Non-contact information acquisition system for moving target in mine

Abstract

The invention relates to the technical field of electrical digital data processing, in particular to a non-contact information acquisition system for a moving target in a mine, which comprises a multi-mode sensing terminal layer, a self-adaptive communication networking layer and an intelligent data fusion processing center. In combination with dynamic networking of a multi-system base station and a mobile relay node, multi-source data fusion and target trajectory reconstruction are realized. The reliability, the coverage range and the positioning precision of underground moving target information acquisition can be improved, and meanwhile, the system power consumption and the deployment cost are reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of electric digital data processing, and in particular to a non-contact information acquisition system for mobile targets in a mine. BACKGROUND

[0002] In the field of mine safety production and intelligent management, it is a key technical support to realize safety early warning, production scheduling and emergency rescue to accurately collect and monitor the real-time position, identity and state information of mobile targets such as underground personnel and vehicles.

[0003] The traditional management mode mainly relies on manual inspection and fixed-point clocking, which is difficult to meet the continuous acquisition demand of mobile target dynamic information in the complex underground environment. At the same time, the existing technology generally faces multiple challenges brought by the harsh underground environment, including that the electromagnetic signal is easily affected by multipath effect and attenuation interference in the complex roadway structure, resulting in identification blind area and positioning deviation; mobile targets are prone to data collision and missed reading when passing at high speed or in groups; the existing system mostly uses a single communication system, which is difficult to adapt to the multi-modal data fusion demand in the heterogeneous network environment underground. SUMMARY

[0004] The purpose of the present application is to provide a non-contact information acquisition system for mobile targets in a mine, which mainly solves the technical problem of identification blind area and positioning deviation caused by electromagnetic signal attenuation and multipath effect in the complex roadway environment in the prior art.

[0005] The present application provides a non-contact information acquisition system for mobile targets in a mine, comprising a perception terminal layer, a communication networking layer and a data fusion processing center; wherein the perception terminal layer comprises a plurality of perception terminals arranged at key nodes of the underground roadway, each perception terminal integrating a radio frequency identification read-write module, an inertial measurement unit and an ultrasonic ranging module; the radio frequency identification read-write module works in the ultra-high frequency band, adopts a time division multiple access and space division multiple access hybrid access mechanism, its transmission power is in the range of 0.1 milliwatt to 100 milliwatt, and is adaptively adjusted with 0.5 milliwatt as a step.

[0006] The communication networking layer is deployed in the multi-standard communication base station and mobile relay node in the roadway, which is used to establish a dynamic reconfigurable data transmission channel between the perception terminal and the data fusion processing center; the multi-standard communication base station is built-in communication protocol adaptive selection module, which monitors the signal strength and bit error rate of each communication channel in real time, and dynamically selects the optimal communication standard according to the preset channel quality evaluation function.

[0007] The data fusion processing center is arranged in a ground monitoring room, and is used for fused calculation and target state estimation on collected original data; the data fusion processing center at least comprises a target trajectory reconstruction module, which generates a continuous and smooth motion trajectory by using a cubic spline interpolation algorithm based on a historical position sequence of the mobile target.

[0008] In some embodiments, the mobile relay node is mounted on a downhole mine car or a patrol robot platform; the mobile relay node is equipped with a dual-mode communication interface, and working modes of the dual-mode communication interface include an active detection mode and an on-demand relay mode; in the active detection mode, the mobile relay node periodically scans a surrounding communication blind area and reports topological information; in the on-demand relay mode, the mobile relay node temporarily acts as a relay to extend network coverage in response to an instruction of the base station.

[0009] In some embodiments, the data fusion processing center further comprises a data fusion module configured with an extended Kalman filtering algorithm, and the extended Kalman filtering algorithm is implemented by taking the position and the speed of the mobile target as a state vector, and taking radio frequency identification ranging values, inertial measurement unit outputs and ultrasonic ranging values as observation vectors; and the optimal estimation value of the state vector is calculated through two recursive steps of state prediction and observation update.

[0010] In some embodiments, the system further comprises a distributed energy management unit composed of an energy acquisition module and a dynamic power consumption management module built in the perception terminal and the base station; the dynamic power consumption management module dynamically adjusts the duty cycle of each module according to the communication load and the priority of the perception task, so that the terminal can continuously work for not less than 30 days under the condition of no external power supply.

[0011] In some embodiments, the communication networking layer adopts a hierarchical distributed topology structure, which includes two layers of a backbone network and an access network; the backbone network is composed of a wired industrial Ethernet, and connects each regional core base station and a ground intelligent data fusion processing center; the access network is composed of wireless multi-standard base stations and mobile relay nodes, and forms a wireless Mesh network covering a mobile area underground; the network topology is dynamically maintained through a link state routing protocol, and a routing update period is 5 seconds.

[0012] In some embodiments, the deployment density of the perception terminal layer is adaptively adjusted according to a roadway structure and a target flow; at a roadway intersection and a main transportation roadway, the deployment interval of the perception terminal is not greater than 10 meters; in a general operation area, the deployment interval is not greater than 20 meters; each perception terminal is registered in the system through a unique identifier, and a geographic position coordinate thereof is pre-recorded in a terminal database of the intelligent data fusion processing center.

[0013] In some embodiments, the seamless switching process of the dual-mode communication interface of the mobile relay node between the industrial wireless local area network and the narrowband Internet of Things.

[0014] In some embodiments, the strategy of dynamically adjusting the duty cycle of each module includes: reducing the scanning frequency of the radio frequency identification read-write module during periods when the target traffic is low; reducing the wireless transmission power or extending the data transmission interval when the communication load is light.

[0015] In some embodiments, the link state routing protocol enables each node in the network to perceive the overall connection state of the network; the routing information update period is set to 5 seconds.

[0016] Compared with the prior art, the present application has the following beneficial effects: first, by integrating three sensing technologies of radio frequency identification, inertial measurement and ultrasonic ranging at the perception terminal layer, a redundant and complementary perception system is constructed, effectively overcoming the problems of missed reading and positioning deviation caused by single technology due to underground electromagnetic multipath fading or target shielding, significantly improving the integrity and reliability of information acquisition.

[0017] Secondly, the communication networking layer adopts multiple mode base stations and mobile relay nodes to work cooperatively, realizing seamless coverage and dynamic topology adaptation in the heterogeneous network environment underground, effectively solving the problems of communication blind area and link interruption, and optimizing the communication quality and energy consumption through the protocol adaptive selection mechanism.

[0018] Thirdly, the data fusion processing center adopts extended Kalman filtering and trajectory reconstruction algorithm to realize high-precision fusion of multi-source heterogeneous data and generation of target continuous motion trajectory, greatly improving the accuracy of mobile target state estimation.

[0019] Finally, the distributed energy management unit prolongs the service life of the terminal device through composite energy collection and dynamic power consumption control, significantly reduces the system deployment and maintenance cost, and provides sustainable technical support for the intelligent construction of the mine. BRIEF DESCRIPTION OF DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiment or prior art description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0021] Fig. 1 is the overall technical scheme architecture diagram of the system of the present application.

[0022] Fig. 2 is the core principle diagram of the fusion perception and adaptive cooperative communication system of the present application. DETAILED DESCRIPTION

[0023] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings. Figs. 1-2 It should be apparent that the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments.

[0024] Embodiment: Please refer to Fig. 1 The system aims to solve the identification blind area and positioning deviation caused by signal attenuation and multipath effect in the complex electromagnetic environment of mine roadway, and cope with the data collision and missing reading challenges when the mobile target high-speed group passes through, and overcome the adaptation difficulties of single communication system in the heterogeneous network environment in the mine, the power supply limit of equipment and the high deployment and maintenance cost.

[0025] It can be understood that the system core architecture is composed of three parts of the perception terminal layer, the communication networking layer and the data fusion processing center, and the high reliability, low power consumption, full-area continuous non-sensing identification and data acquisition of the mobile target in the mine are realized through multi-level technology cooperation.

[0026] The perception terminal layer, as the front end of system data acquisition, is composed of a large number of perception terminals arranged at key nodes in the underground roadway. Each perception terminal is a highly integrated embedded device, which integrates three core sensing modules: radio frequency identification reading and writing module, inertial measurement unit and ultrasonic ranging module.

[0027] The foregoing perception terminal is deployed with adjustable density according to the actual structure of the underground roadway and the expected mobile target flow. In the key areas such as roadway intersections and main transportation roadways, the deployment interval of the perception terminal is strictly controlled within 10 meters to ensure no blind area coverage; in the general operation area, the deployment interval can be relaxed to 20 meters. Each perception terminal is registered in the intelligent data fusion processing center through its globally unique identifier in the system initialization stage. Its accurate geographic position coordinates, including three-dimensional coordinate information, are pre-recorded and stored in the terminal database of the center. This deployment strategy ensures the comprehensive and efficient coverage of the perception network to the underground space.

[0028] The radio frequency identification reading and writing module specifically works in the ultra-high frequency band. The selection of this band balances the penetration ability and reading distance. The module adopts a hybrid access mechanism combining time division multiple access and space division multiple access. This mechanism effectively avoids the data collision problem that occurs when multiple mobile targets carrying radio frequency identification tags simultaneously enter the reading and writing area. The radio frequency transmission power of the module has a high degree of adaptive adjustment capability, with an adjustment range from 0.1 milliwatt to 100 milliwatt, and a minimum step unit of 0.5 milliwatt. The power adjustment strategy is based on real-time communication link quality assessment. When the reading and writing success rate decreases or the error rate increases, the module gradually increases the transmission power to improve the communication conditions; otherwise, when the signal quality is good, the power is reduced to save energy. The module internally solidifies the anti-collision algorithm to ensure orderly reading in a dense tag environment.

[0029] The inertial measurement unit is a key component in the sensing terminal responsible for capturing the motion state of the mobile target itself. The unit includes a 3-axis micro-electromechanical system accelerometer and a 3-axis micro-electromechanical system gyroscope. The measurement range of the accelerometer is set to plus or minus 16g, which can cover the severe acceleration changes that the equipment in the mine may experience. The measurement range of the gyroscope is set to plus or minus 2000 degrees per second, which is used to accurately perceive rotational motion. The sampling frequency of the inertial measurement unit is fixed at 100 Hz, which is sufficient to capture the dynamic characteristics of most underground mobile targets. The raw acceleration and angular velocity data collected are output together with accurate timestamp information after preliminary digital processing inside the unit.

[0030] The ultrasonic ranging module provides the sensing terminal with direct measurement capability of the relative distance of the mobile target. The working frequency of the module is set to 40 kHz, which achieves a good balance between propagation characteristics and anti-interference capability in the air. Its maximum effective detection distance is 10 meters, and the ranging accuracy within this range can reach 2 cm. The module calculates the distance from the target by transmitting ultrasonic pulses and receiving their echoes. In order to reduce the interference of environmental noise and multiple reflections, the module uses digital signal processing technology to filter and verify the echo signals. The ultrasonic ranging data, radio frequency identification reading events, and inertial measurement unit sampling data are preliminarily time-synchronized inside the terminal.

[0031] Please refer to Fig. 2, the adaptive communication networking layer constitutes the main line of system data transmission, responsible for reliably transmitting the raw data collected by the multi-modal sensing terminal layer to the intelligent data fusion processing center located on the ground. This layer is composed of two types of core devices: fixedly deployed multi-standard communication base stations and mobile relay nodes. Multi-standard communication base stations are usually deployed at specific locations on the tunnel wall or roof, forming relatively fixed network access points. Each base station hardware platform integrates the radio frequency front end and baseband processing unit of three communication protocols: industrial wireless local area network, narrowband Internet of Things, and ZigBee. This means that a single base station can simultaneously maintain and manage three different types of wireless networks.

[0032] The core of the base station is the communication protocol adaptive selection module, which continuously monitors the real-time state parameters of all available communication channels, mainly including received signal strength indication and bit error rate. The monitoring data is input into a preset channel quality evaluation function, which outputs a channel quality index between 0 and 1 by weighted comprehensive calculation of multiple input variables. This index quantifies the overall transmission quality of the current channel. The system sets the channel quality index threshold to 0.6, and when the monitored channel quality index of a certain standard is lower than this threshold, the adaptive selection module will immediately trigger the communication standard switching process. The switching process strives to be seamless, ensuring that ongoing data transmission will not be interrupted or lost due to switching. The base station will periodically broadcast the optimal communication standard recommendation to associated sensing terminals, guiding the terminals to access the wireless network with the best quality.

[0033] Mobile relay nodes are dynamic extensions of the adaptive communication networking layer, usually mounted on mobile platforms such as mine cars, rubber-tyred vehicles, or automatic inspection robots operating underground. These nodes are equipped with dual-mode communication interfaces, allowing them to quickly and seamlessly switch between industrial wireless local area networks and narrowband Internet of Things, two different standards with different coverage characteristics and bandwidths. Mobile relay nodes support two basic working modes. In the active detection mode, the node actively scans the wireless signal coverage of its surrounding area according to the preset period, identifies and locates communication blind spots or weak signal areas, and reports these topological change information to the nearest base station or directly back to the processing center. In the on-demand relay mode, the mobile relay node responds to the instructions of the network management unit or base station, temporarily acting as a relay forwarding station in areas with poor signal coverage, extending the effective coverage range of the network and ensuring that the data of the sensing terminal can be successfully returned.

[0034] The communication networking layer of the system adopts a hierarchical distributed topology to organize its network elements, which is clearly divided into two levels of backbone network and access network. The backbone network is composed of high-reliability wired industrial Ethernet, which connects the core base stations in each area of the mine with the intelligent data fusion processing center on the ground, providing a high-speed and stable data backbone. The access network is composed of the wireless multi-mode base stations and mobile relay nodes described above, which form a dynamic and self-organizing wireless Mesh network covering the mobile operation area of the mine. The maintenance of the network topology relies on a link state routing protocol, which enables each node in the network to perceive the overall connection state of the network. The routing information update period is set to 5 seconds, which is sufficient to track the dynamic changes of the network topology caused by the movement of equipment and environmental changes in the mine, while avoiding excessive protocol overhead.

[0035] The data fusion processing center is deployed in the security server cluster of the ground monitoring computer room, which is composed of the core target trajectory reconstruction module and the auxiliary data preprocessing module and multi-source data fusion module. The data preprocessing module is responsible for receiving and uploading the original sensor data stream from the adaptive communication networking layer, and its primary task is to clean and standardize these data, including applying digital filters to the data of the inertial measurement unit and the ultrasonic ranging module to reduce noise and eliminate abnormal pulses; more importantly, the module performs strict timestamp alignment operations, as the data comes from different sensing terminals and different sensing modules, and the transmission path and delay may be different. The preprocessing module uses the time reference provided by the high-precision network time protocol server to label all input data with a uniform time label and align them to a common time axis, laying the foundation for subsequent fusion processing.

[0036] The multi-source data fusion module uses the extended Kalman filter algorithm as its data fusion framework, in which the state of the moving target is defined as a state vector, usually containing the position coordinates and velocity components of the target in three-dimensional space. The observation vector of the system comes from three different sensing sources: the approximate distance information or presence detection provided by the radio frequency identification read-write module, the acceleration and angular velocity data provided by the inertial measurement unit, and the accurate relative distance provided by the ultrasonic ranging module.

[0037] The execution of the extended Kalman filter algorithm follows two recursive steps: state prediction and observation update.

[0038] In the state prediction step, the algorithm predicts the state of the target at the current time based on the state estimation of the target at the last time and the acceleration data measured by the inertial measurement unit through the system state transition model.

[0039] In the observation update step, the algorithm compares the predicted state with the actual observed radio frequency identification data, ultrasonic ranging data, etc. at the current time, and calculates the Kalman gain according to the uncertainty of each sensor, and then corrects the predicted state to obtain the optimal estimation value of the state vector at the current time.

[0040] The algorithm can effectively process the noise in the sensor data and give the best estimation of the target position and velocity. The two key parameter matrices in the algorithm: the process noise covariance matrix and the observation noise covariance matrix, are not fixed. They need to be dynamically calibrated according to the measured data in the specific environment of the mine. For example, in the area with strong electromagnetic interference, the noise covariance of the radio frequency identification observation will increase accordingly; and in the area with strong vibration, the noise parameters of the inertial measurement unit will also be adjusted. This dynamic calibration mechanism ensures that the filter can maintain the best performance under different working conditions.

[0041] The target trajectory reconstruction module receives the discrete target state estimation sequence output from the multi-source data fusion module. Due to sensor errors and data processing delays, these discrete position points may exhibit a certain degree of roughness or jitter. To generate a continuous and smooth motion trajectory that conforms to the actual motion law, the module uses a cubic spline interpolation algorithm. This algorithm uses time as the independent variable and the position coordinates of the target as the dependent variable to construct a set of piecewise cubic polynomial functions. These polynomials not only ensure that the function values are equal at known discrete state points, but also require the first derivative to be continuous, thereby ensuring that the generated trajectory curve is smooth and can better reflect the actual motion trend of the target, such as uniform speed, acceleration, or turning. The reconstructed continuous trajectory data is more convenient for the monitoring system to perform behavior analysis, collision warning, and path planning.

[0042] Further, the system can be continuously operated depending on the distributed energy management unit, which is not an independent physical entity, but its functional components are distributed in each sensing terminal and communication base station. It is mainly composed of two sub-functions: energy collection module and dynamic power consumption management module. The energy collection module uses composite energy collection technology, combining thermoelectric power generation and vibration energy collection. The thermoelectric power generation component uses the small difference between the environmental temperature in the mine tunnel and the temperature of the device itself or the human body temperature to generate electricity. The vibration energy collection component captures environmental vibration energy generated by the operation of underground mechanical equipment, vehicle passing, etc. and converts it into electrical energy. This composite design ensures that a certain amount of energy can be collected under various working conditions, and the maximum comprehensive output power is designed to be 50 milliwatts.

[0043] The dynamic power management module is responsible for intelligently allocating and utilizing this limited energy. It continuously monitors the communication load level of the system and the priority of the currently executed sensing tasks. Based on this information, it dynamically adjusts the duty cycle of each functional module within the sensing terminal. For example, during periods of low mobile target traffic, it can reduce the scanning frequency of the RFID read-write module or put the ultrasonic ranging module into intermittent operation mode. When the communication load is light, it can reduce the wireless transmission power or prolong the data transmission interval. Through this fine power control strategy, the system ensures that each sensing terminal and base station can work autonomously for at least 30 days without relying on external power supply, greatly reducing the operation and maintenance burden and total cost of ownership of the system.

[0044] It is noted that the overall operation process of the system of the present embodiment follows a closed-loop control logic: when a mobile target carrying a RFID tag, such as a miner or equipment, enters the effective detection area of any sensing terminal, the terminal will be activated immediately. The integrated RFID read-write module attempts to read the tag identity information, the inertial measurement unit starts recording the target's motion acceleration and angular velocity, and the ultrasonic ranging module measures the precise distance to the target at the same time. All these multi-modal data are packaged into data packets. The sensing terminal then finds and establishes the optimal data transmission path through the adaptive communication networking layer. This path may be directly connected to a fixed multi-standard base station, or may be relayed through one or more mobile relay nodes. The data is finally transmitted to the ground intelligent data fusion processing center through the backbone industrial Ethernet.

[0045] Inside the center, the data preprocessing module first performs time alignment and filtering on the data. The purified data is sent to the multi-source data fusion module, which runs an extended Kalman filter algorithm to fuse information from different sensors and different time and space points, and calculates the most accurate three-dimensional position coordinates, motion velocity vector and unique identity of the mobile target at the current time. The target trajectory reconstruction module then processes these discrete, fusion-optimized position points, applies a cubic spline interpolation algorithm to generate a continuous, smooth complete motion trajectory from the target entering the monitoring area. Finally, the comprehensive state data containing the target identity, real-time position, velocity and historical trajectory are pushed to the ground safety production monitoring system of the mine in real time. The monitoring system uses these high-precision, high-refresh-rate dynamic information for real-time safety area intrusion warning, personnel and equipment distribution statistics, production efficiency analysis and intelligent dispatching, thereby comprehensively improving the safety production management level and operational efficiency of the mine.

[0046] In summary, this embodiment of the system constructs a redundant and complementary sensing system by integrating three sensing technologies—RFID, inertial measurement, and ultrasonic ranging—at the sensing terminal layer. RFID provides target identification and coarse-grained positioning, the inertial measurement unit provides the target's own motion vector information, and ultrasonic ranging provides precise relative distance reference. This multi-source sensing approach effectively overcomes the problems of missed readings, misreadings, and positioning deviations caused by single technologies in complex downhole environments due to electromagnetic multipath attenuation, signal blockage, or equipment errors, significantly improving the completeness and reliability of information acquisition.

[0047] The communication networking layer adopts a collaborative mode of multi-standard base stations and mobile relay nodes to achieve seamless coverage and dynamic topology adaptation in the heterogeneous network environment of mines. The adaptive selection mechanism of the communication protocol ensures that data transmission always prioritizes the channel with the best quality, effectively solving the problems of communication blind spots and link interruptions, while optimizing the overall communication quality and energy consumption of the system.

[0048] The data fusion processing center employs extended Kalman filtering and cubic spline interpolation trajectory reconstruction algorithms to achieve high-precision fusion of multi-source heterogeneous data and generation of continuous target motion trajectories, greatly improving the accuracy of moving target state estimation and trajectory smoothness.

[0049] The distributed energy management unit significantly extends the working life of terminal equipment and reduces the overall deployment and long-term maintenance costs of the system through a composite energy acquisition and dynamic power consumption control strategy, providing solid and sustainable technical support for the intelligent and unmanned construction of mines.

[0050] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered illustrative and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the scope of the invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0051] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A non-contact information acquisition system for moving targets in a mine, characterized in that, include: The system comprises a sensing terminal layer, a communication network layer, and a data fusion processing center. The sensing terminal layer includes several sensing terminals deployed at key nodes in the underground roadways. Each sensing terminal integrates a radio frequency identification (RFID) read / write module, an inertial measurement unit (IMU), and an ultrasonic ranging module. The RFID read / write module operates in the ultra-high frequency band, employing a hybrid time-division multiple access (TDMA) and space-division multiple access (SDMA) mechanism. Its transmit power ranges from 0.1 milliwatts to 100 milliwatts, adaptively adjusted in 0.5-milliwatt increments. The communication network layer consists of multi-standard communication base stations and mobile relay nodes deployed within the roadways, used for data fusion between the sensing terminals and the network. A dynamically reconfigurable data transmission channel is established between processing centers; the multi-standard communication base station has a built-in adaptive communication protocol selection module, which monitors the signal strength and bit error rate of each communication channel in real time and dynamically selects the optimal communication standard based on a preset channel quality evaluation function; the data fusion processing center is deployed in a ground monitoring room and is used to perform fusion calculation and target state estimation on the collected raw data; the data fusion processing center includes at least a target trajectory reconstruction module, which generates a continuous and smooth motion trajectory based on the historical position sequence of the moving target using a cubic spline interpolation algorithm.

2. The system according to claim 1, characterized in that, The mobile relay node is mounted on an underground mining truck or inspection robot platform; the mobile relay node is equipped with a dual-mode communication interface, and its working modes include active detection mode and on-demand relay mode; in active detection mode, the mobile relay node periodically scans the surrounding communication blind spots and reports topology information; in on-demand relay mode, the mobile relay node responds to the instructions of the base station and temporarily acts as a relay to extend network coverage.

3. The system according to claim 1, characterized in that, The data fusion processing center also includes a data fusion module equipped with an extended Kalman filter algorithm. The extended Kalman filter algorithm is implemented as follows: the position and velocity of the moving target are used as the state vector, and the radio frequency identification ranging value, the inertial measurement unit output and the ultrasonic ranging value are used as the observation vector; through two recursive steps of state prediction and observation update, the optimal estimate of the state vector is calculated.

4. The system according to claim 1, characterized in that, The system also includes a distributed energy management unit, which consists of an energy acquisition module and a dynamic power consumption management module built into the sensing terminal and the base station. The dynamic power consumption management module dynamically adjusts the duty cycle of each module according to the communication load and the priority of the sensing task, so that the terminal can work continuously for no less than 30 days without external power supply.

5. The system according to claim 1, characterized in that, The communication networking layer adopts a hierarchical distributed topology structure, which includes a backbone network and an access network. The backbone network is composed of wired industrial Ethernet, connecting the core base stations in each region with the ground intelligent data fusion processing center. The access network is composed of wireless multi-standard base stations and mobile relay nodes, forming a wireless mesh network covering the underground mobile area. The network topology is dynamically maintained through a link-state routing protocol, with a routing update cycle of 5 seconds.

6. The system according to claim 1, characterized in that, The deployment density of the sensing terminal layer is adaptively adjusted according to the roadway structure and target traffic flow; at roadway intersections and main transport roadways, the deployment interval of sensing terminals is no more than 10 meters; in general operating areas, the deployment interval is no more than 20 meters; each sensing terminal is registered in the system with a unique identifier, and its geographical coordinates are pre-entered into the terminal database of the intelligent data fusion processing center.

7. The system according to claim 2, characterized in that, The dual-mode communication interface of the mobile relay node seamlessly switches between industrial wireless LAN and narrowband IoT.

8. The system according to claim 4, characterized in that, The strategy for dynamically adjusting the duty cycle of each module includes: reducing the scanning frequency of the RFID read / write module during periods of low mobile target traffic; and reducing the wireless transmission power or extending the data transmission interval when the communication load is light.

9. The system according to claim 5, characterized in that, The link-state routing protocol enables each node in the network to be aware of the overall network connectivity status; the routing information update cycle is set to 5 seconds.

Images