Safety management system based on centimeter-level positioning equipment of low-orbit satellite

Abstract

The invention discloses a safety management system based on low-orbit satellite centimeter-level positioning equipment, and relates to the technical field of satellite positioning equipment management, and the system comprises a low-orbit satellite positioning module which employs a phased-array antenna, achieves centimeter-level positioning, and is high in interference resistance; the data transmission module adopts a 5G and satellite link dual mode, and is stable in transmission and low in packet loss rate; the safety monitoring module is used for monitoring in real time and quickly transmitting an alarm by various high-precision sensors; the intelligent analysis module analyzes data through a machine learning algorithm to obtain a report; the monitoring center displays multiple information on a large screen and can process alarms in a grading manner; and the user interaction module is friendly in interaction and safe in authority management through biological recognition. The system is high in positioning precision, stable in data transmission, sensitive in safety monitoring, accurate in intelligent analysis, complete in monitoring center function, convenient in user interaction and safe in authority, effectively improves the multi-scene safety management efficiency, and is wide in application prospect.

Description

Technical Field

[0001] The present invention relates to the technical field of satellite positioning equipment management, and in particular to a safety management system based on centimeter-level positioning equipment of low-orbit satellites. Background Art

[0002] In terms of positioning technology, positioning systems based on ground base stations, such as GPS and mobile communication base station positioning, are easily blocked and reflected in the environment of high-rise buildings in cities, resulting in a significant decrease in positioning accuracy. The error can often reach several meters or even tens of meters, making it difficult to meet the requirements of centimeter-level positioning accuracy such as high-precision navigation for autonomous driving, precise operation guidance of agricultural machinery in precision agriculture, and precise control of large machinery on construction sites. At the same time, in remote areas, oceans, mountainous areas and other regions, ground base stations are insufficiently covered, and there are a large number of blind spots in positioning services, making it impossible to achieve continuous and stable positioning monitoring.

[0003] As for security management systems, most of the existing ones are based on local area networks or single sensor technology. For example, internal security monitoring in enterprises mostly relies on cameras and limited sensor deployment within the local area network, and can only be monitored within a fixed place. Once the target object leaves the area, security management will be out of touch. Moreover, when faced with a variety of complex security risk factors, traditional security management systems lack comprehensive integration and in-depth analysis capabilities. For example, for dangerous goods in the transportation process, it is difficult to simultaneously conduct real-time and accurate monitoring of their location and transportation environment (temperature, vibration, leakage, etc.) and comprehensive risk assessment. Often, only a simple warning can be issued for a single risk indicator, and it is impossible to predict in advance the serious security accidents that may be caused by the coupling of multiple risk factors.

[0004] In addition, with the booming development of global logistics, transportation, field operations and other industries, the safety management of mobile targets has become a key issue that needs to be solved urgently. Traditional systems cannot provide reliable positioning and comprehensive security in a large-scale, dynamically changing environment, and the stability and timeliness of data transmission are difficult to meet the requirements, resulting in inefficient safety management and delayed accident response, which poses a huge potential threat to the safety of life and property as well as the ecological environment.

[0005] In summary, the existing positioning and safety management technologies can no longer meet the urgent needs of modern society for high-precision, wide-coverage, and intelligent safety management. Therefore, it is extremely important to develop a safety management system based on centimeter-level positioning equipment of low-orbit satellites. It can make full use of the advantages of low-orbit satellites, overcome the drawbacks of traditional technologies, and achieve high-precision positioning and all-round safety management of mobile targets around the world, providing solid technical support and guarantee for the safe and efficient operation of many industries. Summary of the invention

[0006] The present invention proposes a safety management system based on a low-orbit satellite centimeter-level positioning device to solve the problems mentioned in the above-mentioned prior art.

[0007] In order to achieve the above-mentioned purpose, the present invention adopts the following technical solution: a safety management system based on a centimeter-level positioning device of a low-orbit satellite, comprising: Low-orbit satellite positioning module: Integrates low-orbit satellite receiving antenna and high-precision positioning chip. The receiving antenna adopts phased array antenna technology, adjusts the antenna beam direction through electronic control, accurately tracks multiple low-orbit satellite signals, and simultaneously receives and processes at least 8 low-orbit satellite signals; the high-precision positioning chip is based on carrier phase differential technology, using the known coordinates of the base station and the propagation time difference of the satellite signal to calculate the position of the target object. The positioning error in an open environment does not exceed 3 cm, and the signal receiving sensitivity reaches -160dBm. Through signal enhancement algorithm and adaptive filtering technology, the positioning update frequency is not less than 10Hz, and the precise position information of the target object is obtained in real time; The signal enhancement algorithm adopts a signal enhancement method based on wavelet transform. First, the received signal is decomposed by wavelet to separate signal components of different frequencies. For low-frequency components, weighted amplification processing is performed according to the signal energy distribution characteristics to enhance the overall signal strength. For high-frequency components, noise interference is removed by threshold processing, and then the signal is reconstructed to improve the signal quality. The adaptive filtering technology uses the least mean square LMS adaptive filtering algorithm, with the minimum mean square value of the error between the reference signal and the received signal as the goal, dynamically adjusts the filter coefficient, and automatically optimizes the filtering parameters according to the real-time changes of the signal to filter out multipath interference and neighboring satellite interference. Data transmission module: 5G and low-orbit satellite communication link dual-mode data transmission is adopted. The 5G module supports two architectures: 5GNR independent networking SA and non-independent networking NSA. It has multiple-input multiple-output MIMO technology and exchanges data with multiple base stations at the same time. When the 5G network signal is good, 5G transmission is given priority; when the 5G signal is blocked or interrupted, it automatically switches to the low-orbit satellite communication link. The low-orbit satellite communication link adopts a modem and uses adaptive coding and modulation technology to dynamically adjust the coding method and modulation order according to the channel quality. At the same time, forward error correction coding and automatic retransmission request mechanism are introduced in the transmission process; Safety monitoring module: connects to a variety of sensors, including acceleration sensors manufactured using MEMS technology, with an internal integrated capacitive detection unit to monitor whether the target has been hit or has abnormal movement; temperature sensors use thermistor principles and use packaging technology to improve thermal response speed and stability; smoke sensors are based on the principle of photoelectric scattering, with a high-sensitivity photodiode and signal amplification circuit inside to detect smoke concentration. When an abnormal situation is detected, an alarm signal is immediately generated, and the relevant data and alarm information are transmitted to the monitoring center; Intelligent analysis module: Receives data from the low-orbit satellite positioning module and the safety monitoring module, and uses machine learning algorithms to analyze the data, including using the DBSCAN clustering algorithm to classify and identify the target object's motion trajectory, and determines whether it deviates from the normal path by building a density-connected data point cluster; uses a multi-layer perceptron neural network algorithm to perform pattern recognition on sensor data. The neural network includes an input layer, a hidden layer, and an output layer. The hidden layer uses the ReLU activation function to enhance the model's nonlinear expression ability, predict potential safety risks, and automatically generate a safety assessment report and response strategy recommendations based on the analysis results; Monitoring center: equipped with large display screen and intelligent monitoring software. The display screen adopts high-resolution liquid crystal display technology, with anti-glare and wide viewing angle characteristics. The intelligent monitoring software is based on distributed architecture design and uses cloud computing technology to achieve flexible allocation of data storage and computing resources. It can display the location information, movement trajectory, sensor monitoring data and safety assessment results of the target object in real time, and centrally monitor and manage multiple target objects. The monitoring software has data storage function and automatically processes alarm information in a graded manner according to preset rules, including giving priority to sending serious safety accident alarms to the mobile phones of relevant persons in charge. User interaction module: provides an interactive interface for managers and operators. The interactive interface is built with HTML5 and CSS3 technologies and can be accessed through a computer terminal or mobile device. It supports user login, permission management, system settings, viewing historical data and report operations. The speech recognition device builds an acoustic model based on a deep neural network, uses a convolutional neural network to extract speech features, and a recurrent neural network to process time series information.

[0008] Furthermore, the low-orbit satellite positioning module adopts multi-frequency signal receiving technology to receive L-band and S-band multi-frequency signals of low-orbit satellites, and further improves positioning accuracy and stability by fusing multi-frequency signals. When fusing multi-frequency signals, the least squares method is used to fit and optimize measurement data of different frequencies.

[0009] Furthermore, when the data transmission module transmits data in a low-orbit satellite communication link, it uses channel coding and interleaving technology to perform error correction coding on the data. The channel coding uses low-density parity check code LDPC, and efficient coding is achieved by constructing a sparse check matrix. The interleaving technology uses block interleaving to disrupt the data transmission order.

[0010] Furthermore, the sensor in the safety monitoring module adopts a low-power design and can work continuously for at least 1 year when powered by a battery. The sensor has an integrated power management chip, uses dynamic voltage regulation and sleep mode switching technology to reduce power consumption, and has a self-diagnosis function.

[0011] Furthermore, the intelligent analysis module adopts an incremental learning algorithm when training the machine learning algorithm, and continuously updates the model parameters according to the newly acquired data. The incremental learning algorithm adopts a model update strategy based on gradient descent, combined with regularization terms to prevent overfitting. There is no need to retrain the entire model, so that the system can quickly adapt to new security management scenarios and changes in target object characteristics.

[0012] Furthermore, the intelligent monitoring software of the monitoring center uses virtual reality VR or augmented reality AR technology to visualize the location and status information of the target object, uses three-dimensional modeling technology to build a virtual model of the target object and the environment, and maps real-time information to the virtual model through spatial positioning technology and sensor data fusion. Managers can immersively view detailed information through VR / AR devices.

[0013] Furthermore, the user interaction module adopts biometric technology in terms of authority management, including fingerprint recognition using a capacitive fingerprint sensor to obtain a fingerprint image by detecting the change in capacitance between the finger skin and the sensor surface, and face recognition using a deep learning algorithm to build a facial feature model to perform dual authentication of the user's identity.

[0014] Compared with the prior art, the present invention has the following beneficial effects: In terms of positioning accuracy, with the help of low-orbit satellites and advanced technologies, the positioning error in open areas is within 3 cm. In complex urban environments, the accuracy is greatly improved due to technologies such as multi-frequency fusion, providing reliable location information for precise navigation of autonomous driving, high-precision mapping, etc. The positioning update frequency is above 10 Hz, ensuring the real-time nature of location data.

[0015] The data transmission is highly stable, with dual modes of 5G and low-orbit satellite communication links. The 5G signal is transmitted at high speed when it is optimal, and the satellite link is seamlessly connected when blocked. The packet loss rate is less than 0.1%, ensuring continuous data flow. Monitoring and location data can be stably transmitted even in remote mountainous areas or offshore operation scenarios.

[0016] Safety monitoring is comprehensive and sensitive, with a variety of high-precision sensors monitoring acceleration, temperature, smoke, etc. For example, the acceleration accuracy of 0.01g can accurately capture abnormal movement, and the abnormal alarm can be issued within 2 seconds and transmitted to the monitoring center, effectively preventing accidents.

[0017] The accuracy rate of intelligent analysis exceeds 95%, and the machine learning algorithm mines the value of data, accurately classifies trajectories and predicts risks. Reports and strategies are generated within 5 seconds to help managers respond to safety hazards in advance.

[0018] The monitoring center is powerful, with a high-resolution large screen and intelligent software to display multiple information, convenient storage and management, alarm classification processing, and response time within 30 seconds, which improves management efficiency. It is user-friendly, with smooth cross-platform interface operation, voice recognition accuracy exceeding 90%, and biometrics to ensure authority security, which is convenient for use in different scenarios. The overall system promotes positioning and security management towards high precision and intelligence, and has broad application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 This is a schematic block diagram of a safety management system based on a low-orbit satellite centimeter-level positioning device proposed by the present invention. DETAILED DESCRIPTION

[0020] 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.

[0021] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.

[0022] In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined. In addition, the terms "installed", "connected" and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, and it can be the internal connection of two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances. The present invention will be further described in detail below in conjunction with the accompanying drawings.

[0023] Reference Figure 1:A safety management system based on low-orbit satellite centimeter-level positioning equipment, including: Low-orbit satellite positioning module: Integrates low-orbit satellite receiving antenna and high-precision positioning chip. The receiving antenna adopts phased array antenna technology, adjusts the direction of the antenna beam through electronic control, accurately tracks multiple low-orbit satellite signals, and can simultaneously receive and process at least 8 low-orbit satellite signals. The high-precision positioning chip is based on carrier phase differential technology, using the known coordinates of the base station and the time difference between the propagation of satellite signals to accurately calculate the position of the target object. The positioning error in an open environment does not exceed 3 cm, and the signal receiving sensitivity reaches -160dBm. Even in complex geographical environments such as mountainous areas and canyons, as well as severe weather conditions such as heavy rain and sandstorms, with the help of signal enhancement algorithms and adaptive filtering technology, it can still work stably, with a positioning update frequency of no less than 10Hz, and obtain the precise location information of the target object in real time; The signal enhancement algorithm adopts a signal enhancement method based on wavelet transform. First, the received signal is decomposed by wavelet to separate signal components of different frequencies. For low-frequency components, weighted amplification processing is performed according to the signal energy distribution characteristics to enhance the overall signal strength. For high-frequency components, noise interference is removed by threshold processing, and then the signal is reconstructed to improve the signal quality. The adaptive filtering technology uses the least mean square LMS adaptive filtering algorithm, with the minimum mean square value of the error between the reference signal and the received signal as the goal, dynamically adjusts the filter coefficient, and automatically optimizes the filtering parameters according to the real-time changes of the signal to filter out multipath interference and neighboring satellite interference.

[0024] Data transmission module: 5G and low-orbit satellite communication link dual-mode data transmission is adopted. The 5G module supports both 5GNR independent networking (SA) and non-independent networking (NSA) architectures, and has multiple-input multiple-output (MIMO) technology. It can interact with multiple base stations at the same time. When the 5G network signal is good, 5G transmission is given priority, and the data transmission rate can reach more than 1Gbps; when the 5G signal is blocked or interrupted, it automatically and seamlessly switches to the low-orbit satellite communication link. The low-orbit satellite communication link adopts a high-speed modem and uses adaptive coding and modulation technology. The coding method and modulation order are dynamically adjusted according to the channel quality. The data transmission rate is not less than 10Mbps. At the same time, forward error correction coding and automatic retransmission request mechanism are introduced during the transmission process. The data packet loss rate is less than 0.1%, ensuring that the positioning data and other related information are transmitted to the monitoring center in real time and stably.

[0025] Safety monitoring module: connects multiple sensors, such as the acceleration sensor manufactured using micro-electromechanical system (MEMS) technology, with an internal integrated high-precision capacitive detection unit, which can accurately measure the target object's motion acceleration, with a measurement range of -10g to 10g and an accuracy of 0.01g, and is used to monitor whether the target has been hit or has abnormal movement; the temperature sensor uses the thermistor principle and a special packaging process to improve the thermal response speed and stability, and can monitor the ambient temperature in real time with a measurement accuracy of ±0.5°C; the smoke sensor is based on the principle of photoelectric scattering, with a high-sensitivity photodiode and signal amplification circuit inside, which can detect smoke concentration with a detection accuracy of 0.1%obs / m. When an abnormal situation is detected, such as a sudden change in acceleration, excessive temperature or excessive smoke, an alarm signal is immediately generated, and the relevant data and alarm information are transmitted to the monitoring center together, and the alarm information transmission delay does not exceed 2 seconds.

[0026] Intelligent analysis module: Receives data from the low-orbit satellite positioning module and the safety monitoring module, and uses machine learning algorithms to analyze the data. For example, it uses the DBSCAN clustering algorithm to classify and identify the motion trajectory of the target object, and determines whether it deviates from the normal path by constructing density-connected data point clusters; it uses a multi-layer perceptron neural network algorithm to perform pattern recognition on sensor data. The neural network consists of an input layer, a hidden layer, and an output layer. The hidden layer uses the ReLU activation function to enhance the nonlinear expression ability of the model and predict potential safety risks. The analysis accuracy of the intelligent analysis module is over 95%, and it can automatically generate a safety assessment report and response strategy recommendations based on the analysis results. The report generation time does not exceed 5 seconds.

[0027] Monitoring center: equipped with a large display screen and intelligent monitoring software. The display screen adopts high-resolution LCD display technology and has anti-glare and wide viewing angle characteristics. The intelligent monitoring software is designed based on a distributed architecture and uses cloud computing technology to achieve flexible allocation of data storage and computing resources. It can display the location information, movement trajectory, sensor monitoring data and security assessment results of the target object in real time, and can centrally monitor and manage multiple target objects. The monitoring software has data storage function with a storage capacity of not less than 10TB and a storage time of not less than 1 year. At the same time, it can automatically classify the alarm information according to preset rules, such as pushing serious safety accident alarms to the mobile phone of the relevant person in charge first, and the response time for the person in charge to receive the alarm information shall not exceed 30 seconds.

[0028] User interaction module: provides an interactive interface for managers and operators. The interactive interface is built with HTML5 and CSS3 technology, has good cross-platform compatibility, and can be accessed through computer terminals or mobile devices. It supports user login, permission management, system settings, viewing historical data and reports, etc. The interactive interface operation response time is less than 1 second, and it has voice interaction function. The voice recognition device builds an acoustic model based on a deep neural network, uses a convolutional neural network to extract voice features, and a recurrent neural network to process time series information. The voice recognition accuracy is higher than 90%, which is convenient for users to use the system conveniently in different scenarios.

[0029] In the present invention, the low-orbit satellite positioning module adopts multi-frequency signal receiving technology, which can receive multi-frequency signals such as L-band and S-band of low-orbit satellites, and further improve the positioning accuracy and stability by fusing the multi-frequency signals. When fusing the multi-frequency signals, the least squares method is used to fit and optimize the measurement data of different frequencies, and the positioning accuracy is improved by more than 20% in complex urban environments.

[0030] The data transmission module of the present invention attaches great importance to the accuracy and reliability of data when using the low-orbit satellite communication link to transmit data, and adopts advanced channel coding and interleaving technology for this purpose.

[0031] In terms of channel coding, low-density parity-check code (LDPC) is used. Its core principle is to achieve efficient coding by constructing a sparse check matrix. First, the check matrix is ​​carefully designed according to the preset coding rules and error correction capability requirements. This check matrix has the characteristics of sparsity, that is, the number of non-zero elements is relatively small, which effectively controls the computational complexity of the coding process. In the actual coding operation, the data to be transmitted is subjected to matrix operations with the check matrix according to a specific grouping method, thereby generating coded data containing redundant information. This redundant information can be used at the receiving end to detect and correct errors that may occur during the transmission process.

[0032] The interleaving technology uses block interleaving. Specifically, the data to be transmitted is first divided into several data blocks of fixed size. Then, according to the pre-set interleaving rules, these data blocks are rearranged and combined to completely disrupt the original transmission order of the data. For example, the data blocks can be written into a two-dimensional matrix by row and then read out by column to disrupt the data order. The cleverness of this interleaving method is that when the data encounters a sudden error during transmission, the error bits that originally appeared continuously will be dispersed to different locations, avoiding the huge challenge caused by the concentration of errors to the error correction capability.

[0033] Through the synergy of channel coding and interleaving technology, the coding gain can reach more than 3dB. This means that the anti-interference ability of the signal during transmission is significantly enhanced, effectively reducing the bit error rate during data transmission. Compared with the case where these technologies are not used, the bit error rate is reduced by more than 50%, thus providing a solid guarantee for the stable and reliable transmission of data, ensuring that the monitoring center can accurately receive key data from various monitoring points, and laying a solid foundation for the effective operation of the entire security management system.

[0034] In the safety monitoring module of the present invention, the sensor is designed with low power consumption as the core highlight, ensuring that it can work stably and continuously for at least 1 year in battery-powered mode. An advanced power management chip is carefully integrated inside it. As the key hub of power consumption control, the chip adopts dynamic voltage regulation technology to intelligently and dynamically adjust the supply voltage according to the real-time workload and operating status of the sensor. For example, when the sensor is in a low data acquisition frequency or a relatively stable monitoring environment, the supply voltage is automatically reduced to the minimum effective working level, thereby significantly reducing energy consumption.

[0035] At the same time, using sleep mode switching technology, when the sensor does not detect obvious environmental changes or data fluctuations within the preset time interval, it automatically switches to sleep mode. At this time, only the extremely low-power wake-up circuit is kept in working state to wait for possible abnormal trigger signals. Once an abnormal situation occurs, the wake-up circuit responds quickly, allowing the sensor to return to normal working state and collect data in time.

[0036] In addition, the sensor has a highly reliable self-diagnosis function. Through the built-in self-test program and redundant detection circuit, the key components of the sensor, such as sensitive elements, signal conditioning circuits, data transmission interfaces, etc., are periodically tested. During the detection process, a variety of methods such as comparison of standard signals, threshold judgment, and data consistency analysis are used to ensure that the accuracy of sensor fault detection reaches more than 98%. Once the sensor is confirmed to have a fault, the fault alarm information containing detailed information such as the fault type, location information, and fault occurrence time is automatically sent to the monitoring center through an independent fault alarm communication channel, so that the monitoring center can take corresponding maintenance or emergency measures in time to ensure the continued effectiveness and reliability of the entire safety monitoring system.

[0037] The intelligent analysis module of the present invention innovatively uses an incremental learning algorithm in the training process of the machine learning algorithm. The incremental learning algorithm is built around a model update strategy based on gradient descent. When new data is received, the gradient effect of the new data on the model parameters is first calculated based on the characteristic distribution of the data and the existing model parameters. By setting a specific learning rate, the model parameters are gradually fine-tuned in the direction of gradient descent, so that the model can quickly adapt to information changes in the new data.

[0038] At the same time, in order to effectively prevent the model from overfitting, the algorithm cleverly combines regularization terms. The L2 regularization method is used to constrain the model parameters, that is, a penalty term of the sum of squared parameters is added to the loss function. This measure limits the excessive growth of model parameters and avoids overfitting of the model on the training data, thereby ensuring that the model still has good generalization ability when facing unknown data.

[0039] During the entire process, since only local model parameters need to be updated for new data rather than retraining the entire model architecture, the time required for model updates is greatly shortened. After actual testing and comparison, the model update time is shortened by more than 80% compared to the traditional full training method. In this way, when the security management system faces new scenarios, such as new monitoring area types, changes in the target object operation mode, etc., the intelligent analysis module can quickly adjust the model based on the new data features, and efficiently provide the system with accurate and up-to-date security analysis and decision-making basis, which effectively improves the system's adaptability and response speed to complex and changing security management needs.

[0040] In the present invention, the intelligent monitoring software equipped in the monitoring center cleverly integrates virtual reality (VR) and augmented reality (AR) technologies to achieve visualization and optimization of the location and status information of the target object. When applying three-dimensional modeling technology, for the target object, first use laser radar scanning technology to obtain its accurate spatial point cloud data, and then combine multi-view image acquisition and computer vision algorithms to reconstruct the target model with rich detailed textures. For environmental modeling, based on satellite remote sensing image data, geographic information system (GIS) data and terrain data collected on the spot, a realistic virtual model of the environment is constructed through digital elevation model (DEM) and texture mapping technology.

[0041] In terms of spatial positioning technology, a solution that integrates the global navigation satellite system (GNSS), inertial measurement unit (IMU) and ultra-wideband (UWB) positioning technology is adopted. GNSS provides macroscopic location information, IMU monitors the attitude changes of the target object in real time, and UWB positioning technology achieves high-precision relative positioning in a local area. The three work together to ensure that the location information of the target object in different scenarios can be accurately obtained and updated in real time.

[0042] In the sensor data fusion phase, a data fusion algorithm based on Kalman filtering is designed. The location information from the low-orbit satellite positioning module, the acceleration sensor, temperature sensor, pressure sensor and other data in the safety monitoring module are fused and processed. First, the time synchronization and coordinate conversion of different sensor data are performed to make them under a unified time and space reference. Then, the Kalman filtering algorithm is used to optimally estimate the data, effectively eliminate noise interference and fill in data gaps, and accurately map the processed data to the corresponding target object and environmental position on the virtual model.

[0043] With the help of professional VR / AR head-mounted display devices and interactive handles, managers can freely shuttle through virtual scenes. The handles can be used to zoom, translate, rotate and other functions of the scene. Clicking on the virtual model can view the detailed attribute information, historical operation data and real-time sensor monitoring data curves of the target object. This immersive monitoring method greatly enhances the intuitiveness and interactivity of monitoring. According to statistical analysis of a large number of actual application cases, compared with the traditional two-dimensional monitoring mode, the safety management efficiency has been significantly improved by more than 30%, which has brought a qualitative leap in safety management work.

[0044] The user interaction module of the present invention attaches great importance to security and confidentiality in the authority management link, and adopts advanced biometric technology for this purpose. Among them, the fingerprint recognition part uses a precise capacitive fingerprint sensor, on the surface of which a large number of tiny capacitive sensing units are distributed. When the user's finger touches the sensor, the capacitance value of each capacitive sensing unit will change due to the difference in distance between the finger skin and the sensor surface and the ups and downs of the skin texture. These capacitance changes are accurately measured and collected through a special circuit system, and then a high-resolution, high-definition fingerprint image is constructed. The image acquisition process has extremely high sensitivity and can capture the subtle features of the fingerprint, such as ridges, valleys, bifurcation points, etc., laying a solid foundation for subsequent accurate identification.

[0045] In terms of face recognition, deep learning algorithms are used to build a face feature model. First, a high-definition camera is used to collect the user's face image. In the image preprocessing stage, grayscale, noise reduction, image enhancement and other operations are performed to improve the image quality. Then, a face detection algorithm based on a convolutional neural network (CNN) is used to quickly and accurately locate the position of the face in the image. Subsequently, through advanced feature extraction networks such as the deep residual network (ResNet), key features of the face, such as the shape, position, texture and other deep feature information of the eyes, nose, mouth and other parts, are extracted, and these feature information is mapped to a high-dimensional feature space to build a highly distinguishable face feature model. During identity authentication, the real-time collected face features are compared and matched with the pre-stored feature model, and the similarity between the two is calculated using measurement methods such as cosine similarity. Only when the similarity exceeds the set threshold, the identity authentication is determined to be successful. Through the dual authentication mechanism of fingerprint recognition and face recognition, the two complement each other and verify together, effectively avoiding the risk of misidentification that may exist in a single biometric method, thereby ensuring that the identity authentication accuracy of system operations reaches more than 99%, building a solid security line for the entire security management system, strictly restricting the access and operation of illegal users, and effectively ensuring the security and confidentiality of various data and functions in the system.

[0046] The above are only preferred specific implementation modes of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical solutions and inventive concepts of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A safety management system based on low-orbit satellite centimeter-level positioning equipment, characterized in that: include: Low-orbit satellite positioning module: Integrates a low-orbit satellite receiving antenna and a high-precision positioning chip. The receiving antenna adopts phased array antenna technology, adjusts the antenna beam direction through electronic control, accurately tracks multiple low-orbit satellite signals, and simultaneously receives and processes at least 8 low-orbit satellite signals; The high-precision positioning chip is based on carrier phase differential technology. It uses the known coordinates of the base station and the time difference between satellite signal propagation to calculate the position of the target object. The positioning error in an open environment does not exceed 3 cm, and the signal receiving sensitivity reaches -160dBm. Through signal enhancement algorithm and adaptive filtering technology, the positioning update frequency is not less than 10Hz, and the precise position information of the target object is obtained in real time. The signal enhancement algorithm adopts a signal enhancement method based on wavelet transform. First, the received signal is decomposed by wavelet to separate signal components of different frequencies. For low-frequency components, weighted amplification processing is performed according to the signal energy distribution characteristics to enhance the overall signal strength. For high-frequency components, noise interference is removed by threshold processing, and then the signal is reconstructed to improve the signal quality. The adaptive filtering technology uses the least mean square LMS adaptive filtering algorithm, with the minimum mean square value of the error between the reference signal and the received signal as the goal, dynamically adjusts the filter coefficient, and automatically optimizes the filtering parameters according to the real-time changes of the signal to filter out multipath interference and neighboring satellite interference. Data transmission module: 5G and low-orbit satellite communication link dual-mode data transmission is adopted. The 5G module supports two architectures: 5GNR independent networking SA and non-independent networking NSA. It has multiple-input multiple-output MIMO technology and exchanges data with multiple base stations at the same time. When the 5G network signal is good, 5G transmission is given priority; when the 5G signal is blocked or interrupted, it automatically switches to the low-orbit satellite communication link. The low-orbit satellite communication link adopts a modem and uses adaptive coding and modulation technology to dynamically adjust the coding method and modulation order according to the channel quality. At the same time, forward error correction coding and automatic retransmission request mechanism are introduced in the transmission process; Safety monitoring module: connects to a variety of sensors, including acceleration sensors manufactured using MEMS technology, with an internal integrated capacitive detection unit to monitor whether the target has been hit or has abnormal movement; temperature sensors use thermistor principles and use packaging technology to improve thermal response speed and stability; smoke sensors are based on the principle of photoelectric scattering, with a high-sensitivity photodiode and signal amplification circuit inside to detect smoke concentration. When an abnormal situation is detected, an alarm signal is immediately generated, and the relevant data and alarm information are transmitted to the monitoring center; Intelligent analysis module: Receives data from the low-orbit satellite positioning module and the safety monitoring module, and uses machine learning algorithms to analyze the data, including using the DBSCAN clustering algorithm to classify and identify the target object's motion trajectory, and determines whether it deviates from the normal path by building a density-connected data point cluster; uses a multi-layer perceptron neural network algorithm to perform pattern recognition on sensor data. The neural network includes an input layer, a hidden layer, and an output layer. The hidden layer uses the ReLU activation function to enhance the model's nonlinear expression ability, predict potential safety risks, and automatically generate a safety assessment report and response strategy recommendations based on the analysis results; Monitoring center: equipped with large display screen and intelligent monitoring software. The display screen adopts high-resolution liquid crystal display technology, with anti-glare and wide viewing angle characteristics. The intelligent monitoring software is based on distributed architecture design and uses cloud computing technology to achieve flexible allocation of data storage and computing resources. It can display the location information, movement trajectory, sensor monitoring data and safety assessment results of the target object in real time, and centrally monitor and manage multiple target objects. The monitoring software has data storage function and automatically processes alarm information in a graded manner according to preset rules, including giving priority to sending serious safety accident alarms to the mobile phones of relevant persons in charge. User interaction module: provides an interactive interface for managers and operators. The interactive interface is built with HTML5 and CSS3 technologies and can be accessed through a computer terminal or mobile device. It supports user login, permission management, system settings, viewing historical data and report operations. The speech recognition device builds an acoustic model based on a deep neural network, uses a convolutional neural network to extract speech features, and a recurrent neural network to process time series information.

2. The safety management system based on low-orbit satellite centimeter-level positioning equipment according to claim 1 is characterized in that: The low-orbit satellite positioning module adopts multi-frequency signal receiving technology to receive L-band and S-band multi-frequency signals of low-orbit satellites. By fusing the multi-frequency signals, the positioning accuracy and stability are further improved. When the multi-frequency signals are fused, the least squares method is used to fit and optimize the measurement data of different frequencies.

3. The safety management system based on low-orbit satellite centimeter-level positioning equipment according to claim 1 is characterized in that: When the data transmission module transmits data in a low-orbit satellite communication link, channel coding and interleaving technology are used to perform error correction coding processing on the data. The channel coding adopts low-density parity check code LDPC and realizes efficient coding by constructing a sparse check matrix. The interleaving technology adopts block interleaving to disrupt the data transmission order.

4. The safety management system based on low-orbit satellite centimeter-level positioning equipment according to claim 1 is characterized in that: The sensor in the safety monitoring module adopts a low-power design and can work continuously for at least 1 year when powered by a battery. The sensor has an integrated power management chip, uses dynamic voltage regulation and sleep mode switching technology to reduce power consumption, and has a self-diagnosis function.

5. The safety management system based on low-orbit satellite centimeter-level positioning equipment according to claim 1 is characterized in that: When training the machine learning algorithm, the intelligent analysis module adopts an incremental learning algorithm to continuously update the model parameters according to the newly acquired data. The incremental learning algorithm adopts a model update strategy based on gradient descent, combined with regularization terms to prevent overfitting. There is no need to retrain the entire model, so that the system can quickly adapt to new security management scenarios and changes in target object characteristics.

6. The safety management system based on low-orbit satellite centimeter-level positioning equipment according to claim 1 is characterized in that: The intelligent monitoring software of the monitoring center adopts virtual reality VR or augmented reality AR technology to visualize the location and status information of the target object, uses three-dimensional modeling technology to build a virtual model of the target object and the environment, and maps real-time information to the virtual model through spatial positioning technology and sensor data fusion. Managers can view detailed information immersively through VR / AR devices.

7. The safety management system based on low-orbit satellite centimeter-level positioning equipment according to claim 1 is characterized in that: The user interaction module adopts biometric technology in terms of authority management, including fingerprint recognition using a capacitive fingerprint sensor to obtain a fingerprint image by detecting the change in capacitance between the finger skin and the sensor surface, and face recognition using a deep learning algorithm to build a facial feature model to perform dual authentication of the user's identity.

Images