A gobi desert ground re-measuring method based on a machine dog carrying a GPS dynamic mode

Abstract

The application discloses a kind of based on machine dog carries GPS dynamic mode's Gobi beach original ground re-measurement method, including step one, according to re-measurement area topography, select the elevation measurement scheme of adaptation, and carry out including GPS module, laser ranging module, leg posture perception module installation and calibration on machine dog;Step two, GPS reference station is laid out;Step three, machine dog walks along the preset walking route, carries out on-site re-measurement, and the data of corresponding module in walking process is collected in real time according to the selection elevation measurement scheme;When walking to two elevation measurement scheme switching nodes, switching measurement scheme;Step four, according to the data collected, plane coordinate and elevation data are calculated, and complete re-measurement data are formed.The application is through the high topographic adaptability of machine dog and the innovative fusion of GPS dynamic positioning, two-scheme elevation measurement technology, completely solve the safety, efficiency, accuracy pain point of Gobi beach original ground re-measurement.

Description

Technical Field

[0001] This invention relates to the field of engineering surveying technology. More specifically, this invention relates to a method for re-surveying the original ground surface in the Gobi Desert based on a robot dog equipped with GPS dynamic mode. Background Technology

[0002] Ground resurveying is a crucial step in engineering construction. Its core task is to accurately obtain the ground's planar coordinates (X, Y) and elevation (H) data, providing a fundamental basis for engineering design, construction layout, and quantity calculation. The Gobi Desert, as a typical harsh engineering area, with its rugged terrain, numerous gravel gullies, and harsh climate (strong winds, large diurnal temperature variations, and unshaded exposure to the sun), presents many intractable challenges to traditional manual resurveying work. These challenges mainly include the following: The safety risks to personnel are extremely high: there are no fixed roads in the Gobi Desert, and personnel are prone to accidents such as falling, spraining their ankles, and getting lost when conducting surveys on foot; extreme high / low temperatures and strong winds and sandstorms can easily cause health problems such as heatstroke, frostbite, and respiratory damage, and rescue is difficult and safety costs remain high.

[0003] Traditional equipment has poor adaptability: manual measurement relies on equipment such as total stations, levels, and RTK rover stations, which are large, heavy, and difficult to transport; the electronic components of the equipment are susceptible to wind and sand erosion and temperature differences, resulting in a high failure rate and a high risk of measurement interruption.

[0004] Low measurement efficiency: Manual measurement requires setting up stations point by point, calibrating equipment, and manually recording readings. A single point measurement takes 3-5 minutes, and the average daily measurement range is only 1-2km. Due to the influence of personnel physical strength and environment, the measurement cycle is long and it is difficult to meet the progress requirements of large-scale projects.

[0005] Unstable data accuracy: Manual operation is susceptible to reading errors due to wind, sand and light interference; leveling rods and prisms are unstable when set up on gravelly ground, further affecting the accuracy of elevation measurement; satellite signal obstruction in some areas (such as gullies and dense gravel piles) leads to GPS data interruption or deviation.

[0006] Currently, the main alternatives for surveying in harsh areas are drone surveying and wheeled robot surveying, but both have significant limitations and cannot meet the accuracy and scenario requirements for resurveying the original ground in the Gobi Desert. Drone surveying: It is only suitable for large-scale rough surveys. Low-altitude flight is easily affected by the turbulence of the Gobi Desert and has a high risk of collision with gravel and gullies. Elevation data relies on image matching calculations, and the accuracy can only reach ±5cm, which is far below the accuracy requirement of ±2cm for engineering resurveys. Furthermore, it cannot penetrate obstructions to obtain accurate local data.

[0007] Wheeled robot surveying: It is prone to slipping and getting stuck in gravel and gully terrain, resulting in poor mobility; insufficient vehicle stability leads to large fluctuations in GPS positioning and elevation measurement data, making it impossible to achieve continuous and stable remeasurement.

[0008] Existing technologies have not yet formed an integrated solution of "high mobility access + high precision measurement + unmanned operation", especially in the field of elevation measurement, there is a lack of flexible and accurate measurement technology solutions adapted to the complex terrain of the Gobi Desert. Summary of the Invention

[0009] One objective of this invention is to provide a method for resurveying the original ground in the Gobi Desert based on a robot dog equipped with GPS dynamic mode. The aim is to completely solve the pain points of safety, efficiency, and accuracy in resurveying the original ground in the Gobi Desert by innovatively integrating the high terrain adaptability of the robot dog with GPS dynamic positioning and dual-scheme elevation measurement technology, and to fill the technological gap of unmanned and accurate resurveying in harsh areas.

[0010] To address the aforementioned technical problems, this invention provides a method for re-surveying the original ground surface in the Gobi Desert based on a robot dog equipped with GPS dynamic mode, comprising the following steps: Step 1: Based on the terrain of the re-survey area, select a suitable elevation measurement scheme, and install and calibrate various devices on the robot dog, including the GPS module, laser ranging module, and leg posture perception module. Step 2: Deploy GPS base stations and preset the robot dog's walking route, walking speed, and measurement interval. There are two elevation measurement schemes: laser ranging elevation measurement scheme and leg posture mean elevation measurement scheme. Set the switching node between the two elevation measurement schemes according to the terrain of the re-measurement area. Step 3: The robot dog walks along the preset walking route to conduct on-site re-measurement. It collects data from the corresponding modules in real time during the walking process according to the selected elevation measurement scheme. When it reaches the switching node between the two elevation measurement schemes, it switches the measurement scheme and collects data from the corresponding modules in real time according to the measurement scheme. Step 4: Process the collected data to calculate the plane coordinates and elevation data, forming complete resurvey data.

[0011] Preferably, an appropriate elevation measurement scheme is selected based on the terrain of the re-survey area. Specifically, for flat road sections, a laser ranging elevation measurement scheme is used, where flat road sections refer to areas with a surface slope of <5°, a gravel coverage rate of <30%, and no gullies or protruding obstacles or obstacles with a height of <5cm. For complex road sections, a leg posture mean elevation measurement scheme is used, where complex road sections refer to areas with a surface slope of ≥5°, a gravel coverage rate of ≥30%, and gullies or protruding obstacles with a height of ≥5cm.

[0012] Preferably, the laser ranging elevation measurement scheme is as follows: First, the robot dog is equipped with a GPS module and a laser rangefinder module. The GPS module is installed directly above the center of the robot dog's torso, and the laser rangefinder module is installed on the robot dog's abdomen, and is vertically coaxial with the GPS module. Secondly, while the robot dog is walking, the GPS module and the laser ranging module respectively collect the ground height of the robot dog's own phase center in real time. The vertical distance h between the laser ranging module's ranging center and the ground; Finally, through the formula Calculate the elevation H of the ground point, where d is the fixed vertical distance from the laser ranging module's ranging center to the GPS phase center, and Δh is the system error correction value of the laser ranging module.

[0013] Preferably, the static calibration method for obtaining the system error correction value Δh of the laser ranging module is as follows: place the robot dog on a standard measurement platform with a known elevation, align the laser ranging module vertically with the platform ground, select 10 calibration points within different height ranges of 0.2-1.0m, record the measured value of the laser ranging module and the actual standard distance value, calculate the measurement error of each point, and take the average of the errors of the 10 points to obtain the system error correction value Δh of the laser ranging module.

[0014] Preferably, the method for measuring the mean elevation of leg posture is as follows: First, the robot dog is equipped with a GPS module, which is located directly above the robot dog's center of gravity, at the geometric center of the robot dog's back corresponding to its four legs; at the same time, angle sensors are installed at the knee and hip joints of the robot dog's four supporting legs. Secondly, while the robot dog is walking, the GPS module and angle sensor collect the ground height of the robot dog's own phase center in real time. The flexion and extension angles corresponding to the knee and hip joints; Finally, through the formula Calculate the ground elevation H, where Δh' is the torso deformation correction value. To obtain the average height data of the supporting legs, the flexion and extension angles of the knee and hip joints were obtained. Combined with the lengths of the robot dog's thigh and lower leg, the vertical height of a single leg was calculated using trigonometric functions. The average height data of the supporting legs was then obtained by averaging the vertical heights of multiple supporting legs. .

[0015] Preferably, the mean of the support leg height data is calculated. At the same time, it is also necessary to determine whether the support leg is suspended or supported, and the data of the support leg in the suspended state is removed during the calculation; specifically, an acceleration sensor is set on the support leg to help determine whether the support leg is suspended or supported.

[0016] Preferably, the static calibration method for obtaining the torso deformation correction value Δh' is as follows: Place the robot dog on a standard horizontal platform, with all four supporting legs in a natural supporting state. Measure the actual vertical height from the robot dog's torso GPS module installation position to the platform ground using a high-precision level. Simultaneously, collect data through the leg posture sensing module and calculate the measured height of the torso from the ground. Repeat the test 20 times, calculate the deviation between the measured height and the actual height for each test, and take the average of the deviation values ​​to obtain the torso deformation correction value Δh'.

[0017] Preferably, the planar coordinate measurement includes: first, setting up GPS reference stations around the re-measurement area and continuously sending differential signals to the robot dog; second, when the robot dog walks along the preset route, the GPS module collects satellite signals and combines them with the differential signals from the GPS reference stations to calculate the original planar coordinate data of the current position in real time; finally, the continuously collected planar coordinate data is smoothed by using a Kalman filter algorithm to remove outliers.

[0018] Preferably, the robot dog is equipped with a GPS module, a laser ranging module, and a leg posture sensing module, and also has a remote control terminal. The remote control terminal contains a data acquisition and transmission module, robot dog posture control software, and data processing software. The data acquisition and transmission module collects data from the GPS module, laser ranging module, and leg posture sensing module, and transmits it to the data processing software for data processing to obtain remeasurement data. The remote control terminal controls the robot dog's movement through the robot dog posture control software. The remote control terminal also controls the robot dog to walk to the switching point between the two elevation measurement schemes to switch the measurement scheme.

[0019] The present invention has at least the following beneficial effects: 1. This invention deeply integrates a quadruped robot dog with GPS dynamic positioning technology for on-site ground resurveying in the Gobi Desert. Leveraging the robot dog's high terrain adaptability (crossing gullies and climbing scree slopes), it overcomes the limitations of traditional equipment, drones, and wheeled robots, achieving full-terrain coverage resurveying in harsh areas. Key advantages include: eliminating the need for personnel to enter dangerous areas of the Gobi Desert, avoiding safety risks such as falls, getting lost, and injuries from extreme weather, significantly reducing engineering safety management pressure and safety assurance costs, and fundamentally solving the safety pain points of surveying in harsh areas; the robot dog's walking speed is 1-2 m / s, and its daily measurement range can reach 5-8 km, which is 5-8 times that of manual surveying; data acquisition, processing, and output are fully automated, requiring no manual intervention. Manual reading, recording, and calculation shorten the single-area retest cycle by more than 60%, effectively ensuring project progress; no need to invest a large amount of manpower in on-site measurement and safety assurance, reducing personnel salaries, insurance costs, etc.; low maintenance cost of the robot dog equipment, improved measurement efficiency, reduced equipment rental time, and overall operating cost reduction of 40%-50%; the robot dog has an IP67 or higher waterproof and dustproof rating, and can adapt to extreme environments such as Gobi desert sandstorms, high temperatures (-40℃~60℃), and low temperatures; the leg shock absorption structure ensures torso stability during walking, ensuring normal operation of the measurement equipment; the GPS module adopts anti-interference communication technology to avoid the impact of sandstorms on signal transmission, ensuring continuous and uninterrupted data acquisition, and is suitable for complex environments in harsh areas.

[0020] 2. This invention addresses the complex terrain of the Gobi Desert by designing two switchable elevation measurement schemes: laser ranging and leg posture averaging. Combined with a differentiated and precise installation layout of GPS modules, it solves the problem of insufficient accuracy of single measurement methods in different terrains, achieving high-precision measurement in flat terrain and stable accuracy in complex terrain. This provides a new technical approach for elevation re-measurement in harsh areas. The horizontal measurement accuracy is ±2cm, and the elevation measurement accuracy is ±1.5-2cm, both superior to traditional manual measurement accuracy. Through Kalman filtering, data averaging, and system error correction algorithms, the impact of environmental interference is reduced, significantly improving data reliability and meeting the stringent requirements of engineering re-measurement.

[0021] 3. This invention constructs a fully unmanned measurement mode with preset routes, real-time monitoring, automatic data acquisition, and intelligent processing, which completely changes the operation logic of traditional manual measurement and promotes the transformation of engineering measurement towards intelligence and safety.

[0022] Other advantages, objectives and features of the present invention will become apparent in part from the following description, and in part from those skilled in the art through study and practice of the invention. Attached Figure Description

[0023] Figure 1 This is a schematic diagram of the structure of the robot dog of the present invention. Detailed Implementation

[0024] To better understand the purpose, structure, and function of this invention, the invention will be further described in detail below with reference to the accompanying drawings, so that those skilled in the art can implement it based on the description.

[0025] It should be noted that, unless otherwise specified, the experimental methods described in the following embodiments are all conventional methods, and the reagents and materials described are all commercially available unless otherwise specified. In the description of this invention, the terms "lateral", "longitudinal", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", and "outer" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0026] like Figure 1 As shown, this invention discloses a method for re-surveying the original ground in the Gobi Desert based on a robot dog equipped with GPS dynamic mode, comprising the following steps: Step 1: Based on the terrain of the re-survey area, select a suitable elevation measurement scheme, and install and calibrate various devices on the robot dog, including the GPS module, laser ranging module, and leg posture perception module. Step 2: Deploy GPS base stations and preset the robot dog's walking route, walking speed, and measurement interval. There are two elevation measurement schemes: laser ranging elevation measurement scheme and leg posture mean elevation measurement scheme. Set the switching node between the two elevation measurement schemes according to the terrain of the re-measurement area. Step 3: The robot dog walks along the preset walking route to conduct on-site re-measurement. It collects data from the corresponding modules in real time during the walking process according to the selected elevation measurement scheme. When it reaches the switching node between the two elevation measurement schemes, it switches the measurement scheme and collects data from the corresponding modules in real time according to the measurement scheme. Step 4: Process the collected data to calculate the plane coordinates and elevation data, forming complete resurvey data.

[0027] This invention relates to original ground resurveying technology, particularly suitable for areas with complex terrain and harsh environments such as the Gobi Desert and other arid regions. The core concept is to use a quadruped robot dog as a mobile carrier, integrating a GPS dynamic positioning module, an elevation measurement module (laser ranging module + leg posture sensing module), a data acquisition and transmission module, a posture control module, and a remote terminal. Through the robot dog's autonomous / remote walking in the Gobi Desert area, synchronous, continuous, and unmanned resurveying of original ground plane coordinates and elevation is achieved. To address the accuracy requirements of elevation measurement, two flexibly switchable measurement schemes are designed, combined with differentiated installation layouts of the GPS module, ensuring elevation measurement accuracy under different terrain conditions. Simultaneously, automated data processing enables real-time output and accuracy calibration of the resurveyed data. By using a robot dog equipped with GPS dynamic positioning technology, combined with dual-scheme elevation measurement, unmanned and efficient synchronous resurveying of original ground plane and elevation is achieved, which can be widely applied to the preliminary survey and post-survey work of large-scale projects such as highways, railways, and water conservancy.

[0028] The retesting method in this application comprises two parts: the mechanical dog hardware carrier and the control software module, which together constitute the retesting system.

[0029] 1. Hardware carrier.

[0030] The quadruped robot dog body 2 features high terrain adaptability, with a three-degree-of-freedom shock-absorbing structure in its legs, allowing it to traverse obstacles over 20cm in height. It boasts an IP67 or higher protection rating, resisting sandstorms and extreme temperatures in the Gobi Desert. Its load capacity is ≥5kg, enabling it to carry various measurement modules. The three-degree-of-freedom hydraulic shock-absorbing structure consists of a hip joint rotational shock-absorbing joint, a knee joint flexion-extension shock-absorbing joint, and elastic buffer pads at the end of the legs. Built-in hydraulic dampers in the hip and knee joints absorb impact vibrations during walking. The high-elasticity rubber buffer pads at the end of the legs further cushion vibrations upon contact with the ground, ensuring trunk stability during walking.

[0031] GPS Dynamic Positioning Module (RTK-GPS): Supports joint positioning of multiple systems such as Beidou / GPS / GLONASS. The GPS module 1 has a horizontal positioning accuracy of ±1cm+1ppm and an elevation positioning accuracy of ±2cm+1ppm. The data update rate is 10Hz. It is used to obtain the geodetic coordinates (X, Y, H) of the robot dog's real-time location.

[0032] Elevation measurement module: includes laser ranging module 4 and leg posture sensing module (including hip joint angle sensor 6 and knee joint angle sensor 3). The two module units work independently and can be switched as needed.

[0033] The laser ranging module is used to measure distances using lasers. The leg posture sensing module is composed of a high-precision angle sensor, a three-axis accelerometer, and a gyroscope sensor. The posture sensing method involves the angle sensor collecting limb joint angles, the accelerometer collecting limb motion acceleration, and the gyroscope sensor collecting limb rotation angular velocity. This allows the posture stabilization control module to control the robot dog's walking movements. The sensed posture types include the flexion and extension angles of the robot dog's knee / hip joints, the spatial tilt angle when the leg is supported, the horizontal and pitch angles of the torso, and the extension and contraction states of a single leg.

[0034] The two elevation measurement modules can work independently and switch on demand through the mode switching unit of the data acquisition and transmission module. The switching condition is the terrain type of the re-measurement area (the laser ranging module is used for flat / slightly obstructed terrain, and the leg posture sensing module is used for complex / severely obstructed terrain). The switching method is that the remote control terminal sends a command to the mode switching unit. After receiving the command, the unit cuts off the power supply to one module and connects the other module, and at the same time completes the switching of the data acquisition channel.

[0035] Data acquisition and transmission module: integrates a data acquisition card and a 4G / 5G wireless communication unit, used to synchronously acquire data from the GPS module, elevation measurement module, and attitude stabilization control module, and transmit the data to the remote control terminal in real time, while also storing backup data locally.

[0036] Remote control terminal 5: Industrial-grade tablet computer or computer, pre-installed with data processing software and robot dog control software, which can preset walking routes, switch elevation measurement modes, and monitor real-time data.

[0037] 2. Software modules.

[0038] Robot dog posture control software: enables preset walking routes, real-time obstacle avoidance, and posture stabilization control.

[0039] Attitude stabilization control module: Implemented through a multi-sensor closed-loop feedback control algorithm. Gyroscopes and accelerometers collect the pitch angle, roll angle, and acceleration of the torso in real time, feeding the data back to the attitude control software. The software sends adjustment commands to the leg joint drive units based on a preset torso level threshold, changing the leg flexion and extension angles in real time to compensate for the torso's tilt. Simultaneously, combined with the mechanical buffering of the leg's three-degree-of-freedom shock absorption structure, it ensures that the robot dog's torso remains horizontally stable whether walking or stationary. This utilizes existing, mature control technologies.

[0040] Data processing software: integrates Kalman filtering algorithm (smoothing plane coordinate data), mean calculation algorithm (correcting elevation data), and error correction algorithm (calibrating system errors), automatically completing data processing and accuracy analysis.

[0041] The remeasurement method of this invention includes both planar measurement and elevation measurement.

[0042] 1. Plane measurement.

[0043] Planar surveying is based on GPS dynamic positioning (RTK) technology. Its core principle is to calculate the plane coordinates (X, Y) of the original ground in real time using differential data between the GPS module and the base station, providing fundamental data support for engineering surveying. The specific process and principle are as follows: Base station deployment: Deploy GPS base stations in locations with open views and no electromagnetic interference around the retest area. After initialization, continuously send differential signals (frequency 10Hz) to the robot dog.

[0044] Dynamic data acquisition: The robot dog walks along a preset route (speed 1-2 m / s), and the GPS module collects 10-20 sets of satellite signals per second. Combined with the differential signal from the base station, the system calculates the raw plane coordinates of the current position in real time. The specific steps for calculating the raw plane coordinates are as follows: ① The GPS module receives the carrier phase and pseudorange observations from the satellites, and simultaneously receives the same satellite observations and known plane coordinates from the base station; ② Differential processing eliminates common errors such as satellite clock errors, atmospheric ionospheric delay, and tropospheric delay; ③ A carrier phase double-difference calculation model is used to calculate the baseline vector (ΔX, ΔY) between the robot dog and the base station; ④ Combined with the known plane coordinates of the base station... The original planar coordinates of the robot dog's current position are obtained through forward calculation of planar coordinates. .

[0045] Data smoothing: To address the potential for brief satellite signal obstruction in the Gobi Desert, a Kalman filter algorithm is used to smooth the continuously acquired planar coordinate data, eliminating outliers (such as data deviations caused by signal interruptions) to ensure the stability and accuracy of the planar coordinate data. The final planar measurement accuracy reaches ±2cm, meeting the basic requirements for engineering re-surveys. Deviance data judgment criteria: ① Planar distance deviation between a single set of coordinate data and the previous 5 smoothed sets ≥ 5cm; ② Coordinate data acquired when the number of satellites locked by the GPS module is < 5; ③ Coordinate data acquired when the differential signal interruption duration is > 0.5s. The algorithm directly discards these three types of data and supplements the missing data with Kalman filter predictions.

[0046] 2. Elevation measurement.

[0047] To address the measurement needs of different terrains in the Gobi Desert (flat areas, gully areas, areas with dense gravel, etc.), two independent elevation measurement schemes were designed. The scheme requirements were matched by the differentiated installation layout of GPS modules to ensure the accuracy of elevation measurement. The two schemes can be switched in real time through a remote control terminal.

[0048] Option 1: Laser ranging and elevation measurement (suitable for flat / less obstructed terrain) GPS module installation requirements: The GPS dynamic positioning module should be installed directly above the center of the robot dog's torso to ensure the stability of the positioning reference point; the laser ranging module should be installed separately on the robot dog's abdomen and should be coaxial with the GPS module in the vertical direction (i.e., the ranging center of the ranging module and the phase center of the GPS should be on the same vertical line). This will prevent the robot dog's limbs from blocking the laser signal and will also allow for accurate control of the vertical distance between the two modules, providing basic data for elevation correction.

[0049] Measurement principle: The laser ranging module emits a laser signal vertically towards the ground, measuring the vertical distance (h) between its ranging center and the ground in real time; the GPS module simultaneously collects the geodetic height (h) of its own phase center. Both the laser ranging module and the GPS module are connected to the data logger of the data acquisition and transmission module. The data logger stores two sets of core data (range value and GPS geodetic height) in real time and transmits them to the remote control terminal. At the same time, it presets and calls the fixed vertical distance (d, accurately obtained through previous static calibration) from the laser ranging module's ranging center to the GPS phase center in the data processing software. Finally, by subtracting the laser ranging value and the vertical distance between the two, and correcting system errors, the accurate elevation (H) of the ground point is calculated.

[0050] The core calculation formula is: Where: H is the elevation of the ground point; d is the geodetic height of the GPS phase center; d is the fixed vertical distance from the laser ranging module's ranging center to the GPS phase center (obtained during pre-calibration and is a fixed value); h is the vertical distance between the ranging module's ranging center and the ground; Δh is the system error correction value of the laser ranging module (obtained through pre-static calibration and fixed at 0.2cm).

[0051] Specific calculation example: After preliminary static calibration, d=0.8m and Δh=0.2cm were determined; when the robot dog measured at a flat point, the GPS module collected... =1256.320m, h=0.520m collected by the laser ranging module; substituting into the formula, we calculate: H=1256.320-0.8-0.520-0.002=1254.998m, that is, the precise elevation of this ground point is 1254.998m.

[0052] Workflow: During the robot dog's movement, the data recorder simultaneously triggers the laser ranging module and GPS module to work. The laser ranging module collects distance data every 0.5 seconds, and the GPS module simultaneously collects geodetic height data. The two sets of data are transmitted to the data recorder in real time for storage and synchronization. The data processing software calls the synchronized data in the recorder, substitutes it into the core calculation formula to complete the elevation calculation, and takes the average of 5 consecutive sets of elevation data to further reduce environmental interference errors and ensure that the elevation measurement accuracy is stably within ±1.5cm, achieving accurate elevation data acquisition.

[0053] Option 2: Mean elevation measurement of leg posture (suitable for complex / multiple obstructions in terrain) GPS module installation requirements: The GPS dynamic positioning module must be installed on the back of the robot dog, and strictly located at the geometric center of the back corresponding to the four legs (i.e., directly above the center of gravity of the robot dog's torso) to ensure the stability of the GPS positioning point and reduce the interference of limb swaying on the data when the robot dog walks.

[0054] Measurement Principle: Angle sensors installed at the knee and hip joints of the robot dog's four legs collect real-time data on the flexion angle and extension state of each leg (the ratio of the actual extension length to the maximum extension length, the extension angle threshold of the leg joint, and the spatial angle between the leg and the torso, used to determine the leg's support effectiveness). Combined with the robot dog's preset limb size parameters (thigh length, lower leg length), trigonometric functions are used to calculate the vertical distance (h, h, h, h) between the torso (GPS installation location) and the ground when each leg is supporting the weight. Abnormal data from suspended legs are removed (accelerometers are used to assist in determining the support / suspension state). The average height data (h) of the supporting legs is then taken and recorded as... Combined with GPS geodetic height ( After deducting the torso deformation correction value (Δh', obtained from previous calibration, fixed at 0.3cm), the ground point elevation H is calculated.

[0055] Core calculation formula: .

[0056] Workflow: During the robot dog's walking process, the angle sensor collects 50 sets of posture data per second. The data processing software calculates the height of each leg and the average height in real time, and simultaneously calculates the elevation by combining the GPS geodetic height. The influence of single-leg posture fluctuations is offset by averaging. This solution can achieve an elevation measurement accuracy of ±2cm, meeting the needs of re-measurement in complex terrain.

[0057] Specific calculation example: Robot dog's thigh length =0.3m, lower leg length =0.3m; During measurement at a complex location, legs 1, 2, and 4 were in a supported state, while leg 3 was suspended; Angle sensors collected the hip extension angle (120°) and knee flexion angle (90°) of leg 1, the hip extension angle (120°) and knee flexion angle (90°) of leg 2, and the hip extension angle (120°) and knee flexion angle (90°) of leg 4; The vertical height of a single leg was calculated using trigonometric functions: ,but = (0.560 + 0.560 + 0.560) / 3 = 0.560m; GPS data =1289.650m, Δh'=0.3cm=0.003m; Substituting into the formula, we get: H=1289.650-0.560-0.003=1289.087m, which means the precise elevation of this ground point is 1289.087m.

[0058] Standards for defining flat / complex road sections: Flat road sections are areas with flat / minimally obstructed terrain, characterized by a surface slope of <5°, gravel coverage of <30%, and no gullies / protruding obstacles or obstacles with a height of <5cm; Complex road sections are areas with complex / multiple obstructions, characterized by a surface slope of ≥5°, gravel coverage of ≥30%, and the presence of gullies / protruding obstacles with a height of ≥5cm.

[0059] Two elevation measurement schemes can be switched flexibly: normally, the switching is carried out according to the preset switching nodes. If temporary changes in terrain are found during the process, the terrain type is automatically determined by the terrain identification data of the lidar, or the elevation measurement scheme is manually switched after remote manual determination.

[0060] Example: The following examples use the resurvey of the original ground surface of a highway project in the Gobi Desert as a specific application scenario to explain in detail the implementation steps, equipment selection, parameter settings and effect verification of the present invention, ensuring the feasibility and repeatability of the present invention.

[0061] 1. Overview of the implementation scenario.

[0062] The re-survey area is a section of highway construction in the Gobi Desert, with a total length of 5km, including 3km of flat road (with little surface gravel and no obvious gullies) and 2km of complex road (containing many shallow gullies, dense gravel piles, and rugged terrain).

[0063] The re-measurement requirements are: horizontal measurement accuracy ±2cm, vertical measurement accuracy ±2cm, and the re-measurement of the entire road section must be completed within 3 days to avoid the risks of on-site personnel operations.

[0064] 2. Equipment selection and parameter configuration.

[0065] Quadruped robot dog: The quadruped robot dog is hydraulically driven, with a load capacity of 6kg, a walking speed of 1.5m / s, three degrees of freedom in the legs, and can cross obstacles up to 25cm. It has an IP68 protection rating and is suitable for the extreme environment of the Gobi Desert.

[0066] GPS dynamic positioning module: RTK-GPS module, supports BeiDou / GPS dual-mode positioning, with a horizontal accuracy of ±1cm+1ppm, an elevation accuracy of ±2cm+1ppm, a data update rate of 10Hz, and a communication distance of ≥3km.

[0067] Elevation Measurement Module: A. Laser Ranging Module: Measurement range 0.1-10m, accuracy ±0.5mm, ranging frequency 20Hz, resistant to strong light interference, suitable for outdoor measurement. B. Leg Posture Sensing Module: High-precision angle sensor (measurement range 0-360°, accuracy ±0.1°), accelerometer (measurement range ±16g), data update rate 50Hz.

[0068] Data acquisition and transmission module: Industrial-grade data acquisition card, supporting multi-channel synchronous acquisition, 4G / 5G wireless communication, and 128GB storage capacity, ensuring stable data transmission and secure backup.

[0069] Remote control terminal: Industrial-grade tablet PC (Intel Core i7 processor, 8GB RAM, 512GB storage), pre-installed with the data processing software and robot dog posture control software that accompany this invention.

[0070] 3. Detailed ground re-survey method, including the following steps.

[0071] (1) Preliminary preparation: Based on the terrain of the re-survey area, select a suitable elevation measurement scheme, complete the installation and calibration of the GPS module at the corresponding location; set up GPS base stations and complete communication pairing with the robot dog; preset the walking route, measurement interval (0.5-1 second), and error correction parameters through the remote terminal. The following steps are included.

[0072] Step 1: Solution Selection and Equipment Installation and Calibration (1 day before implementation) A. Based on the terrain of the re-survey area, the division of labor is clear: for flat road sections (3km), Scheme 1 (laser ranging) is adopted, and for complex road sections (2km), Scheme 2 (average leg posture) is adopted.

[0073] B. Equipment Installation: Option 1 Installation - Fix the GPS module directly above the center of the robot dog's torso, and fix the laser rangefinder module in the center of the robot dog's abdomen, ensuring that the two are coaxial; Option 2 Installation - Fix the GPS module to the geometric center of the robot dog's back, and install the angle sensor and acceleration sensor at the knee and hip joints of the four supporting legs, connect the wiring and fix them for protection.

[0074] C. Equipment Calibration: Perform static calibration on the laser ranging module to determine the system error correction value Δh = 0.2cm; calibrate the leg posture sensing module and determine the torso deformation correction value Δh' = 0.3cm through a static standing test of the robot dog; debug the communication between the GPS module and the base station to ensure the stability of the differential signal.

[0075] Static calibration method for laser ranging unit: Place the robot dog on a standard measurement platform with a known elevation, and align the laser ranging module vertically with the platform ground. Select 10 calibration points within different height ranges of 0.2-1.0m, record the measured values ​​of the laser ranging module and the actual standard distance values, calculate the measurement error at each point, and take the average of the errors at the 10 points to obtain the system error correction value Δh of the laser ranging module.

[0076] Static calibration method for leg posture sensing unit: Place the robot dog on a standard horizontal platform with all four legs in a natural supporting state. Measure the actual vertical height from the GPS installation position of the robot dog's torso to the ground using a high-precision level. Simultaneously, collect data through the leg posture sensing unit and calculate the measured height of the torso from the ground. Repeat the test 20 times, calculate the deviation between the measured height and the actual height for each test, and take the average of the deviation values ​​to obtain the torso deformation correction value Δh'.

[0077] Step 2: Base Station Deployment and Route Preset (Morning of the Implementation Day) Deploy a GPS base station in an open area 600m from the starting point of the retest area, connect the power supply and antenna, complete the initialization, and set the differential signal output frequency to 10Hz.

[0078] Import the design route map of the re-measurement area through a remote terminal, preset the robot dog's walking speed to 1m / s, the measurement interval to 0.5 seconds, and set the switching node for the two elevation measurement schemes (at the end of the flat section).

[0079] (2) On-site retest: The robot dog walks autonomously / remotely along a preset route, and the posture control module ensures stable walking; the GPS module and elevation measurement module simultaneously collect plane coordinates, geodetic height, and elevation-related data (laser distance / leg posture), and the data acquisition and transmission module transmits the data to the remote terminal in real time and backs it up locally. The process includes the following steps.

[0080] Step 3: On-site retesting (morning-afternoon of the day of implementation) The robot dog starts from the starting point and walks along the preset route. On flat roads, it automatically starts Option 1 (laser ranging). On complex roads, after reaching the switching node, it switches to Option 2 (average leg posture) through the remote control terminal.

[0081] During operation, the remote control terminal monitors the robot dog's walking status and data acquisition status in real time. If it encounters a sudden obstacle (such as a large pile of rubble), it can adjust the walking route remotely. The data acquisition and transmission module simultaneously transmits plane coordinates, ground height, laser distance / leg posture data, etc. to the remote control terminal and backs them up locally.

[0082] After the entire route retest is completed, the robot dog and base station equipment are retrieved, and local backup data is exported to ensure data integrity.

[0083] (3) Data processing: The data processing software of the remote control terminal automatically performs Kalman filtering to smooth the plane coordinate data, calculates the mean and corrects the error of the elevation data, and matches the plane coordinates and elevation data to form complete re-measurement data (X, Y, H). It includes the following steps.

[0084] Step 4: Data Processing and Result Verification (Evening of the day of implementation) Data processing: The planar coordinate data was smoothed by Kalman filtering using terminal software to remove three sets of outliers caused by brief signal obstruction.

[0085] For elevation data of flat road sections, retrieve the GPS geodetic height (which is synchronously stored in the data logger) ), and substitute the laser ranging value (h) into the corrected formula. = - d - h - 0.2cm (d is the vertical distance from the previously calibrated ranging module to the GPS phase center) is calculated, and the average of 5 consecutive sets of data is taken.

[0086] For elevation data of complex road sections, calculate the average height of the support legs. Substitute into the formula = - - Calculate 0.3cm.

[0087] Results verification: 20 randomly selected re-measurement points were compared with the high-precision manual measurement data. The plane coordinate error was ≤1.8cm and the elevation error was ≤1.9cm, which met the re-measurement accuracy requirements. The re-measurement of the entire road section took 2 days, which was 80% shorter than the manual measurement (estimated to take 10 days). There were no personnel working on site, and the safety risks were completely avoided.

[0088] (4) Output of results: Automatically generate a re-survey report, including a data list, accuracy analysis, and a 3D terrain model, supporting data comparison and verification with engineering design drawings. It includes the following steps.

[0089] Step 5: Output Results (Day 3 of Implementation) Generate a complete resurvey report, including all resurvey point data (X, Y, H) for the 5km road section, an accuracy analysis report, and a 3D terrain model; compare the resurvey data with the highway design drawings, mark the deviation areas between the terrain and the design, and provide accurate basis for subsequent construction.

[0090] It is understood that the present invention has been described through some embodiments, and those skilled in the art will recognize that various changes or equivalent substitutions can be made to these features and embodiments without departing from the spirit and scope of the invention. Although embodiments of the present invention have been disclosed above, they are not limited to the applications listed in the specification and embodiments. It can be applied to various fields suitable for the present invention, and other modifications can be easily implemented by those skilled in the art. Therefore, without departing from the general concept defined by the claims and their equivalents, the present invention is not limited to the specific details and examples shown and described herein.

Claims

1. A method for re-surveying the original ground surface in the Gobi Desert based on a robot dog equipped with GPS dynamic mode, characterized in that, Includes the following steps: Step 1: Based on the terrain of the re-survey area, select a suitable elevation measurement scheme, and install and calibrate various devices on the robot dog, including the GPS module, laser ranging module, and leg posture perception module. Step 2: Deploy GPS base stations and preset the robot dog's walking route, walking speed, and measurement interval. There are two elevation measurement schemes: laser ranging elevation measurement scheme and leg posture mean elevation measurement scheme. Set the switching node between the two elevation measurement schemes according to the terrain of the re-measurement area. Step 3: The robot dog walks along the preset walking route and conducts on-site re-measurement. It collects data from the corresponding modules in real time during the walking process according to the selected elevation measurement scheme. When the walking reaches the switching point between the two elevation measurement schemes, the measurement scheme is switched, and the data of the corresponding module during the walking process is collected in real time according to the measurement scheme. Step 4: Process the collected data to calculate the plane coordinates and elevation data, forming complete resurvey data.

2. The method for re-surveying the original ground surface in the Gobi Desert based on a robot dog equipped with GPS dynamic mode as described in claim 1, characterized in that, Based on the topography of the re-survey area, a suitable elevation measurement scheme was selected. Specifically, for flat road sections, a laser ranging elevation measurement scheme was used. Flat road sections refer to areas with a surface slope of <5°, a gravel coverage rate of <30%, and no gullies or protruding obstacles, or obstacles with a height of <5cm. For complex road sections, a leg posture mean elevation measurement scheme was used. Complex road sections refer to areas with a surface slope of ≥5°, a gravel coverage rate of ≥30%, and gullies or protruding obstacles with an obstacle height of ≥5cm.

3. The method for re-surveying the original ground surface in the Gobi Desert based on a robot dog equipped with GPS dynamic mode as described in claim 1, characterized in that, The specific laser ranging elevation measurement scheme is as follows: First, the robot dog is equipped with a GPS module and a laser rangefinder module. The GPS module is installed directly above the center of the robot dog's torso, and the laser rangefinder module is installed on the robot dog's abdomen, and is vertically coaxial with the GPS module. Secondly, while the robot dog is walking, the GPS module and the laser ranging module respectively collect the ground height of the robot dog's own phase center in real time. The vertical distance h between the laser ranging module's ranging center and the ground; Finally, through the formula Calculate the elevation H of the ground point, where d is the fixed vertical distance from the laser ranging module's ranging center to the GPS phase center, and Δh is the system error correction value of the laser ranging module.

4. The method for re-surveying the original ground surface in the Gobi Desert based on a robot dog equipped with GPS dynamic mode as described in claim 3, characterized in that, The static calibration method for obtaining the system error correction value Δh of the laser ranging module is as follows: Place the robot dog on a standard measurement platform with a known elevation, align the laser ranging module vertically with the platform ground, select 10 calibration points within different height ranges of 0.2-1.0m, record the measured value of the laser ranging module and the actual standard distance value, calculate the measurement error of each point, and take the average of the errors of the 10 points to obtain the system error correction value Δh of the laser ranging module.

5. The method for re-surveying the original ground surface in the Gobi Desert based on a robot dog equipped with GPS dynamic mode as described in claim 1, characterized in that, The specific plan for measuring the mean elevation of leg posture is as follows: First, the robot dog is equipped with a GPS module, which is located directly above the robot dog's center of gravity, at the geometric center of the robot dog's back corresponding to its four legs; at the same time, angle sensors are installed at the knee and hip joints of the robot dog's four supporting legs. Secondly, while the robot dog is walking, the GPS module and angle sensor collect the ground height of the robot dog's own phase center in real time. The flexion and extension angles corresponding to the knee and hip joints; Finally, through the formula Calculate the ground elevation H, where Δh' is the torso deformation correction value. To obtain the average height data of the supporting legs, the flexion and extension angles of the knee and hip joints were obtained. Combined with the lengths of the robot dog's thigh and lower leg, the vertical height of a single leg was calculated using trigonometric functions. The average height data of the supporting legs was then obtained by averaging the vertical heights of multiple supporting legs. .

6. The method for re-surveying the original ground surface in the Gobi Desert based on a robot dog equipped with GPS dynamic mode as described in claim 5, characterized in that, Calculate the mean of the support leg height data At the same time, it is also necessary to determine whether the support leg is suspended or supported, and the data of the support leg in the suspended state is removed during the calculation; specifically, an acceleration sensor is set on the support leg to help determine whether the support leg is suspended or supported.

7. The method for re-surveying the original ground surface in the Gobi Desert based on a robot dog equipped with GPS dynamic mode as described in claim 5, characterized in that, The static calibration method for obtaining the torso deformation correction value Δh' is as follows: Place the robot dog on a standard horizontal platform, with all four supporting legs in a natural supporting state. Measure the actual vertical height from the robot dog's torso GPS module installation position to the platform ground using a high-precision level. Simultaneously, collect data through the leg posture sensing module and calculate the measured height of the torso from the ground. Repeat the test 20 times, calculate the deviation between the measured height and the actual height for each test, and take the average of the deviation values ​​to obtain the torso deformation correction value Δh'.

8. The method for re-surveying the original ground surface in the Gobi Desert based on a robot dog equipped with GPS dynamic mode as described in claim 1, characterized in that, The planar coordinate measurement includes: First, setting up GPS reference stations around the re-measurement area and continuously sending differential signals to the robot dog; Second, as the robot dog walks along the preset route, the GPS module collects satellite signals and combines them with the differential signals from the GPS reference stations to calculate the original planar coordinate data of the current position in real time; Finally, the continuously collected planar coordinate data is smoothed using the Kalman filter algorithm to remove outliers.

9. The method for re-surveying the original ground surface in the Gobi Desert based on a robot dog equipped with GPS dynamic mode as described in claim 1, characterized in that, The robot dog is equipped with a GPS module, a laser ranging module, and a leg posture sensing module. It also has a remote control terminal, which contains a data acquisition and transmission module, robot dog posture control software, and data processing software. The data acquisition and transmission module collects data from the GPS module, laser ranging module, and leg posture sensing module and transmits it to the data processing software for data processing to obtain remeasurement data. The remote control terminal controls the robot dog's movement through the robot dog posture control software. The remote control terminal also controls the robot dog to walk to the switching node between two elevation measurement schemes to switch between the two schemes.

Images