A new type of electronic address point reader / writer terminal device
The new electronic boundary point reader/writer terminal device, which integrates a UWB ranging module and a machine learning module, solves the problem of low automation in boundary point measurement and realizes efficient and accurate boundary point measurement and monitoring of land changes in villages and towns.
Patent Information
- Application Number
- CN202211430407.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-11-08
- Publication Date
- 2026-11-17
- Estimated Expiration
- 2042-11-08
AI Technical Summary
Existing boundary point measurement methods have low automation, low data collection efficiency, and high professional thresholds, making it impossible to monitor rapid changes in village and town land.
The new electronic boundary point reader/writer terminal device integrates UWB ranging module, wireless communication module, machine learning module, etc., to achieve high-precision positioning and automated measurement. It uses AIoT and TinyML technology for lightweight machine learning inference to reduce communication power consumption and improve battery life.
It has enabled the automation and intelligentization of boundary point surveys, lowered the technical threshold, improved measurement efficiency and accuracy, and enhanced the monitoring capabilities of village and town land changes.
Smart Images

Figure CN115660008B_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the technical field of boundary point spatial information survey equipment, specifically relating to a novel electronic boundary point reader / writer terminal device. Background Technology
[0002] Boundary survey is a crucial part of real estate registration survey work. Boundary setting has legal effect, and the accuracy, realism, and authenticity of the boundaries (points and lines) affect the quality of the real estate registration survey. Boundary points: the turning points of land parcel or ownership boundaries, i.e., inflection points. Boundary lines: the ownership boundaries surrounding a land parcel, i.e., the broken lines or curves formed by connecting the boundary points of the land parcel. The boundary survey process is strictly carried out according to regulations. The core content of a land parcel boundary survey is to ascertain the location, boundaries, boundary line lengths, boundary point locations, types of boundary markers, and their relative positions with neighboring land parcels.
[0003] Boundary point measurement involves surveying boundary points identified through boundary investigations to provide location information for rural real estate cadastral mapping. Currently, there are various boundary point measurement methods, the most widely used being fully digital field surveying and GPS-RTK dynamic surveying. These methods require professional surveyors to operate surveying instruments, resulting in low automation, significant separation between field and office work, and inefficient data collection with a high risk of data loss. Data entry requires substantial human and material resources, increasing the overall workload. Furthermore, these methods demand a certain level of technical expertise from surveyors, hindering the monitoring of rapid changes in rural land use.
[0004] This invention was developed to address the problems of low automation, low data acquisition efficiency, high professional requirements, and untimely monitoring of rapid changes in soil blocks in existing boundary point measurement operations. Summary of the Invention
[0005] To address the aforementioned problems, this invention discloses a novel electronic boundary point reader / writer terminal device, used in conjunction with electronic boundary points. The electronic boundary point is a terminal device comprising a UWB ranging module and a wireless communication module, which is placed on the boundary point to be measured by the operator during use. The electronic boundary point reader / writer can communicate with and measure distances to the target electronic boundary point, achieving high-precision positioning of the boundary point to be measured.
[0006] To achieve the above objectives, the technical solution of the present invention is as follows: A novel electronic boundary point reader / writer terminal device includes a reader / writer terminal housing, hardware circuitry, an antenna, and a mounting bracket. The hardware circuitry includes a controller, a memory, a power management module, an RTK module, a UWB ranging module, a first wireless communication module, a second wireless communication module, a machine learning module, and an input / output module. The controller controls the operation of each module. The memory stores data. The power management module converts the input power supply voltage to the rated industrial voltage for other modules. The RTK module connects to GNSS satellites and ground base stations to acquire the terminal's location information. The first wireless communication module transmits interaction information between the electronic boundary point reader / writer and the electronic boundary points. The second wireless communication module transmits communication between the electronic boundary point reader / writer and a cloud server. The machine learning model is trained using AIoT and TinyML technologies. After being lightweighted on a PC, it is deployed in the machine learning module, allowing the ranging parameters from the UWB ranging module to be used as model input for local inference. The input / output module controls the terminal's status and indicates the device's status to operators. The hardware circuitry is installed inside the reader / writer terminal housing.
[0007] In one example, during a land survey, electronic boundary points are placed at multiple locations to be measured. Each electronic boundary point is a terminal device comprising a UWB ranging module and a wireless communication module. During use, the operator places the electronic boundary point at the desired boundary point location. The electronic boundary point interacts wirelessly with an electronic boundary point reader and a cloud server. The electronic boundary point uses its built-in UWB ranging module to perform distance measurement and positioning with the electronic boundary point reader. The cloud server uniquely encodes the reader terminal device and all electronic boundary points within the measurement area and distributes the encoded data to all terminal devices. The electronic boundary points and the electronic boundary point reader configure communication parameters based on the encoded data, enabling one-to-many communication between the electronic boundary point reader and multiple electronic boundary points, thus avoiding crosstalk between devices.
[0008] The machine learning module, based on AIoT and TinyML technologies, enables lightweight MCU-level machine learning inference. By performing inference on ranging data at the edge, it reduces the communication power consumption of the reader, shortens the ranging and positioning task time, and ultimately improves the overall device's battery life and task efficiency. To achieve the aforementioned objectives, this invention employs the following technical solutions: Step 1: Collect and categorize data obtained from the UWB module in different environments. Step 2: Perform statistical feature analysis on the data collected by the UWB module, and calculate the impact coefficient of different features on the environment in the scene environment; Step 3: Use a deep learning framework to train the model with relevant features, and convert the resulting model file into a file format that can be deployed on an MCU; Step 4: Deploy and infer the model according to the MCU development board of different terminal devices; Step 5: Perform inference based on the data input from the UWB module.
[0009] On the other hand, the reader can send commands through communication with the electronic boundary point to wake up and put the electronic boundary point device into sleep mode, thereby reducing the power consumption of the electronic boundary point device.
[0010] Optionally, the power management module can be powered by a built-in lithium battery or an external carrier platform, and users can configure a suitable power supply method according to the actual usage conditions.
[0011] Optionally, the input / output module includes two parts: an input module and an output module. The input module allows the user to modify the status of the reader / writer, including but not limited to buttons and an interface for communication with a host computer or other hardware device that can modify the reader / writer's configuration. The output module includes but is not limited to status indicator lights, an LED display screen, and a communication interface with a device that can display the reader / writer's status.
[0012] The beneficial effects achieved by this invention are: (1) The electronic boundary point reader / writer terminal device proposed in this invention can be installed and fixed on a drone / ground robot. The reader / writer guides the carrier to automatically go to the area to be measured where the electronic boundary point is located. It performs distance measurement and positioning with the target electronic boundary point, automatically calculates and obtains the information of the target boundary point, and uploads it to the cloud server. During use, there is no need for too much manual operation, which reduces the technical threshold of boundary point survey and realizes the automation of boundary point survey.
[0013] (2) The electronic boundary point reader terminal device can be set with a unique number through the cloud server and receive the approximate coordinate data and unique number of the target electronic boundary point from the server. The approximate coordinates of the target electronic boundary point can guide the UAV or ground robot to carry the reader to the target area. The unique number solves the signal collision problem when communicating between the electronic boundary point and the reader.
[0014] (3) After the electronic boundary point reader / writer terminal device arrives at the area to be measured, it successfully communicates with the target electronic boundary point and sends a command to wake up the target electronic boundary point. After the ranging and positioning task is completed, a hibernation command is sent to put the target electronic boundary point into hibernation, thereby reducing power consumption and increasing battery life.
[0015] (4) The electronic boundary point reader / writer terminal device has a built-in machine learning module, which can run machine learning models on the terminal device to infer the ranging data input by the controller and output prediction results. The prediction results have two functions: First, they can be used as an evaluation basis for short-range UAVs or ground robots to find static measurement points; second, they can filter out gross errors in the ranging process and improve the accuracy of the final ranging and positioning.
[0016] (5) Running deep machine learning inference on the MCU can reduce the communication power consumption of the reader, shorten the ranging and positioning task time, and improve the overall battery life and task efficiency of the device. Attached Figure Description
[0017] Figure 1 This is a schematic diagram of the structure of the present invention; Figure 2 A hardware circuit diagram is provided for this invention; Figure 3 This is a hardware block diagram of the present invention.
[0018] Explanation of icon numbers: 1-Reader terminal box; 2-Hardware circuit; 21-Controller; 22-Memory; 23-Power management module; 24-RTK module; 25-UWB ranging module; 26-Wireless communication module one; 27-Machine learning module; 28-Input / output module; 29-PCB board; 210-Wireless communication module two; 3-Antenna group; 4-Fixing bracket. Detailed Implementation
[0019] The present invention will be further illustrated below with reference to the accompanying drawings and specific embodiments. It should be understood that the following specific embodiments are for illustrative purposes only and are not intended to limit the scope of the invention.
[0020] Example 1: A novel electronic boundary point reader / writer terminal device includes a reader / writer housing 1, hardware circuitry 2, antenna assembly 3, and mounting bracket 4. The hardware circuitry 2 includes a controller 21, a memory 22, a power management module 23, an RTK module 24, a UWB ranging module 25, a wireless communication module 1 26, a machine learning module, an input / output module 28, a PCB board 29, and a wireless communication module 210. The controller 21 controls the operation of each module of the reader / writer. The memory 22 stores data. The power management module 23 converts the input power voltage to the rated operating voltage of other modules. The RTK module 24 connects to GNSS satellites and ground base stations to obtain the terminal's location information. The reader / writer sends ranging data to a server. The server processes the data and sends the processing result back to the reader / writer. The controller 21 executes the next task based on the processing result. The input / output module 28 controls the terminal status and indicates the device status to the operator. The hardware circuitry is installed inside the reader / writer terminal housing 1.
[0021] Example 2: The antenna group 3 includes at least four models such as RTK antenna, UWB antenna, LoRa antenna, NB-IoT antenna, and Cat1 antenna. Each model is configured with different driving circuits and operates at different frequencies. The antenna group 3 can be installed on the inner or outer surface of the reader housing according to actual needs. The RTK antenna and UWB antenna should be deployed on the outer side, and a certain distance should be maintained between the two antenna groups to avoid mutual interference. During use, the electronic boundary point reader should ensure that the RTK antenna and UWB antenna are not obstructed by the carrier.
[0022] The machine learning module 27 includes at least one MCU and a non-volatile memory communicatively connected to the MCU. The MCU is connected to the controller via a communication interface. The MCU receives input data from the controller as a machine learning model for inference and outputs the inference results to the controller. The non-volatile memory stores the network structure and weight data of the model. The controller receives instructions from the cloud server and starts the ranging module of the terminal device through a built-in program. The MCU transmits the inference results to the controller 21 via the communication interface for further calculation and processing to obtain the precise coordinate information of the boundary point to be measured. After standardization, the coordinates are transmitted to the cloud server. The fixed bracket 4 fixes the reader terminal device on a specific carrier, which includes, but is not limited to, drones, ground robots, and handheld platforms.
[0023] The wireless communication module 26 is a LoRa communication module, used for transmitting frequent data and command interactions between the reader terminal device and the electronic address point. The wireless communication module 210 is an NB-IoT communication module or a Cat-1 communication module, used for transmitting data interactions between the reader terminal device and the cloud server. Combining different communication methods can further reduce task latency and device power consumption.
[0024] Example 3: The machine learning module 27 is integrated on the PCB board 29. The steps for deploying the machine learning model on the machine learning module 27 are as follows: Step 1: Collect UWB ranging data in different environments; Step 2: Feature extraction and analysis of UWB data: In this implementation example, the controller 21 can acquire all ranging parameter data from the UWB module.
[0025] Step 3: Use a machine learning framework to train the model and convert the model file for the relevant features; In this implementation example, model training and conversion are performed on a PC.
[0026] Step 4: Deploy the model at the edge; The converted model file is written into the machine learning module using software.
[0027] Step 5: The MCU processes the UWB ranging data for inference; In the use of the electronic boundary point reader / writer, the controller 21 receives instructions from the cloud server and starts the terminal device to activate the UWB ranging module 25 to collect data through a built-in program. The MCU transmits the inference results to the controller 21, which further processes the results to obtain accurate information of the boundary point to be measured and transmits it to the cloud server. The power management module 29 connects to the power supply interface of the carrier platform through a standard power supply interface to power the electronic boundary point reader / writer. The fixing bracket 4 fixes the reader / writer terminal device on a specific carrier, which includes, but is not limited to, drones, ground robots, and handheld platforms. In this embodiment, a method for guiding the carrier platform to find suitable static measurement points is also provided. Specifically, The coordinates of the static observation point are obtained by the RTK module 24. If the RTK cannot converge to a fixed solution, the precise coordinates of the static observation point cannot be obtained, and the static observation point is deemed unreasonable.
[0028] The machine learning module 27 can identify whether there is an NLOS environment during wireless ranging between the electronic boundary point reader and the target electronic boundary point. If the NLOS environment affects the positioning accuracy, the static observation point is determined to be unreasonable.
[0029] The electronic boundary point reader connects to drones or ground robots via a standard communication interface, guiding the drones or electronic boundary points to a reasonable static measurement point.
[0030] Additionally, in this embodiment, when the physical communication interfaces of the carrier platform are insufficient or incompatible, the reader can also transmit the evaluation results of the static measurement points to the cloud server. The cloud server then sends instruction data to guide the drone or ground robot to find suitable static measurement points. Furthermore, if the carrier platform is a handheld platform without an external power interface, the power management module can also be powered by a built-in lithium battery, and the reader can directly display the evaluation results of the static measurement points to the operator through the input / output module.
[0031] In summary, this invention provides a novel electronic boundary point reader / writer device that utilizes wireless communication between the terminal device, the cloud server, and the electronic boundary points to automatically advance the measurement task of the boundary points to be measured. This enables intelligent boundary point surveying. The device operates without requiring excessive human intervention or specialized surveying skills, lowering the barrier to boundary point surveying and enhancing the ability to rapidly monitor land parcel changes.
[0032] It should be noted that the above content merely illustrates the technical concept of the present invention and should not be construed as limiting the scope of protection of the present invention. For those skilled in the art, various improvements and modifications can be made without departing from the principle of the present invention, and all such improvements and modifications fall within the scope of protection of the claims of the present invention.
Claims
1. An electronic address point and line reader terminal device, characterized in that The device includes a reader terminal box (1), hardware circuit (2), antenna group (3), and fixed bracket (4); the hardware circuit (2) includes a controller (21), memory (22), power management module (23), RTK module (24), UWB ranging module (25), wireless communication module one (26), machine learning module (27), input / output module (28), PCB board (29), and wireless communication module two (210); the controller controls the working status of each module of the reader terminal device and parses and processes the collected data, the memory (22) stores the working data of the terminal device, the power management module (23) converts the input power working voltage into the rated industrial voltage of other modules, the RTK module (24) connects to GNSS satellites and ground base stations to obtain the real-time location information of the reader terminal device, the wireless communication module one (26) communicates with electronic boundary points, and the wireless communication module two (210) communicates with cloud servers to obtain the data required by the reader terminal device; The machine learning module (27) deploys a machine learning model to perform inference on the collected data; the input / output module (28) modifies the configuration of the reader terminal device and displays the working status of the device to the staff; the hardware circuit is electrically connected to the PCB board by plug-in or patch, and the hardware circuit is installed inside the reader terminal box (1); The machine learning module (27) includes at least one MCU and a non-volatile memory that is communicatively connected to the MCU. The MCU is connected to the controller (21) through a communication interface. The non-volatile memory stores the network structure and weight data of the machine learning model. The MCU receives the data stream from the controller (21) as the input of the machine learning model and performs inference at the edge. The original input data and inference results are transmitted to the cloud server for secondary development via wireless communication. The RTK module (24) obtains the real-time precise location of the electronic boundary point reader terminal device, and the UWB ranging module (25) determines the coordinates of the target electronic boundary point through the ranging positioning method; Its ranging and positioning workflow features the following steps: Step 1: After assembling the electronic boundary point reader terminal equipment, the operator uses the fixed bracket (4) to install and fix the reader terminal equipment on the carrier platform, and turns on the power supply of the wireless communication module one (26) and wireless communication module two (210) of the reader terminal equipment through the input / output module (28); Step 2: The surveyor places the electronic boundary point on the boundary point to be measured, turns on the power supply of the communication module of the electronic boundary point, and sets the electronic boundary point to the standby state. Step 3: The carrier platform carries the reader terminal device and, guided by the electronic boundary point and the cloud server, runs to the vicinity of the target electronic boundary point; Step 4: The controller (21) sends the power supply command data for the UWB ranging module of the electronic boundary point to the electronic boundary point. After receiving the command data, the electronic boundary point performs the corresponding action and returns the response command data to the electronic boundary point reader terminal device. The controller (21) parses the response command data and then determines whether the command has been executed successfully. Step 5: After the controller (21) determines that the command has been executed successfully, it turns on the power supply to the RTK module (24) and UWB ranging module (25) of the reader terminal device; the carrier platform carries the reader terminal device to a suitable static measurement point; if the controller determines that the command execution has failed in step 4, the reader terminal device sends a device fault code to the background. Step 6: The real-time coordinates of the static measurement point in Step 5 are obtained by the RTK module. The reader terminal device is stationary at the static measurement point. After completing the UWB ranging task, the controller (21) obtains all the ranging parameter data in the UWB ranging module. The ranging parameters are used as the input of the machine learning module. The rationality of the static measurement point selection is determined by a comprehensive evaluation of whether the RTK module converges to a fixed solution and the reasoning results of the machine learning module. Step 7: Set a certain number of static measurement points according to the ranging and positioning requirements. The carrier platform is equipped with a reader terminal device and moves to a reasonable static measurement point to obtain the ranging data of the location and the target electronic boundary point. After all static measurement points are collected, the data is input into the controller (21) to calculate the coordinate data of the target electronic boundary point. The MCU of the machine learning module processes the data collected by the UWB ranging module (25), including all ranging parameter data; the output inference results have the following functions, including: the MCU identifies whether there is an NLOS environment when wirelessly ranging between the electronic boundary point reader terminal device and the target electronic boundary point, assesses whether it will affect the positioning accuracy, and the inference results serve as one of the standards for the rationality of the static measurement point of the electronic boundary point reader terminal device; the MCU identifies the error factors when wirelessly ranging between the electronic boundary point reader terminal device and the target electronic boundary point, infers the ranging error calibration factor, and the inference results serve as subsequent ranging and positioning error compensation; the inference results are input to the controller (21) for processing, and transmitted to the vehicle through the communication interface to guide the vehicle to perform the next action; when the electronic boundary point reader terminal device is installed on the handheld platform, it is transmitted to the output device to guide the handheld operator to perform the next action.
2. An electronic site and spot reader terminal device as in claim 1, wherein, The antenna group (3) includes four types of antennas: RTK antenna, UWB antenna, LoRa antenna, and NB-IoT antenna. Each antenna is configured with a different driving circuit and operates at a different frequency. The antenna group (3) is installed at a suitable position on the outer surface of the reader terminal box according to actual performance requirements.
3. An electronic site and sound reader terminal device as claimed in claim 1, characterized in that The wireless communication module one (26) is a LoRa communication module, which uses LoRa communication as the first communication method to process the data and command exchange between the electronic boundary point reader terminal device and the electronic boundary point; the wireless communication module two (210) is an NB-IoT communication module or a Cat-1 communication module, which uses NB-IoT communication as the second communication method to process the data and command exchange between the electronic boundary point reader terminal device and the cloud server.
4. An electronic site and sound reader terminal device as claimed in claim 1, characterized in that The fixed bracket (4) fixes the reader terminal device on a specific carrier, which includes drones, ground robots and handheld platforms; the power management module (23) is set to be powered by a built-in battery or by an external carrier platform.
5. An electronic site and sound reader terminal device as claimed in claim 1, characterized in that: The machine learning module (27) described above has a machine learning model deployed internally that performs inference on the input data at the edge. The model training and deployment steps are as follows: Step 1: Collect and categorize data from the UWB ranging module in different environments; Step 2: Perform statistical feature analysis on the data collected by the UWB ranging module, and analyze the relationship between different features and environmental factors in the scene environment; Step 3: Use a deep learning framework to train the model using the relevant features; Step 4: Convert the obtained model file on the PC into a file format that can be deployed on the MCU; Step 5: Deploy and infer the model according to the MCU development board of different terminal devices; Step 6: Perform inference based on the data input from the UWB ranging module.
Citation Information
Patent Citations
Multi-sensor multi-environment monitoring method based on AIot and TinyML technologies
CN113408744A
RFID electronic boundary point short-distance high-precision distance measurement equipment and method
CN114462436A
System and Methods for Real Time Kinematic Surveying Using GNSS and Ultra Wideband Ranging
US20100066603A1
