contents are as follows: If the error growth trend value Wczs ≤ the error growth trend threshold W, it means that the error growth trend of the tunneling equipment does not have the risk of divergence, the equipment operates normally, and the calibration strategy remains unchanged; If the error growth trend value Wczs>the error growth trend threshold W, it means that the error growth trend of the tunneling equipment has a risk of divergence. At this time, the calibration strategy and parameter settings are adjusted in real time, including sending a compensation activation instruction to the path prediction and compensation unit.

[0029] The path prediction and compensation unit adopts an adaptive path calibration algorithm to predict the deviation of the excavation path of the next 5-10 meters based on the calibration error evaluation coefficient Gz and the error growth trend value Wczs; the calibration strategy is dynamically adjusted through a multi-objective optimization algorithm to generate a set of compensation parameters including propulsion speed, cutter head torque and hydraulic cylinder pressure.

[0030] The machine learning optimization unit is used to build and deploy a random forest and neural network fusion model, taking the error growth trend value Wczs, the dynamic environment data set and the compensation parameter set as input; outputs the optimal compensation parameter weight coefficient, and updates the compensation parameter mapping relationship in real time through the online learning mechanism.

[0031] The environmental perception module includes a geological condition quantification unit, an environmental analysis unit, and a deep learning adjustment unit; The geological condition quantification unit is used to collect geological characteristic data including stratum hardness, rock type, groundwater level, and structural differences in coal mine tunnels in real time through multi-sensor data fusion technology to construct a geological data set; extract the geological data set and perform dimensionless processing, including rock hardness value Geo, rock type number Pha, groundwater level depth Mul, and fault displacement value Ion, and calculate the environmental adaptability index Szp through the following formula: ; In the formula, They represent the rock hardness value Geo, the number of rock types Pha, the groundwater level depth Mul and the fault displacement value Ion respectively; n represents the number of variables involved in the calculation.

[0032] The environmental analysis unit compares and evaluates the preset environmental adaptability threshold E with the environmental adaptability index Gjs to determine the environmental adaptability. The specific evaluation contents are as follows: If the environmental adaptability index Gjs ≥ environmental adaptability threshold E, the environmental adaptability is judged to be "qualified", indicating that the geological conditions in the coal mine tunnel have little impact on the equipment calibration system, and the equipment continues to use the current calibration parameters for positioning; If the environmental adaptability index Gjs is less than the environmental adaptability threshold E, the environmental adaptability is judged as "unqualified", indicating that the geological conditions in the coal mine tunnel may cause the calibration system to be inaccurate, and the deep learning model needs to be started for dynamic adjustment.

[0033] The deep learning adjustment unit analyzes the impact of complex geological conditions in coal mine tunnels on the calibration system and dynamically adjusts the calibration parameters based on real-time collected environmental data. The specific process is as follows: When the environmental adaptability assessment is "qualified", maintain the existing calibration parameters; When the environmental adaptability assessment is "unqualified", the deep learning model is started to learn the relationship between the geological characteristics in the coal mine tunnel and the calibration error, adjust the calibration parameters in real time, and optimize the fusion weight of the inertial navigation and lidar system.

[0034] The interference compensation module is used to receive the output data of the inertial navigation system, including acceleration-related data, angular velocity-related data and attitude information-related data, and then extract the acceleration Ijs, angular velocity Isd, device attitude angle Izt, magnetic field intensity Icc and working environment temperature Iwd and perform dimensionless processing to calculate the inertial navigation system error coefficient Ins. The specific calculation formula is as follows: .

[0035] The interference compensation module presets the inertial navigation system error threshold R and the inertial navigation system error coefficient Ins for comparative evaluation. The specific contents are as follows: When the inertial navigation system error coefficient Ins ≤ the inertial navigation system error threshold R, it means that the system error is within an acceptable range and the inertial navigation system is in normal working condition; When the inertial navigation system error coefficient Ins> the inertial navigation system error threshold R, it means that the system error exceeds the acceptable range, the system error is abnormal, and the navigation accuracy is inaccurate. At this time, compensation processing is performed. The specific compensation processing is as follows: When the error coefficient Ins of the inertial navigation system exceeds the error threshold R of the inertial navigation system, the electromagnetic interference signal is first processed through filtering algorithms including Kalman filtering and particle filtering, especially the electromagnetic wave interference that affects the performance of the inertial navigation system; during the filtering process, the electromagnetic interference signal is identified and filtered out in the real-time collected data, and dynamic correction is performed according to the relevant data of the inertial sensor and the external environment information; secondly, the system parameters are adjusted through the compensation algorithm according to the error of the inertial navigation system; then the error of the inertial navigation system is corrected by adjusting the calibration parameters, angle deviation and position error of the sensor; finally, the electromagnetic interference is compensated.

[0036] The feedback optimization module includes a data preprocessing subunit and an intelligent optimization subunit; The data preprocessing subunit is used to receive the time series data stream of the calibration error evaluation index Gz, the signal quality optimization index Szp and the inertial navigation system error coefficient Ins, and extract the latest 20 sets of historical data through the sliding window mechanism; then build and use the Prophet time series model to detect abnormal fluctuation points, and mark them as key abnormal events when the following conditions are met: The calibration error evaluation index Gz exceeds the threshold Gz_threshold for three consecutive sampling values ​​and the gradient change rate is greater than 5% / s; The absolute value of the correlation coefficient between the signal quality optimization index Szp and the inertial navigation system error coefficient Ins is less than 0.3 and lasts for more than 10 seconds; Finally, a weighted error trend analysis report is generated, in which the weight coefficients of the calibration error evaluation index Gz, the signal quality optimization index Szp and the inertial navigation system error coefficient Ins are dynamically allocated according to the current geological type.

[0037] The intelligent optimization subunit is used to build a parameter optimization model based on a multi-agent reinforcement learning framework. The parameter optimization model is used to generate a set of correction suggestions through offline training and online fine-tuning, including: When the calibration error evaluation index Gz is detected as the dominant anomaly, the calibration strategy is adjusted first: shorten the laser target sampling interval to 200ms, and increase the gyroscope data weight by 40%; when the signal quality optimization index Szp abnormality persists, switch to the multi-band signal fusion mode, and inject spread spectrum coding check bits into the UWB channel; trigger composite compensation for the inertial navigation system error coefficient Ins abnormality: enable CEEMDAN filtering and inertial navigation zero bias online calibration at the same time.

[0038] In this embodiment, the calibration optimization module effectively improves the calibration accuracy and stability of the system through the error evaluation unit, the path prediction and compensation unit and the machine learning optimization unit; the error evaluation unit calculates the calibration error evaluation coefficient Gz, based on the light intensity change Ggl, the electromagnetic field interference intensity Gdq, the temperature change Gwb and the humidity change Gsb in the dynamic environment data set, and evaluates the calibration error after dimensionless processing to ensure that the calibration error meets the preset threshold, and then decides whether to adjust the calibration accuracy of the equipment; the path prediction and compensation unit predicts the path deviation of the tunneling equipment through an adaptive path calibration algorithm, and adjusts the calibration strategy in combination with the error growth trend value Wczs, dynamically optimizes the compensation parameter set, and improves the adaptability and positioning accuracy of the equipment in complex environments; the machine learning optimization unit uses a random forest and neural network fusion model to dynamically learn the error growth trend value Wczs and the environmental data set, and updates the compensation parameter mapping relationship in real time, thereby further optimizing the performance of the equipment; The environmental perception module collects geological characteristic data in coal mine tunnels in real time based on multi-sensor data fusion technology through geological condition quantification unit, environmental analysis unit and deep learning adjustment unit, calculates the environmental adaptability index Szp, and evaluates environmental adaptability to ensure that the calibration system can dynamically adjust parameters according to different geological conditions and optimize the fusion weight of inertial navigation and lidar systems; the interference compensation module collects inertial navigation system output data, such as acceleration Ijs, angular velocity Isd, attitude angle Izt, etc., calculates the inertial navigation system error coefficient Ins, and combines Kalman filtering, particle filtering and other algorithms to compensate for electromagnetic interference and sensor errors in real time to ensure navigation accuracy; the feedback optimization module monitors calibration errors, signal quality and inertial navigation errors in real time through data preprocessing and intelligent optimization sub-units, detects key abnormal events through time series analysis, generates correction suggestion sets based on multi-agent reinforcement learning models, and provides targeted calibration strategy adjustments and compensation solutions to ensure the stability and accuracy of the system in long-term operation.

[0039] Example 4 See also Figure 2 , a method for calibrating the position of a tunneling device using a cursor frame, comprising the following steps: Step 1: Collect coal mine tunnel environmental data based on multi-sensor fusion technology, including illumination, electromagnetic field, temperature and humidity, and geological structure characteristics, and use adaptive filtering and hierarchical modeling to construct a dynamic environmental data set; Step 2: Calculate the calibration error evaluation coefficient Gz based on the dynamic environment data set, use the adaptive path calibration algorithm to predict the path deviation and calculate the error trend value Wczs, adjust the calibration strategy based on multi-objective optimization, and optimize the error compensation parameters in combination with the machine learning model; Step 3: Combining inertial navigation, lidar SLAM and visual recognition technology, a multimodal fusion algorithm is used to achieve position calibration, and the cursor frame weight is adjusted based on the calibration error evaluation index Gz; Step 4: Use multi-sensor data fusion technology to perceive the environment in the coal mine tunnel, further perceive the geological characteristics including stratum hardness, rock type, groundwater level, and structural differences, and calculate and evaluate the environmental adaptability index Szp based on real-time geological data; use a deep learning model to analyze the impact of complex geological conditions in coal mine tunnels on the calibration system, dynamically adjust calibration parameters, and optimize the fusion weight of inertial navigation and lidar systems; Step 5: Calculate the error coefficient Ins of the inertial navigation system, use a deep learning model to analyze the electromagnetic interference characteristics, combine time series prediction to optimize the compensation strategy, and adjust the weights of inertial navigation and lidar data through adaptive filtering; Step 6. Based on the evaluation contents of the calibration error evaluation index Gz, the signal quality optimization index Szp and the inertial navigation system error coefficient Ins, the time series anomaly detection is used to analyze the calibration historical data, automatically generate error trend analysis, and combine reinforcement learning to optimize the calibration parameters, and put forward correction suggestions, including adjusting the calibration strategy, optimizing the signal enhancement scheme and the interference compensation strategy.

[0040] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A tunneling equipment position calibration system using a cursor frame, characterized in that: It includes environment acquisition module, calibration optimization module, cursor positioning module, environment perception module, interference compensation module and feedback optimization module; The environment collection module is used to collect coal mine tunnel environmental data and build a dynamic environment data set; The calibration optimization module calculates the calibration error evaluation coefficient Gz based on the dynamic environment data set, uses an adaptive path calibration algorithm to predict the path deviation and calculate the error trend value Wczs, and optimizes the error compensation parameters by combining multi-objective optimization and machine learning models; The cursor positioning module is used to calibrate the position of the tunneling equipment using a multimodal fusion algorithm and adjust the cursor frame weight; The environmental perception module uses multi-sensor data fusion technology to perceive the environment in the coal mine tunnel and calculate and evaluate the environmental adaptability index Szp; optimize the fusion weight of the inertial navigation and lidar systems; The interference compensation module calculates the error coefficient Ins of the inertial navigation system, uses a deep learning model to analyze the electromagnetic interference characteristics, combines time series prediction to optimize the compensation strategy, and adjusts the weights of inertial navigation and lidar data through adaptive filtering; The feedback optimization module uses time series anomaly detection to analyze and calibrate historical data, combines it with reinforcement learning to optimize calibration parameters, and proposes correction suggestions.

2. A tunneling equipment position calibration system using a cursor frame according to claim 1, characterized in that: The environment acquisition module sets high-precision photosensitive sensors around the tunneling equipment to collect real-time data related to the illumination inside the coal mine tunnel, and samples data according to the set time interval, and uses the time series filtering algorithm to eliminate abnormal illumination related data to form illumination intensity time series data; secondly, by using a broadband electromagnetic field sensor, the electromagnetic field intensity changes in the coal mine tunnel environment are detected, and the construction equipment interference signal and external electromagnetic interference are distinguished based on the multi-band signal analysis algorithm, and the electromagnetic field related data are output in real time; then a high-precision temperature and humidity sensor is used to collect the temperature and humidity in the coal mine tunnel in real time and perform multi-layer data fusion to obtain temperature and humidity related data; finally, based on ultrasonic geological radar and multi-point sensor array, the layered scanning technology is used to obtain the rock structure information of the coal mine tunnel, in which the ultrasonic signal is emitted through the pulse excitation mode, and the echo signal is used to calculate the rock density, water content and fault zone position, thereby generating data related to the geological characteristics of the coal mine tunnel; After the coal mine tunnel environmental data is collected, the adaptive filtering method is used to denoise the light intensity time series data, electromagnetic field related data, temperature and humidity related data, and coal mine tunnel geological characteristics related data, and a dynamic environmental data set is constructed through hierarchical modeling technology.

3. A tunneling equipment position calibration system using a cursor frame according to claim 1, characterized in that: The calibration optimization module includes an error evaluation unit, a path prediction and compensation unit, and a machine learning optimization unit; The error evaluation unit is used to calculate the calibration error evaluation coefficient Gz based on the dynamic environment data set; the calibration error evaluation coefficient Gz is obtained by extracting the ambient light intensity change Ggl, electromagnetic field interference intensity Gdq, temperature change Gwb and humidity change Gsb in the dynamic environment data set and performing dimensionless processing, and then combining the following formula to calculate: ； Then, the calibration error is evaluated; by comparing and evaluating the preset calibration error evaluation threshold Q with the calibration error evaluation coefficient Gz, it is determined whether the calibration error is qualified. The specific evaluation contents are as follows: If the calibration error evaluation coefficient Gz ≤ the calibration error evaluation threshold Q, it means that the calibration error is qualified, indicating that the current calibration accuracy meets the requirements, and the equipment continues to work normally at this time; If the calibration error evaluation coefficient Gz> calibration error evaluation threshold Q, it means that the calibration error is unqualified, indicating that the position calibration error of the equipment exceeds the preset range, and adjustment or correction is performed at this time; When the evaluation results show that the calibration error is unqualified, the future error growth trend value Wczs of the tunneling equipment is calculated. First, according to the motion trajectory of the tunneling equipment and the dynamic environment data set, the path error model is constructed to obtain the equipment and path position error value Wwz, tunneling equipment position deviation value Wpc, equipment sensor error Wsc and coal mine tunnel soil hardness index Wtc. Then, after the amount-free processing, the error growth trend value Wczs is calculated by the following formula: ； In the formula, y o They represent the equipment and path position error value Wwz, the tunneling equipment position deviation value Wpc, the equipment sensor error Wsc and the coal mine tunnel soil layer hardness index Wtc respectively, and m represents the number of variables involved in the calculation; By presetting the error growth trend threshold W, the error growth trend value Wczs of the tunneling equipment is compared and evaluated, and the calibration strategy and parameter settings of the tunneling equipment are adjusted in real time. The specific evaluation contents are as follows: If the error growth trend value Wczs ≤ the error growth trend threshold W, it means that the error growth trend of the tunneling equipment does not have the risk of divergence, the equipment operates normally, and the calibration strategy remains unchanged; If the error growth trend value Wczs>the error growth trend threshold W, it means that the error growth trend of the tunneling equipment has a risk of divergence. At this time, the calibration strategy and parameter settings are adjusted in real time, including sending a compensation activation instruction to the path prediction and compensation unit.

4. A tunneling equipment position calibration system using a cursor frame according to claim 3, characterized in that: The path prediction and compensation unit adopts an adaptive path calibration algorithm to predict the deviation of the excavation path of the next 5-10 meters based on the calibration error evaluation coefficient Gz and the error growth trend value Wczs; the calibration strategy is dynamically adjusted through a multi-objective optimization algorithm to generate a set of compensation parameters including propulsion speed, cutter head torque and hydraulic cylinder pressure.

5. A tunneling equipment position calibration system using a cursor frame according to claim 3, characterized in that: The machine learning optimization unit is used to build and deploy a random forest and neural network fusion model, taking the error growth trend value Wczs, the dynamic environment data set and the compensation parameter set as input; outputs the optimal compensation parameter weight coefficient, and updates the compensation parameter mapping relationship in real time through the online learning mechanism.

6. A tunneling equipment position calibration system using a cursor frame according to claim 1, characterized in that: The environmental perception module includes a geological condition quantification unit, an environmental analysis unit, and a deep learning adjustment unit; The geological condition quantification unit is used to collect geological characteristic data including stratum hardness, rock type, groundwater level, and structural differences in coal mine tunnels in real time through multi-sensor data fusion technology to construct a geological data set; extract the geological data set and perform dimensionless processing, including rock hardness value Geo, rock type number Pha, groundwater level depth Mul, and fault displacement value Ion, and calculate the environmental adaptability index Szp through the following formula: ； In the formula, They represent the rock hardness value Geo, the number of rock types Pha, the groundwater level depth Mul and the fault displacement value Ion respectively; n represents the number of variables involved in the calculation.

7. A tunneling equipment position calibration system using a cursor frame according to claim 6, characterized in that: The environmental analysis unit compares and evaluates the preset environmental adaptability threshold E with the environmental adaptability index Gjs to determine the environmental adaptability. The specific evaluation contents are as follows: If the environmental adaptability index Gjs ≥ environmental adaptability threshold E, the environmental adaptability is judged to be "qualified", indicating that the geological conditions in the coal mine tunnel have little impact on the equipment calibration system, and the equipment continues to use the current calibration parameters for positioning; If the environmental adaptability index Gjs is less than the environmental adaptability threshold E, the environmental adaptability is judged as "unqualified", indicating that the geological conditions in the coal mine tunnel may cause the calibration system to be inaccurate, and the deep learning model needs to be started for dynamic adjustment.

8. A tunneling equipment position calibration system using a cursor frame according to claim 6, characterized in that: The deep learning adjustment unit analyzes the impact of complex geological conditions in coal mine tunnels on the calibration system and dynamically adjusts the calibration parameters based on real-time collected environmental data. The specific process is as follows: When the environmental adaptability assessment is "qualified", maintain the existing calibration parameters; When the environmental adaptability assessment is "unqualified", the deep learning model is started to learn the relationship between the geological characteristics in the coal mine tunnel and the calibration error, adjust the calibration parameters in real time, and optimize the fusion weight of the inertial navigation and lidar system.

9. A tunneling equipment position calibration system using a cursor frame according to claim 1, characterized in that: The interference compensation module is used to receive the output data of the inertial navigation system, including acceleration-related data, angular velocity-related data and attitude information-related data, and then extract the acceleration Ijs, angular velocity Isd, device attitude angle Izt, magnetic field intensity Icc and working environment temperature Iwd and perform dimensionless processing to calculate the inertial navigation system error coefficient Ins. The specific calculation formula is as follows: 。 10. A tunneling equipment position calibration system using a cursor frame according to claim 1, characterized in that: The interference compensation module presets the inertial navigation system error threshold R and the inertial navigation system error coefficient Ins for comparative evaluation. The specific contents are as follows: When the inertial navigation system error coefficient Ins ≤ the inertial navigation system error threshold R, it means that the system error is within an acceptable range and the inertial navigation system is in normal working condition; When the inertial navigation system error coefficient Ins> the inertial navigation system error threshold R, it means that the system error exceeds the acceptable range, the system error is abnormal, and the navigation accuracy is inaccurate. At this time, compensation processing is performed. The specific compensation processing is as follows: When the error coefficient Ins of the inertial navigation system exceeds the error threshold R of the inertial navigation system, the electromagnetic interference signal is first processed through filtering algorithms including Kalman filtering and particle filtering, especially the electromagnetic wave interference that affects the performance of the inertial navigation system; during the filtering process, the electromagnetic interference signal is identified and filtered out in the real-time collected data, and dynamic correction is performed according to the relevant data of the inertial sensor and the external environment information; secondly, the system parameters are adjusted through the compensation algorithm according to the error of the inertial navigation system; then the error of the inertial navigation system is corrected by adjusting the calibration parameters, angle deviation and position error of the sensor; finally, the electromagnetic interference is compensated.

11. A tunneling equipment position calibration system using a cursor frame according to claim 1, characterized in that: The feedback optimization module includes a data preprocessing subunit and an intelligent optimization subunit; The data preprocessing subunit is used to receive the time series data stream of the calibration error evaluation index Gz, the signal quality optimization index Szp and the inertial navigation system error coefficient Ins, and extract the latest 20 sets of historical data through the sliding window mechanism; then build and use the Prophet time series model to detect abnormal fluctuation points, and mark them as key abnormal events when the following conditions are met: The calibration error evaluation index Gz exceeds the threshold Gz_threshold for three consecutive sampling values ​​and the gradient change rate is greater than 5% / s; The absolute value of the correlation coefficient between the signal quality optimization index Szp and the inertial navigation system error coefficient Ins is less than 0.3 and lasts for more than 10 seconds; Finally, a weighted error trend analysis report is generated, in which the weight coefficients of the calibration error evaluation index Gz, the signal quality optimization index Szp and the inertial navigation system error coefficient Ins are dynamically allocated according to the current geological type.

12. A tunneling equipment position calibration system using a cursor frame according to claim 11, characterized in that: The intelligent optimization subunit is used to build a parameter optimization model based on a multi-agent reinforcement learning framework. The parameter optimization model is used to generate a set of correction suggestions through offline training and online fine-tuning, including: When the calibration error evaluation index Gz is detected as the dominant anomaly, the calibration strategy is adjusted first: shorten the laser target sampling interval to 200ms, and increase the gyroscope data weight by 40%; when the signal quality optimization index Szp abnormality persists, switch to the multi-band signal fusion mode, and inject spread spectrum coding check bits into the UWB channel; trigger composite compensation for the inertial navigation system error coefficient Ins abnormality: enable CEEMDAN filtering and inertial navigation zero bias online calibration at the same time.

13. A method for calibrating the position of a tunneling device using a cursor frame, a system for calibrating the position of a tunneling device using a cursor frame according to any one of claims 1 to 12, characterized in that: The following steps are involved: Step 1: Collect coal mine tunnel environmental data based on multi-sensor fusion technology, including illumination, electromagnetic field, temperature and humidity, and geological structure characteristics, and use adaptive filtering and hierarchical modeling to construct a dynamic environmental data set; Step 2: Calculate the calibration error evaluation coefficient Gz based on the dynamic environment data set, use the adaptive path calibration algorithm to predict the path deviation and calculate the error trend value Wczs, adjust the calibration strategy based on multi-objective optimization, and optimize the error compensation parameters in combination with the machine learning model; Step 3: Combining inertial navigation, lidar SLAM and visual recognition technology, a multimodal fusion algorithm is used to achieve position calibration, and the cursor frame weight is adjusted based on the calibration error evaluation index Gz; Step 4: Use multi-sensor data fusion technology to perceive the environment in the coal mine tunnel, further perceive the geological characteristics including stratum hardness, rock type, groundwater level, and structural differences, and calculate and evaluate the environmental adaptability index Szp based on real-time geological data; use a deep learning model to analyze the impact of complex geological conditions in coal mine tunnels on the calibration system, dynamically adjust calibration parameters, and optimize the fusion weight of inertial navigation and lidar systems; Step 5: Calculate the error coefficient Ins of the inertial navigation system, use a deep learning model to analyze the electromagnetic interference characteristics, combine time series prediction to optimize the compensation strategy, and adjust the weights of inertial navigation and lidar data through adaptive filtering; Step 6. Based on the evaluation contents of the calibration error evaluation index Gz, the signal quality optimization index Szp and the inertial navigation system error coefficient Ins, the time series anomaly detection is used to analyze the calibration historical data, automatically generate error trend analysis, and combine reinforcement learning to optimize the calibration parameters, and put forward correction suggestions, including adjusting the calibration strategy, optimizing the signal enhancement scheme and the interference compensation strategy.

Images