GPS-based methods and systems for precise surveying and positioning of highway routes

Abstract

This invention discloses a GPS-based method and system for precise surveying and positioning of highway routes, belonging to the field of highway route surveying and positioning technology. The method includes collecting multi-source sensor data during highway surveying, performing real-time zero-bias calibration on the angular velocity data output by the gyroscope to obtain calibrated angular velocity data, dynamically adjusting visual weights based on the matching residuals of visual feature data and the zero-bias calibration value of the gyroscope, and obtaining a fused positioning result for the highway route through weighted fusion calculation based on absolute position data, the zero-bias calibration value, and the dynamically adjusted visual positioning weights. The fused positioning result includes plane coordinates, elevation values, and azimuth angles. A precise survey report of the highway route is generated based on the fused positioning result. This invention can dynamically calibrate the gyroscope zero bias, dynamically allocate visual weights, achieve adaptive matching of "sensor reliability - weights," and perform closed-loop fusion of multi-source data, improving positioning accuracy and elevation measurement reliability.

Description

Technical Field

[0001] This invention relates to the field of highway route surveying and positioning technology, and in particular to a GPS-based method and system for precise highway route surveying and positioning. Background Technology

[0002] Highway route surveying is the foundation of highway design, construction, and operation and maintenance. The core requirement is to obtain key parameters such as the horizontal position, elevation, and azimuth of the route. Traditional methods rely on manual operation equipment such as total stations and levels, which are inefficient and limited by terrain. With the development of GPS technology, RTK-GPS has become the mainstream with its centimeter-level absolute positioning accuracy. However, in complex scenarios (mountainous areas, urban canyons, tunnels), GPS is susceptible to obstruction and multipath effects, which can lead to positioning interruptions or error jumps.

[0003] Currently, to compensate for the shortcomings of GPS, the industry has introduced inertial navigation-gyroscopes to provide high-frequency attitude data (solving the problem of low GPS sampling rate) and photogrammetry / video measurement to provide environmental feature constraints (assisting in elevation and plane calibration). However, multi-source sensor data have different error characteristics. Specifically, GPS has high absolute accuracy but poor continuity, gyroscopes have good continuity but zero bias drift, and vision is easily affected by lighting / occlusion. Existing fusion methods are difficult to achieve centimeter-level accuracy in all scenarios because they have not established dynamic error calibration and weight allocation mechanisms. Summary of the Invention

[0004] The technical problem to be solved by this invention is that the existing technology has the disadvantages of inertial navigation drift, fixed sensor weights and insufficient fusion accuracy. To address this, we propose a GPS-based method and system for precise surveying and positioning of highway routes.

[0005] In a first aspect, one embodiment of the present invention provides a GPS-based method for precise surveying and positioning of highway routes, comprising the following steps:

[0006] During highway surveying, multi-source sensor data is collected, including absolute position data output by a GPS receiver, angular velocity data output by a gyroscope, and visual feature data output by a photogrammetric device.

[0007] The angular velocity data output by the gyroscope is calibrated in real time with zero bias to obtain calibrated angular velocity data;

[0008] Based on the matching residuals of the visual feature data and the zero-bias calibration value of the gyroscope after calibration, the visual weights are dynamically adjusted to obtain the visual positioning weights.

[0009] Based on the absolute position data, the zero-bias calibration value, and the dynamically adjusted visual positioning weights, a fusion positioning result for the highway route is obtained through weighted fusion calculation. The fusion positioning result includes plane coordinates, elevation values, and azimuth angles.

[0010] A precise survey report of the highway route is generated based on the fusion positioning results.

[0011] Preferably, the step of performing real-time zero-bias calibration on the angular velocity data output by the gyroscope includes:

[0012] The horizontal position change of the absolute position data within a preset sampling time interval is obtained, and the theoretical angular velocity is calculated based on the horizontal position change and the Earth's radius of curvature.

[0013] The angular velocity deviation value is obtained by subtracting the angular velocity data output by the gyroscope from the theoretical angular velocity.

[0014] Based on the product of the angular velocity deviation value and the preset calibration coefficient, the zero bias calibration value of the gyroscope is dynamically corrected, and the real-time angular velocity data is subtracted from the zero bias calibration value to obtain the calibrated angular velocity data.

[0015] Preferably, the method for calculating the theoretical angular velocity is as follows:

[0016] The theoretical angular velocity is obtained by dividing the change in horizontal position by the product of the preset sampling time interval and the radius of curvature of the Earth, wherein the radius of curvature of the Earth is calculated in real time based on the latitude information in the absolute position data output by the GPS receiver.

[0017] Preferably, the step of dynamically adjusting visual weights includes:

[0018] Visual feature points of a highway scene are extracted using photogrammetry equipment, and the visual feature point matching residuals are calculated. The visual feature point matching residuals represent the degree of deviation between the actual coordinates and the predicted coordinates of the feature points.

[0019] Obtain the zero-bias calibration value of the calibrated angular velocity data, and obtain the gyroscope accuracy correction term based on the zero-bias calibration value and the preset maximum zero-bias threshold.

[0020] The visual residual correction term is obtained based on the ratio of the preset visual feature matching residual threshold to the visual feature point matching residual.

[0021] The visual positioning weight is dynamically adjusted and obtained based on the influence of the product of the gyroscope accuracy correction term and the visual residual correction term on the preset visual weight benchmark value. The visual positioning weight increases as the visual feature point matching residual decreases and decreases as the zero bias calibration value increases.

[0022] Preferably, the adjustment range of the visual positioning weight is limited to between 0.1 and 0.8, and the visual positioning weight, GPS weight and gyroscope weight satisfy the normalization condition. The GPS weight is dynamically set according to the GPS signal quality, and the gyroscope weight is input and set by subtracting the sum of the GPS weight and the visual positioning weight from 1.

[0023] Preferably, the weighted fusion calculation step includes:

[0024] Multiply the absolute position data by the GPS weight to obtain the GPS positioning component;

[0025] The calibrated angular velocity data is integrated to obtain the gyroscope integrated position, and then multiplied by the gyroscope weight to obtain the gyroscope positioning component;

[0026] The visual feature data is reconstructed to obtain visual three-dimensional coordinates, which are then multiplied by the dynamically adjusted visual positioning weights to obtain visual positioning components.

[0027] The GPS positioning component, the gyroscope positioning component, and the visual positioning component are summed to obtain the fused positioning result, which is recorded as the final three-dimensional coordinates.

[0028] Preferably, the fused positioning result is further corrected through closed-loop feedback, the specific process of which is as follows:

[0029] The fused positioning result is used as the position reference for the zero-bias calibration of the gyroscope. The change in horizontal position is recalculated, and a calibration-fusion loop optimization is performed.

[0030] Secondly, embodiments of the present invention also provide a GPS-based highway route precision survey and positioning system, including:

[0031] The multi-source data acquisition module is used to acquire absolute position data output by the GPS receiver, angular velocity data output by the gyroscope, and visual feature data output by the photogrammetric equipment.

[0032] The gyroscope calibration module is used to calculate the theoretical angular velocity based on the change in the horizontal position of the GPS, dynamically correct the zero bias calibration value of the gyroscope by comparing the deviation between the real-time angular velocity and the theoretical angular velocity, and output the calibrated angular velocity data.

[0033] The weight dynamic allocation module is used to adjust the visual weights according to the matching residuals of the visual feature data and the zero bias calibration value, and output the visual positioning weights.

[0034] The multi-source fusion positioning module is used to dynamically weight and sum the GPS positioning components, gyroscope positioning components, and visual positioning components, and output the fusion positioning result.

[0035] The results output module is used to generate a highway route survey report based on the fusion positioning results.

[0036] Preferably, the gyroscope calibration module includes:

[0037] The theoretical angular velocity calculation unit is used to calculate the theoretical angular velocity based on the change in horizontal position, the sampling time interval, and the radius of curvature of the Earth.

[0038] The deviation analysis unit is used to compare the real-time angular velocity of the gyroscope with the theoretical angular velocity to obtain the angular velocity deviation value;

[0039] The zero-bias correction unit is used to dynamically adjust the zero-bias calibration value based on the angular velocity deviation value and a preset calibration coefficient, and to correct the calibrated angular velocity data.

[0040] Preferably, the multi-source fusion positioning module further includes a closed-loop feedback unit, used to feed back the fusion positioning result to the gyroscope calibration module to update the calculation benchmark of the GPS horizontal position change.

[0041] The technical effects and advantages of this invention are as follows:

[0042] In this invention, the absolute position data output by the GPS receiver is first used as a reference to deduce the theoretical angular velocity, which is then compared with the actual output of the gyroscope to calculate the zero-bias calibration value in real time. This dynamically corrects the systematic error of the gyroscope, ensuring the long-term reliability of inertial navigation. Secondly, the weight of vision is dynamically adjusted through "visual residual correction term" and "gyroscope accuracy correction term" to achieve adaptive adjustment where "the higher the visual reliability and the lower the gyroscope reliability, the greater the visual weight." This ensures that the fusion result always relies on the most reliable sensor currently available. Finally, using the calibrated angular velocity data and visual positioning weight as input, combined with the absolute position data of GPS, the advantages of "GPS absolute accuracy, gyroscope continuity, and visual environmental constraints" are integrated through weighted fusion, and the controllability of the error is ensured through closed-loop feedback. Attached Figure Description

[0043] Figure 1 This is a flowchart illustrating the steps of the GPS-based method for precise surveying and positioning of highway routes according to the present invention.

[0044] Figure 2 This is a schematic diagram illustrating the acquisition of fused positioning results through precise surveying and positioning in accordance with the present invention. Detailed Implementation

[0045] The present invention will now be described in further detail with reference to the accompanying drawings and preferred embodiments.

[0046] Reference Figure 1 and Figure 2 As shown, the present invention provides a technical solution: a GPS-based method for precise surveying and positioning of highway routes, comprising the following steps:

[0047] Step I: Collect multi-source sensor data during highway surveying. The multi-source sensor data includes:

[0048] Absolute position data output by the GPS receiver;

[0049] Angular velocity data output by the gyroscope;

[0050] Visual feature data output by photogrammetry equipment (such as binocular cameras or structured light cameras).

[0051] The GPS receiver preferably uses RTK-GPS, which has a positioning accuracy of up to the centimeter level. The gyroscope uses MEMS gyroscope or fiber optic gyroscope, and the output angular velocity unit is rad / s. While acquiring images, the photogrammetric equipment obtains visual feature points, feature point pixel coordinate sets and visual feature matching residuals through feature extraction algorithms.

[0052] Preferably, as an embodiment of this application, the sampling frequency can be specifically set as needed in actual application. Specifically, the implementer can adjust the sampling frequency according to the actual survey scenario, and this application does not impose any special restrictions.

[0053] This completes the acquisition of multi-source sensor data for precise surveying of highway routes.

[0054] Step II: Perform real-time zero-bias calibration on the angular velocity data output by the gyroscope to obtain calibrated angular velocity data.

[0055] The steps for real-time zero-bias calibration of the angular velocity data output by the gyroscope include:

[0056] Acquire the change in horizontal position of absolute position data within a preset sampling time interval, and calculate the theoretical angular velocity based on the change in horizontal position and the radius of curvature of the Earth;

[0057] The angular velocity deviation value is obtained by subtracting the angular velocity data output by the gyroscope from the theoretical angular velocity.

[0058] The zero-bias calibration value of the gyroscope is dynamically corrected by multiplying the angular velocity deviation value and the preset calibration coefficient. The real-time angular velocity data is then subtracted from the zero-bias calibration value to obtain the calibrated angular velocity data.

[0059] The step of performing real-time zero-bias calibration on the angular velocity data output by the gyroscope is described in the following formula:

[0060] ;

[0061] In the formula: t p is the zero-bias calibration value, j is the calibration coefficient, which is experimentally calibrated and ranges from 0.8 to 1.2, reflecting the strength of GPS correction to the gyroscope's zero bias, w gyro ΔP is the angular velocity of the gyroscope's original output. GPS Let be the change in horizontal position of GPS over time Δt, where Δt is the sampling time interval and r is the radius of curvature of the Earth at the location of the highway.

[0062] Preferably, as an embodiment of this application, when a highway survey vehicle is in motion (especially in mountainous areas or on curves), the gyroscope needs to provide high-frequency angular velocity data to reflect the vehicle's real-time steering and pitch attitude. However, the gyroscope has "zero drift" (such as zero-point shift caused by temperature changes). After using it alone for 10 seconds, the error may exceed 1°, resulting in position estimation deviation.

[0063] Understandably, although GPS has a low sampling rate, it provides absolute position (latitude and longitude), and can calculate the change in horizontal position ΔP within the sampling time interval Δt. GPS And based on the formula for circular motion: change in angle = arc length / radius, the theory can be derived in reverse. This theoretical value is compared with the actual angular velocity w output by the gyroscope. gyro The difference is the zero bias error, and the calibrated zero bias calibration value t is obtained through the calibration coefficient j. p The calibrated angular velocity data is the calibrated angular velocity = angular velocity w gyro -Zero bias calibration value t p This information is then used for subsequent integration calculations of the gyroscope position, thereby avoiding the divergence of position errors in traditional uncalibrated gyroscopes when GPS loses lock. Furthermore, the absolute position data of GPS is used to anchor the gyroscope to zero bias, transforming inertial navigation from "divergence over time" to "short-term high-precision continuous navigation," thus solving the pain point of "data interruption in signal obstruction scenarios" in highway surveying.

[0064] Thus, the zero-bias calibration value t of the gyroscope was achieved. p Real-time dynamic calibration improves the long-term reliability of inertial navigation.

[0065] Step III: Based on the matching residuals of the visual feature data and the zero-bias calibration value of the gyroscope after calibration, dynamically adjust the visual weights to obtain the visual positioning weights.

[0066] The steps for dynamically adjusting visual weights include:

[0067] Visual feature points of a highway scene are extracted using photogrammetry equipment, and the visual feature point matching residuals are calculated. The visual feature point matching residuals represent the degree of deviation between the actual coordinates and the predicted coordinates of the feature points.

[0068] Obtain the zero-bias calibration value of the calibrated angular velocity data, and obtain the gyroscope accuracy correction term based on the zero-bias calibration value and the preset maximum zero-bias threshold.

[0069] The visual residual correction term is obtained based on the ratio of the preset visual feature matching residual threshold to the visual feature point matching residual.

[0070] Based on the influence of the product of the gyroscope accuracy correction term and the visual residual correction term on the preset visual weight benchmark value, the visual positioning weight is dynamically adjusted and obtained. The visual positioning weight increases as the visual feature point matching residual decreases and decreases as the zero bias calibration value increases.

[0071] Based on the step of dynamically adjusting visual weights, the formula for calculating visual positioning weights is as follows:

[0072] ;

[0073] In the formula: q vis For visual positioning weights, q 0,vis The visual weight baseline value is a preset initial weight, calibrated to 0.5 through experiments. 0,vis A residual threshold is matched to the visual features; this threshold represents the maximum permissible residual and is calibrated using a photogrammetric system. vis The visual feature point matching residual is the deviation between the actual and predicted positions of feature points in photogrammetry, and is calculated using a stereo camera / SfM algorithm. 0,p This is the maximum zero bias threshold for the gyroscope, which is provided in the device datasheet, and gyroscope data exceeding this value is unreliable.

[0074] In urban canyons / tree-lined roads, GPS multipath effects cause location jumps, but visual features are rich (lane lines, streetlight textures are clear), thus requiring increased visual weighting; while inside tunnels: visual features are singular (only walls), but gyroscope data... The high accuracy after calibration necessitates reducing the visual weight and relying on the gyroscope.

[0075] Preferably, as an embodiment of this application, for the visual residual correction term In terms of visual feature point matching residual e vis The smaller the value (the more accurate the feature point matching), the larger this term becomes, and the greater the visual localization weight q. vis The higher the zero bias calibration value t, the better. p The smaller the gyroscope size (the more reliable it is), the better. The closer the gyroscope accuracy correction term is to 1, the higher the visual positioning weight q.vis The baseline value is basically maintained, if the zero offset calibration value t p Approaching the maximum zero bias threshold t 0,p (Severe gyroscope drift) This item approaches 0, and the visual weight is automatically increased (to compensate for the decrease in gyroscope accuracy). In this way, the weight of visual data in the fusion positioning is dynamically adjusted, so that "more reliable sensor data dominates the positioning".

[0076] It is understandable that the visual feature point matching residual e vis The smaller the (feature point matching deviation), the better the visual residual correction term. Large, visual positioning weight q vis The higher the altitude, such as on an open highway, the greater the visual positioning weight q. vis Enlarged, backlit scene, visual positioning weight q vis Then it is suppressed so that the visual localization weight q vis The gyroscope achieves significant improvements in "scenes with clear features" (such as highways) (dominant positioning, filtering out GPS multipath errors) and reduces these errors in "scenes with unclear features" (such as backlighting, occlusion) (avoiding mismatch interference), thus improving scene adaptability and increasing the gyroscope's zero-bias calibration value t. p The smaller the value (the higher the accuracy), the better the gyroscope accuracy correction term. The closer it is to 1, the higher the visual positioning weight q. vis The zero-bias calibration value t is mainly determined by its own reliability (residual). p The larger the value (the lower the accuracy), the more accurate the gyroscope correction term. The smaller the value, the higher the visual positioning weight q. vis The gyroscope weights are automatically increased to compensate for the reduced weights, ensuring inertial navigation remains dominant (leveraging its vibration resistance), thereby utilizing the gyroscope accuracy correction term. The weights are adjusted in a coordinated manner to achieve complementary reliability from multiple sources.

[0077] Thus, the visual positioning weight q is achieved. vis The dynamic adaptive adjustment increases as visual matching accuracy improves and decreases as gyroscope error increases.

[0078] Step IIII: Based on absolute position data, zero-bias calibration value, and dynamically adjusted visual positioning weights, the fusion positioning result of the highway route is obtained through weighted fusion calculation. The fusion positioning result includes plane coordinates, elevation values, and azimuth angles.

[0079] The steps of weighted fusion calculation include:

[0080] Multiply the absolute position data by the GPS weight to obtain the GPS positioning component;

[0081] Integrate the calibrated angular velocity data to obtain the gyroscope integrated position, multiply it by the gyroscope weight to obtain the gyroscope positioning component;

[0082] Visual feature data is reconstructed into three-dimensional visual coordinates, which are then multiplied by dynamically adjusted visual positioning weights to obtain visual positioning components.

[0083] The GPS positioning components, gyroscope positioning components, and visual positioning components are summed to obtain the fused positioning result, which is recorded as the final three-dimensional coordinates.

[0084] The adjustment range of the visual positioning weight is limited to between 0.1 and 0.8, and the visual positioning weight, GPS weight and gyroscope weight meet the normalization condition. The GPS weight is dynamically set according to the GPS signal quality, and the gyroscope weight is input by subtracting the sum of the GPS weight and the visual positioning weight from 1.

[0085] Preferably, as an embodiment of this application, the final three-dimensional coordinate calculation formula is as follows:

[0086] ;

[0087] In the formula: P fusion For the multi-source fusion positioning results - the final three-dimensional coordinates of the highway route, P GPS This is the raw GPS positioning result, P gyro P is the integral position of the gyroscope. vis q represents the visual 3D coordinates. GPS For GPS weights, q gyro q represents the gyroscope weights. vis For visual positioning weights, and q GPS +q gyro +q vis =1.

[0088] Understandably, GPS positioning components Reflecting the "absolute position contribution" provided by GPS, it embodies its global reference role, provides absolute position anchors for the fusion results, and avoids long-term drift. (Gyroscope positioning component) This reflects the "continuous motion contribution" provided by the gyroscope, demonstrating its high-frequency tracking capability, and addresses the low sampling rate deficiency of GPS, as well as the visual positioning component. This reflects the "contribution of environmental feature constraints" provided by vision, embodying its relative distance measurement accuracy, to correct GPS elevation fitting errors (especially in mountainous terrain) and gyroscope heading drift, thereby reducing fusion elevation errors. Furthermore, q GPS +q gyro +q vis =1 ensures that the sum of the weights of multiple data sources is 1 (dimensionless), and that physical quantities are unified (each location item is in meters, and remains in meters after weighting). This avoids dimensional errors or numerical overflow in the positioning results due to weight superposition, ensuring that the final 3D coordinates P of the fused result are consistent.fusion It always has a "meter-level" physical location, which can be directly used for highway route design.

[0089] This achieves high-precision fusion positioning of multi-source sensor data.

[0090] Step III: Generate a precise survey report of the highway route based on the fusion positioning results.

[0091] The fused positioning results are also corrected through closed-loop feedback, the specific process of which is as follows:

[0092] The fused positioning results are used as the position reference for gyroscope zero-bias calibration. The change in horizontal position is recalculated, and a calibration-fusion loop optimization is performed.

[0093] Among them, △P GPS =P GPS(t） -P GPS(t-△t） ;

[0094] In the formula: P GPS(t） P represents the GPS three-dimensional coordinates at the current sampling time. GPS(t-△t） Let P be the GPS 3D coordinates at the previous sampling time, and in the iterative process after discarding the initial survey... GPS(t-△t） =P fusion .

[0095] Preferably, as an embodiment of this application, if the horizontal position change ΔP is directly calculated using raw GPS data... GPS Multipath effects (such as those occurring in urban canyons) can cause position jumps, leading to deviations in the theoretically derived angular velocity. Errors in zero-bias calibration can further affect the gyroscope's integral position P. gyro Error accumulation, and the final three-dimensional coordinates P fusion GPS noise and visual mismatches have been eliminated; horizontal position change ΔP GPS Reflecting the actual motion of the vehicle, the deviation in theoretical angular velocity calculation is reduced, and the accuracy of zero-bias calibration is improved; the final three-dimensional coordinates P fusion The accuracy of the zero bias calibration will also directly affect the zero bias calibration, and the calibrated zero bias calibration value t p Determine the gyroscope's integral position P gyro The accuracy of the gyroscope integral position P gyro As P fusion =q GPS ×P GPS +q gyro ×P gyro +q vis ×P vis The input items feed back into the fusion accuracy, forming a closed-loop control chain of "fusion accuracy → zero bias calibration accuracy → inertial position accuracy → fusion accuracy".

[0096] Understandably, by dynamically calibrating the gyroscope to zero bias to solve "inertial drift", dynamically allocating weights to solve "sensor reliability matching", and using multi-source fusion to achieve "high-precision positioning", the three work together to form a technical system of "error calibration-weight allocation-data fusion", thereby further improving the system accuracy to the centimeter level and ensuring long-term stability, ultimately supporting the "all-scenario, high-efficiency, and high-precision" requirements of highway route surveying.

[0097] This completes the entire process of precise surveying and positioning of highway routes based on GPS.

[0098] Based on the same inventive concept as the above method, embodiments of the present invention also provide a GPS-based highway route precision survey and positioning system, including:

[0099] The multi-source data acquisition module is used to acquire absolute position data output by the GPS receiver, angular velocity data output by the gyroscope, and visual feature data output by the photogrammetric equipment.

[0100] The gyroscope calibration module is used to calculate the theoretical angular velocity based on the change in the horizontal position of the GPS, dynamically correct the zero bias calibration value of the gyroscope by comparing the deviation between the real-time angular velocity and the theoretical angular velocity, and output the calibrated angular velocity data.

[0101] The dynamic weight allocation module is used to adjust the visual weights based on the matching residuals and zero-bias calibration values ​​of the visual feature data, and output the visual localization weights.

[0102] The multi-source fusion positioning module is used to dynamically weight and sum the GPS positioning components, gyroscope positioning components, and visual positioning components, and output the fusion positioning result.

[0103] The results output module is used to generate a highway route survey report based on the fusion positioning results.

[0104] Thus, the multi-source data acquisition module, gyroscope calibration module, weight dynamic allocation module, multi-source fusion positioning module, and result output module of the GPS-based highway route precision survey and positioning system jointly execute and implement any one of the steps of the above-mentioned GPS-based highway route precision survey and positioning method.

[0105] It should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this invention should also be within the scope of protection of this invention.

Claims

1. A GPS-based highway line precision surveying and positioning method, characterized in that, The method comprises the following steps: Collecting multi-source sensor data in the process of highway surveying, the multi-source sensor data comprising absolute position data output by a GPS receiver, angular velocity data output by a gyroscope, and visual feature data output by a photogrammetry device; Performing real-time zero offset calibration on the angular velocity data output by the gyroscope to obtain calibrated angular velocity data; According to the matching residual of the visual feature data and the zero offset calibration value of the calibrated gyroscope, dynamically adjusting the visual weight to obtain a visual positioning weight; Based on the absolute position data, the zero offset calibration value, and the dynamically adjusted visual positioning weight, calculating a fusion positioning result of the highway route through weighted fusion, the fusion positioning result comprising plane coordinates, elevation values, and azimuth angles; Generating a precise survey report of the highway route according to the fusion positioning result; The step of dynamically adjusting the visual weight comprises: Extracting visual feature points of the highway scene through the photogrammetry device, and calculating visual feature point matching residuals of the visual feature points, the visual feature point matching residuals being the deviation degree of the actual coordinates of the feature points from the predicted coordinates; Obtaining the zero offset calibration value of the calibrated angular velocity data, and based on the zero offset calibration value and a preset maximum zero offset threshold, obtaining a gyroscope accuracy correction term; According to the preset visual feature matching residual threshold and the ratio of the visual feature point matching residuals, obtaining a visual residual correction term; According to the product influence of the gyroscope accuracy correction term and the visual residual correction term on the preset visual weight reference value, dynamically adjusting and obtaining the visual positioning weight, the visual positioning weight increasing as the visual feature point matching residuals decrease and decreasing as the zero offset calibration value increases.

2. The GPS based highway alignment surveying and positioning method as claimed in claim 1, wherein, The step of performing real-time zero offset calibration on the angular velocity data output by the gyroscope comprises: Obtaining the horizontal position change amount of the absolute position data within a preset sampling time interval, and calculating a theoretical angular velocity according to the horizontal position change amount and the earth curvature radius; Subtracting the theoretical angular velocity from the angular velocity data output by the gyroscope to obtain an angular velocity deviation value; Based on the product of the angular velocity deviation value and a preset calibration coefficient, dynamically correcting the zero offset calibration value of the gyroscope, subtracting the zero offset calibration value from the real-time angular velocity data to obtain the calibrated angular velocity data.

3. The GPS based highway alignment surveying and positioning method as claimed in claim 2, wherein, The calculation method of the theoretical angular velocity is: Dividing the horizontal position change amount by the product of the preset sampling time interval and the earth curvature radius to obtain the theoretical angular velocity, wherein the earth curvature radius is calculated in real time according to the latitude information in the absolute position data output by the GPS receiver.

4. The GPS based highway alignment surveying and positioning method as claimed in claim 1, wherein, The adjustment range of the visual positioning weight is limited to between 0.1 and 0.8, and the visual positioning weight, GPS weight, and gyroscope weight satisfy the normalization condition, wherein the GPS weight is dynamically set according to the GPS signal quality, and the gyroscope weight is input set by subtracting the sum of the GPS weight and the visual positioning weight from 1.

5. The GPS based highway alignment surveying and positioning method as claimed in claim 4, wherein, The step of weighted fusion calculation comprises: Multiplying the absolute position data by the GPS weight to obtain a GPS positioning component; integrating the calibrated angular velocity data to obtain a gyroscopic integral position, multiplying the gyroscopic integral position by the gyroscopic weight to obtain a gyroscopic positioning component; reconstructing the visual feature data to obtain visual three-dimensional coordinates, multiplying the visual three-dimensional coordinates by the dynamically adjusted visual positioning weight to obtain a visual positioning component; summing the GPS positioning component, the gyroscopic positioning component and the visual positioning component to obtain the fusion positioning result, and recording the fusion positioning result as a final three-dimensional coordinate.

6. The GPS based highway alignment surveying and positioning method as claimed in claim 2, wherein, The fusion positioning result is also corrected through closed-loop feedback, and the specific process is: taking the fusion positioning result as a position reference for gyroscopic zero offset calibration, recalculating the horizontal position change amount, and performing cyclic optimization of calibration and fusion.

7. A GPS-based highway route precise surveying and positioning system for performing the GPS-based highway route precise surveying and positioning method according to any one of claims 1 to 6, characterized by, It includes: a multi-source data acquisition module for acquiring absolute position data output by a GPS receiver, angular velocity data output by a gyroscope, and visual feature data output by a photogrammetric device; a gyro calibration module for calculating a theoretical angular velocity according to a horizontal position change amount of the GPS, dynamically correcting a zero offset calibration value of the gyroscope by comparing a deviation between a real-time angular velocity and the theoretical angular velocity, and outputting calibrated angular velocity data; a weight dynamic allocation module for adjusting a visual weight according to a matching residual error of the visual feature data and the zero offset calibration value, and outputting a visual positioning weight; a multi-source fusion positioning module for weighting and summing a GPS positioning component, a gyroscopic positioning component and a visual positioning component according to dynamic weights, and outputting a fusion positioning result; a result output module for generating a highway route survey report according to the fusion positioning result; The use steps of the weight dynamic allocation module include: extracting visual feature points of a highway scene through a photogrammetric device, and calculating a visual feature point matching residual error of the visual feature points, the visual feature point matching residual error being a deviation degree between actual coordinates and predicted coordinates of the feature points; obtaining the zero offset calibration value of the calibrated angular velocity data, and obtaining a gyroscopic precision correction term based on the zero offset calibration value and a preset maximum zero offset threshold value; obtaining a visual residual error correction term according to a preset visual feature matching residual error threshold value and a ratio of the visual feature point matching residual error; dynamically adjusting and obtaining the visual positioning weight according to a product influence of the gyroscopic precision correction term and the visual residual error correction term on a preset visual weight reference value, the visual positioning weight increasing as the visual feature point matching residual error decreases and decreasing as the zero offset calibration value increases.

8. The GPS based highway alignment survey and positioning system as claimed in claim 7 wherein, The gyro calibration module includes: a theoretical angular velocity calculation unit for calculating a theoretical angular velocity according to a horizontal position change amount, a sampling time interval and a radius of the earth curvature; a deviation analysis unit for comparing a real-time angular velocity of the gyroscope with the theoretical angular velocity to obtain an angular velocity deviation value; a zero offset correction unit for dynamically adjusting the zero offset calibration value based on the angular velocity deviation value and a preset calibration coefficient, and correcting the calibrated angular velocity data.

9. The GPS based highway alignment surveying and positioning system as claimed in claim 7, wherein, The multi-source fusion positioning module further includes a closed-loop feedback unit for feeding back the fusion positioning result to the gyro calibration module to update a calculation reference of the horizontal position change amount of the GPS.

Images