A method and equipment for quickly removing snow cover on ice surface with adaptive adjustment of working depth

Abstract

The present application relates to the technical field of regulation and control, and particularly relates to a method and equipment for quickly removing snow on ice surface with adaptive adjustment of working depth. The method comprises the following steps: obtaining ice-snow layer thickness correlation characteristic data by performing snow layer thickness inversion calculation and correlation mapping between ice surface grooves and snow layer thickness based on ice surface three-dimensional topography point cloud data; obtaining ice-snow layer thickness gradient data corresponding to different ice-snow layer groove partitions by performing ice surface groove shape division and snow layer difference analysis based on the ice-snow layer thickness correlation characteristic data; constructing a working depth adjustment model and performing snow removal execution action predetermination to obtain ice-snow layer removal area execution preset data; and starting ice-snow layer quick removal operation and performing ice-snow layer quick removal iteration optimization to output corresponding ice-snow layer quick removal result. The present application can achieve efficient snow removal on ice surface and accurate balance of ice surface protection.

Description

Technical Field

[0001] This invention relates to the field of adjustment and control technology, and in particular to a method and equipment for rapid snow removal from ice surfaces with adaptive adjustment of working depth. Background Technology

[0002] Snow removal is a complex and crucial task within glacial regions, particularly in scientific research, climate monitoring, and glacier protection. Snow cover on glaciers not only affects their natural state but also impacts long-term climate change monitoring, glacier thickness measurement, and related research. Existing removal methods primarily include physical, chemical, and thermal methods, but these methods are often insufficient to efficiently and accurately address the unique environmental conditions and removal requirements of glacial regions.

[0003] Currently, snow removal in glacier areas mostly relies on manual snowplows and mechanical vehicles. While these methods can remove some snow to a certain extent, the traditional methods are inefficient due to the harsh environment of glacier areas, with extremely low temperatures, strong winds, and uneven snow thickness. They also often require a lot of manual labor. In addition, mechanical vehicles are easily damaged in extremely cold environments and are inconvenient to operate in complex terrain, making it difficult to make precise control on the ice surface, thus reducing the efficiency of snow removal. Summary of the Invention

[0004] Therefore, it is necessary for the present invention to provide a method and equipment for rapid snow removal on ice surfaces with adaptive adjustment of working depth, in order to solve at least one of the above-mentioned technical problems.

[0005] To achieve the above objectives, a rapid snow removal method for ice surfaces with adaptive working depth adjustment includes the following steps: Step S1: Obtain the corresponding three-dimensional ice surface topography point cloud data through the target ice surface area, and perform snow cover thickness inversion calculation on the three-dimensional ice surface topography point cloud data to generate ice surface-snow cover thickness distribution data; perform correlation mapping between ice surface grooves and snow cover thickness based on the three-dimensional ice surface topography point cloud data and the ice surface-snow cover thickness distribution data to obtain ice surface-snow cover thickness correlation feature data. Step S2: Divide the target ice surface region into ice surface groove morphology based on ice surface-snow cover thickness correlation feature data to generate different ice surface-snow cover groove morphology partitions; perform snow cover difference analysis on ice surface-snow cover thickness distribution data based on different ice surface-snow cover groove morphology partitions to obtain ice surface-snow cover thickness gradient data corresponding to different ice surface-snow cover groove partitions. Step S3: Construct a working depth adjustment model based on the ice surface-snow cover layer groove morphology partition and the corresponding ice surface-snow cover layer thickness gradient data to output the corresponding depth adjustment reference parameter data, and perform snow removal execution actions on the ice surface-snow cover layer thickness distribution data based on the depth adjustment reference parameter data to obtain the pre-set data for the ice surface-snow cover layer removal area. Step S4: Start the rapid ice-snow layer removal operation according to the preset data of the ice-snow layer removal area, and collect the snow layer removal residual data corresponding to the removal process through real-time feedback sensors; perform iterative optimization of the working depth adjustment model for ice-snow layer removal based on the snow layer removal residual data and combined with the ice-snow layer thickness correlation feature data, so as to output the corresponding rapid ice-snow layer removal results.

[0006] Furthermore, step S1 includes the following steps: Step S11: The target ice surface area is scanned by deploying a lidar scanner to collect the corresponding ice surface elevation data and ice surface outline point cloud data within each grid cell; Step S12: Based on the ice surface elevation data and ice surface contour range point cloud data corresponding to each grid cell, perform three-dimensional topographic coordinate registration on the target ice surface area to obtain three-dimensional topographic point cloud data of the ice surface. Step S13: Perform snow cover thickness inversion calculation on the three-dimensional topographic point cloud data of the ice surface to generate ice surface-snow cover thickness distribution data; Step S14: Perform spatial alignment processing on the three-dimensional ice surface topography point cloud data and the ice surface-snow cover thickness distribution data in the same three-dimensional coordinate system to obtain the corresponding ice surface topography point cloud data and ice surface-snow cover thickness distribution data in the same three-dimensional spatial coordinate system. Step S15: Based on the ice surface topography point cloud data and ice surface-snow cover thickness distribution data in the same three-dimensional spatial coordinate system, perform the correlation mapping between ice surface grooves and snow cover thickness, so as to obtain the correlation mapping between the geometric morphology parameters of the irregular ice surface grooves and the corresponding snow cover thickness distribution through the ice surface topography point cloud data, and obtain the ice surface-snow cover thickness correlation feature data.

[0007] Furthermore, step S13 includes the following steps: Step S131: Perform ice surface snow cover point cloud density statistics on each grid cell in the ice surface three-dimensional topography point cloud data to obtain the corresponding ice surface snow cover point cloud distribution density in each grid cell; Step S132: Based on the distribution density of the snow-covered point cloud on the ice surface in each grid cell, perform spatial clustering of the snow-covered area of ​​the three-dimensional shape point cloud data of the ice surface to obtain the snow-covered spatial region of each three-dimensional shape of the ice surface. Step S133: Extract the snow cover zone boundary and describe the morphological features of each three-dimensional snow cover space region of the ice surface to generate a geometric boundary feature library of the ice surface snow cover region; perform ice surface-snow cover interface reflection detection based on the ice surface-snow cover region geometric boundary feature library to calculate the point cloud reflectivity between the ice surface-snow cover interface through local comparison and identify the corresponding ice surface-snow cover interface candidate points; at the same time, perform three-dimensional spatial interpolation based on the ice surface-snow cover interface candidate points to generate the initial ice surface-snow cover interface surface. Step S134: Perform voxel mesh downsampling on the three-dimensional topographic point cloud data of the ice surface to generate simplified three-dimensional point cloud data of the ice surface; based on the simplified three-dimensional point cloud data of the ice surface, perform ground point and non-ground point segmentation to obtain the basic three-dimensional point cloud data of the ice surface; Step S135: The initial ice-snow-covered interface surface is optimized and fitted using the least squares method to remove the corresponding misjudged points of the ice-snow-covered layer using the Mahalanobis distance discrimination criterion, and the corresponding ice-snow-covered interface model is generated based on the remaining ice-snow-covered interface points; the thickness vertical distance between the ice-snow-covered interface models is calculated based on the three-dimensional basic point cloud data of the ice surface to generate ice-snow-covered layer thickness distribution data.

[0008] Furthermore, step S2 includes the following steps: Step S21: Perform morphological analysis on the ice surface groove contour corresponding to the ice surface-snow cover thickness correlation feature data, so as to identify the opening width, opening depth, tilt angle and cross-sectional shape parameters of the corresponding irregular groove on the ice surface, so as to obtain the basic morphological data of the ice surface groove. Step S22: Based on the basic morphological data of the ice surface grooves, the corresponding ice surface groove areas within the target ice surface area are divided into ice surface groove morphology. The grooves are clustered into different morphology categories according to the opening size and depth ratio of different ice surface groove contours, including groove areas corresponding to inverted triangular cones and inverted quadrangular prisms, to generate different ice surface-snow cover layer groove morphology partitions. Step S23: Based on the different ice surface-snow cover layer groove morphology partitions, the ice surface-snow cover layer thickness distribution data is divided into different grooves to determine the snow layer thickness distribution data in the ice surface-snow cover layer grooves corresponding to different ice surface-snow cover layer groove partitions; Step S24: Obtain the snow layer thickness distribution data outside the grooves in the same area corresponding to different ice surface-snow cover layer groove partitions through the ice surface-snow cover layer thickness distribution data, and calculate the snow cover layer thickness difference between the snow layer thickness distribution data outside the grooves in the same area corresponding to different ice surface-snow cover layer groove partitions and the corresponding snow thickness distribution data inside the ice surface-snow cover layer grooves to obtain the ice surface-snow cover layer thickness gradient data corresponding to different ice surface-snow cover layer groove partitions.

[0009] Furthermore, step S3 includes the following steps: Step S31: Obtain the corresponding irregular groove depth parameter data by partitioning the grooves of different ice surface-snow cover layers; Step S32: Based on the irregular groove depth parameter data corresponding to different ice surface-snow cover groove partitions, perform relative thickening statistics within the grooves on the corresponding ice surface-snow cover thickness gradient data to obtain the relative thickening of the snow cover within the grooves corresponding to different ice surface-snow cover groove partitions. Step S33: Based on the irregular groove depth parameter data corresponding to different ice surface-snow cover groove partitions and combined with the corresponding relative thickening of the snow cover in the groove, construct the corresponding working depth adjustment model to determine the adjustment mathematical relationship between the working clearing depth and the groove depth and the relative thickening of the snow cover. Obtain the corresponding ice surface hardness parameter in the ice surface-snow cover groove partition. At the same time, set the corresponding working depth adjustment rate limit based on the ice surface hardness parameter, and impose adjustment limit constraints on the working depth adjustment model based on the working depth adjustment rate limit to avoid damage to the ice surface in the groove due to excessively rapid working depth adjustment. Calculate and output the corresponding depth adjustment benchmark parameter data. Step S34: Divide the snow removal area based on the depth adjustment reference parameter data of the ice surface-snow cover thickness distribution data, so as to merge the areas with similar depth adjustment reference parameter data into the same removal adjustment unit, so as to generate ice surface-snow cover adaptive removal area data, including the working depth adjustment range corresponding to each removal adjustment unit. Step S35: Based on the depth adjustment reference parameter data, pre-set the snow removal execution action for the ice surface-snow cover layer adaptive removal area data to obtain the pre-set execution data for the ice surface-snow cover layer removal area.

[0010] Furthermore, step S32 includes the following steps: The depth parameter data of irregular grooves corresponding to different ice surface-snow cover layer groove partitions are deconstructed into groove topology to convert the corresponding irregular groove depth parameter data into three-dimensional surface coordinates. A polar coordinate system is established with the lowest point of the groove bottom as the origin. At the same time, the radial distribution of the depth value at different angles is calculated to obtain the radial distribution of the irregular groove depth corresponding to different ice surface-snow cover layer groove partitions. Based on the radial distribution of irregular groove depth corresponding to different ice surface-snow cover groove zones, obtain the edge inflection points and edge turning points of irregular grooves corresponding to different ice surface-snow cover groove zones; Based on the irregular groove edge inflection points and irregular groove edge turning points corresponding to different ice surface-snow cover groove partitions, the corresponding ice surface-snow cover groove partitions are divided into groove sub-regions to obtain the irregular groove snow cover sub-partitions corresponding to different ice surface-snow cover groove partitions. Based on the irregular groove sub-regions within the irregular grooves corresponding to different ice-snow-covered groove partitions, the depth of the irregular groove sub-regions and the increase in the difference in ice-snow-covered thickness are selected from the corresponding irregular groove depth parameter data and ice-snow-covered thickness gradient data. Then, based on the depth of the irregular groove sub-regions and the increase in the difference in ice-snow-covered thickness corresponding to the same sub-region, the relative thickening within the grooves is calculated cumulatively to obtain the relative thickening of the snow-covered layer within the grooves corresponding to different ice-snow-covered groove partitions.

[0011] Furthermore, step S35 includes the following steps: Step S351: By extracting the spatial coordinates and working depth adjustment range of each clearing adjustment unit from the ice surface-snow cover adaptive clearing area data, and combining the groove spatial distribution corresponding to the ice surface-snow cover thickness distribution data, basic data for ice surface-snow cover clearing area path planning is generated. Step S352: Construct a path optimization model based on the basic data of the ice surface-snow cover removal area path planning, with the objective function being to maximize the removal coverage and minimize the path length. Introduce priority removal weight rules corresponding to the ice surface-snow cover grooves, including that the weight of deep grooves is higher than that of shallow grooves. At the same time, generate corresponding path optimization objective parameter data for the path optimization model based on the objective function and the priority removal weight rules. Step S353: The A* algorithm is used in conjunction with the path optimization target parameter data to solve the path optimization model, so as to obtain the optimal clearing trajectory from the starting point to the end point of the adaptive clearing area of ​​ice surface-snow cover, which includes the order and turning angle of passing through each clearing adjustment unit, so as to obtain the initial path planning data. Step S354: Perform a smoothness check on the initial path planning data to calculate the turning radius and curvature change rate corresponding to the optimal clearing trajectory. When the curvature change rate exceeds a preset threshold, insert the corresponding transition line segment to obtain the optimized path planning data. Step S355: Based on the corresponding trajectory nodes in the optimized path planning data and combined with the depth adjustment reference parameter data, determine the working clearing depth, tool angle and travel speed corresponding to each trajectory node, and obtain the preset data for the ice surface-snow layer clearing area.

[0012] Furthermore, step S4 includes the following steps: Step S41: Based on the preset data for the ice-snow layer removal area, start the removal execution mechanism to perform the corresponding rapid ice-snow layer removal operation, and collect the corresponding ice-snow layer removal residue image data during the removal process through real-time feedback sensors; Step S42: Perform snow cover layer residual analysis on the ice surface-snow cover layer removal residual image data to identify the areas of snow accumulation that have not been removed from the ice surface-snow cover layer using image segmentation algorithms, and calculate the corresponding residual area and snow thickness based on the areas of snow accumulation that have not been removed from the ice surface-snow cover layer to obtain snow cover layer removal residual data. Step S43: Based on the snow cover layer removal residual data and combined with the ice surface-snow cover layer thickness correlation feature data, perform iterative optimization of the working depth adjustment model for ice surface-snow cover layer removal to output the corresponding rapid ice surface-snow cover layer removal results.

[0013] Furthermore, step S43 includes the following steps: Based on the snow cover layer removal residual data and combined with the ice surface-snow cover layer thickness correlation feature data, the depth correction calculation is performed on the depth adjustment benchmark parameter data output by the working depth adjustment model to obtain the working depth adjustment correction amount data. The working depth adjustment correction data is input into the working depth adjustment model and the adjustment amount corresponding to the clearing actuator is calculated by the PID control algorithm. Based on the adjustment amount corresponding to the clearing actuator, the corresponding control signal is output to obtain the real-time control signal data for clearing execution. The real-time control signal data for clearing is sent to the corresponding drive module of the clearing actuator to drive the depth adjustment component to perform the corresponding depth adaptive adjustment, thereby obtaining real-time depth adjustment data; based on the real-time depth adjustment data, the precise clearing operation corresponding to the ice surface-snow layer clearing area is completed, and the ice surface-snow layer clearing area is simultaneously scanned a second time by a laser scanning device to obtain the secondary residual thickness of the snow layer and the area of ​​ice surface damage traces corresponding to the ice surface-snow layer clearing area, thereby obtaining the original data of the clearing effect; Obtain the original snow cover thickness distribution and the total area of ​​the ice surface corresponding to the ice-snow cover removal area. Based on the original snow cover thickness distribution and the total area of ​​the ice surface, calculate the removal effect of the original data to determine the snow residue rate and ice surface damage degree corresponding to the ice-snow cover removal area. The snow residue rate is the ratio of the secondary residual thickness of the snow cover to the original thickness distribution of the snow cover, while the ice surface damage degree is the ratio of the area of ​​ice surface damage traces to the total area of ​​the ice surface. This yields the removal effect evaluation data. The clearing effect evaluation data is compared with the preset effect threshold to filter out the sub-areas that did not meet the clearing standards for the ice-snow layer clearing zone. Based on the sub-areas that did not meet the clearing standards, the parameters of the working depth adjustment model are reversed to generate model optimization parameters. Based on the model optimization parameters, the corresponding working depth adjustment model is updated to perform iterative optimization of ice-snow layer clearing to output the corresponding rapid clearing results of ice-snow layer.

[0014] Furthermore, the present invention also provides a rapid ice and snow removal device with adaptive working depth adjustment, including a processor, a memory, and a computer program stored in the memory and executable on the processor, for performing the rapid ice and snow removal method with adaptive working depth adjustment as described above.

[0015] The beneficial effects of this invention are: The proposed method for rapid snow removal on ice surfaces with adaptive working depth adjustment, compared with existing technologies, offers the following advantages: By acquiring three-dimensional topographic point cloud data of the target ice surface area, it provides spatial information on ice surface details for subsequent research. Through snow cover thickness inversion calculations on the three-dimensional topographic data of the ice surface, the relationship between the ice surface and the snow cover layer can be revealed, providing accurate ice-snow cover thickness distribution data for subsequent analysis. The inversion calculation not only provides thickness information at various locations on the ice surface but also reflects the thickness variation trend of the snow cover layer at different locations. In this way, high-precision data mapping between the ice surface and the snow cover layer can be achieved, laying a solid foundation for ice removal operations and precise adjustments in the target area. By combining the three-dimensional topographic point cloud data of the ice surface with the ice-snow cover thickness distribution data, further correlation mapping between ice surface grooves and snow cover thickness can reveal the specific relationship between different morphological features and thickness variations within the area. The effective implementation of this process helps optimize operational methods and achieve more efficient removal results in large-scale ice removal operations. Secondly, by utilizing the correlation feature data of ice surface-snow cover thickness, the ice surface is divided into groove shapes, revealing the differences in ice surface and snow cover thickness within each zone. This allows for targeted application of different treatment methods. Snow cover difference analysis for different groove shape zones not only accurately determines the ice surface coverage in different areas but also helps identify areas with different thickness gradients. In snow removal operations, thickness differences often affect the efficiency of snow cover removal. This not only serves as a basis for formulating removal plans but also effectively avoids low efficiency or uneven removal caused by thickness differences. Then, based on the ice surface-snow cover groove shape zones and corresponding thickness gradient data, a working depth adjustment model is constructed. The core function of this depth adjustment model is to adjust the removal depth according to the characteristics of different groove shapes, thereby maximizing the removal effect. Precise depth adjustment avoids incomplete removal due to insufficient depth or waste of resources or unnecessary damage to the ice surface due to excessive depth. The model also incorporates snow cover gradient data, enabling personalized adjustments based on the snow cover thickness of different areas. Meanwhile, by pre-setting snow removal actions based on the ice surface-snow cover thickness distribution data using depth adjustment benchmark parameter data, specific removal paths, actions, and predetermined targets can be set in advance, thereby achieving optimal operational results during actual operations. This step not only improves the accuracy of the removal operation but also makes the overall operation planning more efficient and operable, avoiding the waste of resources and time caused by blind operations.Finally, based on a pre-set execution plan, the ice-snow layer removal operation is quickly initiated. This process uses real-time feedback sensors to monitor and collect residual data on snow layer removal during the operation, enabling timely understanding of deviations in the removal process. Analysis of this real-time data allows for adjustments and optimization of the removal operation, ensuring that each stage of the removal work achieves its intended goals. In particular, based on the correlation characteristics of ice-snow layer thickness, the real-time feedback data accurately reflects the complex relationship between the ice surface and the snow layer, thus enabling effective iterative optimization of the working depth adjustment model. This continuous optimization process ensures that each round of removal operations continuously improves its effectiveness, avoiding over-removal or incomplete removal. Through dynamic adjustment and iterative optimization, the removal operation can not only quickly respond to thickness changes in different areas but also adapt to changes in the ice surface and snow layer under different environmental conditions. This achieves efficient and precise rapid removal operations, thereby improving the corresponding snow removal efficiency on the ice surface. Attached Figure Description

[0016] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 This is a schematic diagram of the steps of the rapid snow removal method for ice surface with adaptive working depth adjustment according to the present invention. Figure 2 for Figure 1 A detailed flowchart of step S1. Detailed Implementation

[0017] The technical method of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0018] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0019] It should be understood that although the terms "first," "second," etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are used merely to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0020] To achieve the above objectives, please refer to Figures 1 to 2 This invention provides a method for rapid snow removal from ice surfaces with adaptive adjustment of working depth, the method comprising the following steps: Step S1: Obtain the corresponding three-dimensional ice surface topography point cloud data through the target ice surface area, and perform snow cover thickness inversion calculation on the three-dimensional ice surface topography point cloud data to generate ice surface-snow cover thickness distribution data; perform correlation mapping between ice surface grooves and snow cover thickness based on the three-dimensional ice surface topography point cloud data and the ice surface-snow cover thickness distribution data to obtain ice surface-snow cover thickness correlation feature data. Step S2: Divide the target ice surface region into ice surface groove morphology based on ice surface-snow cover thickness correlation feature data to generate different ice surface-snow cover groove morphology partitions; perform snow cover difference analysis on ice surface-snow cover thickness distribution data based on different ice surface-snow cover groove morphology partitions to obtain ice surface-snow cover thickness gradient data corresponding to different ice surface-snow cover groove partitions. Step S3: Construct a working depth adjustment model based on the ice surface-snow cover layer groove morphology partition and the corresponding ice surface-snow cover layer thickness gradient data to output the corresponding depth adjustment reference parameter data, and perform snow removal execution actions on the ice surface-snow cover layer thickness distribution data based on the depth adjustment reference parameter data to obtain the pre-set data for the ice surface-snow cover layer removal area. Step S4: Start the rapid ice-snow layer removal operation according to the preset data of the ice-snow layer removal area, and collect the snow layer removal residual data corresponding to the removal process through real-time feedback sensors; perform iterative optimization of the working depth adjustment model for ice-snow layer removal based on the snow layer removal residual data and combined with the ice-snow layer thickness correlation feature data, so as to output the corresponding rapid ice-snow layer removal results.

[0021] In the embodiments of this invention, please refer to Figure 1 The diagram shown is a flowchart illustrating the steps of the rapid snow removal method for ice surface with adaptive working depth adjustment according to the present invention. In this example, the rapid snow removal method for ice surface with adaptive working depth adjustment includes the following steps: Step S1: Obtain the corresponding three-dimensional ice surface topography point cloud data through the target ice surface area, and perform snow cover thickness inversion calculation on the three-dimensional ice surface topography point cloud data to generate ice surface-snow cover thickness distribution data; perform correlation mapping between ice surface grooves and snow cover thickness based on the three-dimensional ice surface topography point cloud data and the ice surface-snow cover thickness distribution data to obtain ice surface-snow cover thickness correlation feature data. In this embodiment of the invention, a 100m×100m target ice surface area is scanned using a lidar scanner (wavelength 1550nm, scanning frequency 200Hz, ranging accuracy ±2cm) in a grid pattern (1m×1m grid). The elevation data (Z-axis accuracy ±0.5cm) and contour point cloud data (X, Y accuracy ±1cm) of each grid are collected, generating 5 million points of three-dimensional ice surface topography point cloud data. Snow cover thickness inversion calculation is performed on this data: ice surface points with reflection intensity >8000 are distinguished from snow-covered points with reflection intensity <5000. The vertical distance from the snow-covered point to the ice surface point directly below it is calculated (e.g., for a snow-covered point Z=1.5m, the corresponding ice surface point Z=1.2m, thickness 0.3m), generating ice surface-snow cover thickness distribution data (containing thickness values ​​of 10,000 grids, accuracy 0.01m). Based on the above data, an association mapping is performed: 80 ice surface grooves (depth >0.1m, area >0.5m²) are identified. 2 Extract the depth (0.1-0.5m) and area (0.5-10m²) of each groove. 2 The correlation coefficient (0.6-0.9) between the ice surface and the corresponding snow cover thickness (0.15-0.85m) is calculated to obtain the correlation characteristic data between the ice surface and the snow cover thickness. This data includes the groove ID, geometric parameters, snow cover parameters and correlation coefficients. All calculations are based on strict geometric formulas and are without subjective judgment.

[0022] Step S2: Divide the target ice surface region into ice surface groove morphology based on ice surface-snow cover thickness correlation feature data to generate different ice surface-snow cover groove morphology partitions; perform snow cover difference analysis on ice surface-snow cover thickness distribution data based on different ice surface-snow cover groove morphology partitions to obtain ice surface-snow cover thickness gradient data corresponding to different ice surface-snow cover groove partitions. In this embodiment of the invention, based on the correlation feature data of ice surface-snow cover thickness, 80 ice surface grooves are morphologically divided: K-means clustering (features are opening width × depth and depth / width ratio) is used to obtain inverted triangular pyramidal grooves (35, opening width 1.2-3.0m, depth 0.1-0.3m, ratio 0.1-0.2) and inverted quadrangular prism grooves (45, opening width 3.0-5.0m, depth 0.3-0.5m, ratio 0.1-0.15), generating two ice surface-snow cover groove morphological partitions. Snow cover difference analysis was performed on the ice-snow cover thickness distribution data for each zone: In zone A (inverted triangular cone type), the average snow cover thickness within the grooves was 0.35m, and the flat snow layer within 1m of the perimeter was 0.25m, with a thickness difference of 0.1m; in zone B (inverted quadrangular prism type), the average snow cover thickness within the grooves was 0.45m, and the flat snow layer within 1.5m of the perimeter was 0.30m, with a thickness difference of 0.15m. The thickness gradient (difference / distance) was calculated, with a value of 0.1m / m for zone A and 0.1m / m for zone B (0.15m difference over a 1.5m distance). This yielded ice-snow cover thickness gradient data for different zones, including zone ID, average thickness difference, and gradient value. Data was statistically analyzed by groove zone to ensure a fixed area for difference calculation and that the gradient value accurately reflects the rate of thickness change.

[0023] Step S3: Construct a working depth adjustment model based on the ice surface-snow cover layer groove morphology partition and the corresponding ice surface-snow cover layer thickness gradient data to output the corresponding depth adjustment reference parameter data, and perform snow removal execution actions on the ice surface-snow cover layer thickness distribution data based on the depth adjustment reference parameter data to obtain the pre-set data for the ice surface-snow cover layer removal area. In this embodiment of the invention, a working depth adjustment model is constructed based on the morphological partitioning and thickness gradient data of the ice surface-snow cover layer grooves. The model inputs are groove depth, snow cover thickness, and gradient value, and the output is a depth adjustment benchmark parameter (working depth = groove depth + snow cover thickness - flat snow thickness + 0.02m safety margin). In partition A, a groove depth is 0.22m, snow cover is 0.35m, and flat snow thickness is 0.25m, so the benchmark parameter = 0.22 + 0.35 - 0.25 + 0.02 = 0.34m; in partition B, a groove depth is 0.4m, snow cover is 0.45m, and flat snow thickness is 0.3m, so the benchmark parameter = 0.4 + 0.45 - 0.3 + 0.02 = 0.57m. Model constraints are set: adjustment rate ≤ 0.05m / s (partition A), ≤ 0.08m / s (partition B). Based on baseline parameters, clearing units were divided (parameter differences ≤ 0.05m were merged), resulting in 6 units (A1-A3, B1-B3). The working depth adjustment range for each unit was ±0.02m of the baseline parameters. Clearing actions were pre-defined for these units: A1 working depth 0.34m, tool pressure 300N, speed 0.5m / s; B1 working depth 0.57m, pressure 250N, speed 0.3m / s. Pre-set data for the ice-snow layer clearing zone was generated. The action parameters were calculated based on physical formulas, without any subjective settings.

[0024] Step S4: Start the rapid ice-snow layer removal operation according to the preset data of the ice-snow layer removal area, and collect the snow layer removal residual data corresponding to the removal process through real-time feedback sensors; perform iterative optimization of the working depth adjustment model for ice-snow layer removal based on the snow layer removal residual data and combined with the ice-snow layer thickness correlation feature data, so as to output the corresponding rapid ice-snow layer removal results.

[0025] In this embodiment of the invention, the snow removal mechanism (milling cutter speed 1000 r / min, depth adjustment screw pitch 5 mm) is activated based on preset data for the ice-snow layer removal zone, operating in the sequence A1→A2→A3→B1→B2→B3. Real-time feedback sensors (4096×3072 camera, 30fps; 50Hz laser rangefinder, ±0.5mm accuracy) collect data: the camera captures images of unremoved snow (grayscale 180-255), and the laser measures the residual thickness. Residual analysis is performed on the data: the percentage of unremoved snow pixels in A1 is 15%, with a residual area of ​​0.6 m². 2 Average thickness 0.03 μm; B1 pixel ratio 10%, residual area 0.8 μm. 2The thickness is 0.025m. Based on the residual data and associated features (A1 groove depth 0.22m, relative thickness increase 0.023m), the optimization model is as follows: the correction amount is calculated (A1=0.03×(0.22 / 0.22)=0.03m), the baseline parameter is updated (0.37m), and a control signal (12mA) is generated through a PID algorithm (Kp=6, Ki=0.12, Kd=0.6) to drive the mechanism for secondary operation. After iteration, the residual thickness of A1 is 0.01m and B1 is 0.008m, both ≤0.02m threshold. The output results include the final residual and the clearing time is 45 minutes. All optimization steps are based on a fixed algorithm, and the residual directly drives the parameter update.

[0026] Furthermore, as an embodiment of the present invention, reference is made to... Figure 2 As shown, Figure 1 A detailed flowchart of step S1 is shown below. In this embodiment, step S1 includes the following steps: Step S11: The target ice surface area is scanned by deploying a lidar scanner to collect the corresponding ice surface elevation data and ice surface outline point cloud data within each grid cell; In this embodiment of the invention, a laser radar scanner (wavelength 1550nm, scanning frequency 200Hz, ranging range 0.5-200m, ranging accuracy ±2cm) is deployed to perform a grid-based scan of the target ice surface area (100m×100m). The grid unit size is set to 1m×1m, for a total of 10,000 grids. The scanner is mounted on a mobile platform (traveling speed 0.5m / s) and scans line by line along the grid, with a 1m interval between each line. The angular resolution of the emitted laser beam during scanning is 0.1°, and each scan line contains 1000 sampling points. The ice surface elevation data (Z-axis coordinate, accuracy ±0.5cm) corresponding to each grid unit is collected. For example, the average elevation of 500 points in a certain grid is 1.25m. The point cloud data of the ice surface contour range is obtained by scanning the grid edges. The contour of each grid consists of point clouds of 4 sides, with each side containing 100 points. The X and Y coordinates of the points are recorded (planar accuracy ±1cm). Generate raw scan data with timestamps (accurate to milliseconds). Data for each grid is stored separately, including the coordinate set of the mean elevation, maximum elevation, minimum elevation, and contour point cloud. This ensures that the data covers all grids without overlap or omission. The parameters of the LiDAR are fixed during the scanning process and are not dynamically adjusted.

[0027] Step S12: Based on the ice surface elevation data and ice surface contour range point cloud data corresponding to each grid cell, perform three-dimensional topographic coordinate registration on the target ice surface area to obtain three-dimensional topographic point cloud data of the ice surface. In this embodiment of the invention, three-dimensional topographic coordinate registration of the target ice surface region is performed based on the ice surface elevation data and ice surface contour range point cloud data corresponding to each grid cell, using the ICP (Iterative Closest Point) algorithm. Using the lower left grid cell (coordinates 0,0) as the reference coordinate system, the contour point clouds of other grids are matched with the contour point cloud of the reference grid. Rotation matrices (rotation angle errors around the X, Y, and Z axes < 0.01°) and translation vectors (translation errors in the X, Y, and Z directions < 0.5cm) are calculated to ensure that the overlap rate of the contour point clouds of adjacent grids is ≥ 95%. During the registration process, the average distance between point clouds is calculated in each iteration, and iteration stops when the average distance is < 0.3cm (maximum 50 iterations). After registering 10,000 grids, a three-dimensional point cloud data of the ice surface was obtained, containing 100 million points. The accuracy of the three-dimensional coordinates (X: 0-100m, Y: 0-100m, Z: 1-1.5m) of each point reached ±0.5cm. The point cloud was stored in voxel divisions of 5cm×5cm×5cm to ensure that the three-dimensional shape can completely reflect the undulation of the ice surface. There were no obvious breaks at the stitching points of the registered point cloud.

[0028] Step S13: Perform snow cover thickness inversion calculation on the three-dimensional topographic point cloud data of the ice surface to generate ice surface-snow cover thickness distribution data; In this embodiment of the invention, snow cover thickness is calculated by inverting the three-dimensional topographic point cloud data of the ice surface. First, the ice surface and snow cover are distinguished by the reflection intensity of the point cloud (ice surface reflection intensity > 8000, snow cover reflection intensity < 5000), and the reflection intensity is directly output by the lidar (range 0-10000). Points with a reflection intensity < 5000 (snow cover points) are clustered using the Euclidean distance clustering algorithm (distance threshold 0.3m), and points within the same cluster are considered to be the same snow cover area. The vertical distance (Z-direction difference) from each snow cover point to the ice surface point directly below it (the point with a reflection intensity > 8000 and the smallest Z value) is calculated, which is the snow cover thickness at that point. For example, if a snow cover point has Z = 1.5m, the corresponding ice surface point has Z = 1.2m, and the thickness is 0.3m. For each 1m×1m grid, the average snow cover thickness (arithmetic mean of the thickness of all snow cover points within the grid) is calculated, generating ice surface-snow cover thickness distribution data. The thickness values ​​are retained to two decimal places (in meters), and areas without snow cover are marked as 0.00m. The threshold is fixed during the inversion process to ensure the consistency of the thickness calculation.

[0029] Step S14: Perform spatial alignment processing on the three-dimensional ice surface topography point cloud data and the ice surface-snow cover thickness distribution data in the same three-dimensional coordinate system to obtain the corresponding ice surface topography point cloud data and ice surface-snow cover thickness distribution data in the same three-dimensional spatial coordinate system. In this embodiment of the invention, the three-dimensional topographic point cloud data of the ice surface and the ice-snow cover thickness distribution data are spatially aligned in the same three-dimensional coordinate system. The coordinate system adopts the Gauss-Kruger projection (3-degree zone, central meridian 117°), with the origin set at the lower left corner of the target ice surface area (X=0, Y=0, Z=0). The coordinates of the three-dimensional topographic point cloud data of the ice surface have been unified to this coordinate system through registration. The grid center point coordinates of the ice-snow cover thickness distribution data (X=0.5m, Y=0.5m is the first grid center) correspond to the grid coordinates of the point cloud data. During alignment, the coordinate deviation between the grid center of the thickness data and the nearest point in the point cloud data is calculated (both <0.1m). A translation transformation (X direction correction +0.02m, Y direction correction -0.03m) is used to make the grid centers of the two coincide. After alignment, each thickness data grid is matched one-to-one with the corresponding grid of the point cloud data. The thickness value is associated with all point clouds in the corresponding grid, generating ice surface morphology point cloud data (containing 100 million points) and ice surface-snow cover thickness distribution data (10,000 grids) in the same three-dimensional spatial coordinate system. The alignment error is <0.1m, ensuring the spatial consistency of subsequent correlation analysis.

[0030] Step S15: Based on the ice surface topography point cloud data and ice surface-snow cover thickness distribution data in the same three-dimensional spatial coordinate system, perform the correlation mapping between ice surface grooves and snow cover thickness, so as to obtain the correlation mapping between the geometric morphology parameters of the irregular ice surface grooves and the corresponding snow cover thickness distribution through the ice surface topography point cloud data, and obtain the ice surface-snow cover thickness correlation feature data.

[0031] In this embodiment of the invention, a correlation mapping between ice surface depressions and snow cover thickness is performed based on the ice surface topography point cloud data and ice surface-snow cover thickness distribution data corresponding to the same three-dimensional spatial coordinate system. An ice surface depression is defined as: an area where the ice surface elevation is more than 0.1m lower than the average ice surface elevation within a 1m radius, and the area is >0.5m². 2 By calculating the average elevation of the neighborhood (radius 1m) of each point, 80 ice surface indentations were identified. Geometric parameters were extracted for each indentation: depth (maximum elevation difference, e.g., 0.25m) and area (area enclosed by the indentation's outline, e.g., 2.5m²). 2 The parameters are: perimeter (length of the outline, e.g., 6m), and slope (inclination angle of the groove wall, e.g., 30°). These parameters are correlated with the snow cover thickness distribution data (average thickness 0.3m, maximum thickness 0.45m) within the groove area, and Pearson correlation coefficients are calculated (correlation coefficient between depth and thickness is 0.85, and between area and thickness is 0.6). This generates ice surface-snow cover thickness correlation feature data, including the ID, geometric parameters, corresponding snow cover thickness parameters, and correlation coefficients for each groove. The correlation process is based on numerical calculations, without subjective judgment, ensuring that the mapping relationship is quantifiable.

[0032] Furthermore, step S13 includes the following steps: Step S131: Perform ice surface snow cover point cloud density statistics on each grid cell in the ice surface three-dimensional topography point cloud data to obtain the corresponding ice surface snow cover point cloud distribution density in each grid cell; In this embodiment of the invention, the point cloud density of the ice surface covered by snow is statistically analyzed for each grid cell within the three-dimensional topographic point cloud data of the ice surface. The point cloud data contains 10 million points, and the grid cell size is 0.1m × 0.1m (a total of 1 million grids). The point cloud density within each grid is calculated as follows: the number of points divided by the grid volume (0.5m in the height direction). For example, if a grid contains 500 points, its density = 500 ÷ (0.1 × 0.1 × 0.5) = 100,000 points / m. 3 The density threshold is set at 80,000 points / m². 3 (Snow-covered areas are typically above this value), and grids below the threshold are marked as bare ice. During the statistical process, the number of point clouds in each grid is quickly retrieved using a KD-tree (query radius 0.1m), ensuring that the density calculation for a single grid is completed within 1 second. The final result is the distribution density of the snow-covered point cloud within each grid cell, with density values ​​retained as integers, generating a 1 million-dimensional density vector for subsequent snow-covered area clustering. All calculations are strictly based on preset thresholds, without human intervention.

[0033] Step S132: Based on the distribution density of the snow-covered point cloud on the ice surface in each grid cell, perform spatial clustering of the snow-covered area of ​​the three-dimensional shape point cloud data of the ice surface to obtain the snow-covered spatial region of each three-dimensional shape of the ice surface. In this embodiment of the invention, the three-dimensional topographic point cloud data of the ice surface is spatially clustered based on the distribution density of the snow-covered point cloud within each grid cell, using the DBSCAN algorithm (neighborhood radius 0.3m, minimum number of points 50). Points with a density ≥ 80,000 points / m are clustered. 3 The grid is considered as the core point, with a density of ≥60,000 points / m in its neighborhood. 3 The grid is considered as boundary points, below 60,000 points / m 3 The points are considered noise points. During clustering, all core points (approximately 600,000) are first labeled. Then, connected core points are merged into clusters using a breadth-first search, and boundary points are assigned to the nearest cluster. This results in 20 three-dimensional snow-covered spatial regions of the ice surface, each containing at least 1000 grids. Regions are numbered from 1 to 20, and each region is labeled with its spatial extent (minimum bounding cube, accuracy 0.01m) and average density (rounded to two decimal places), ensuring that the snow-covered areas are completely and unambiguously divided, and that the clustering results do not rely on subjective judgment.

[0034] Step S133: Extract the snow cover zone boundary and describe the morphological features of each three-dimensional snow cover space region of the ice surface to generate a geometric boundary feature library of the ice surface snow cover region; perform ice surface-snow cover interface reflection detection based on the ice surface-snow cover region geometric boundary feature library to calculate the point cloud reflectivity between the ice surface-snow cover interface through local comparison and identify the corresponding ice surface-snow cover interface candidate points; at the same time, perform three-dimensional spatial interpolation based on the ice surface-snow cover interface candidate points to generate the initial ice surface-snow cover interface surface. In this embodiment of the invention, the snow cover zone boundaries and morphological features of each three-dimensional snow-covered spatial region of the ice surface are extracted and described. The MarchingCubes algorithm (cube size 0.05m) is used to generate a regional surface mesh. During boundary extraction, the normal vector of each mesh face is calculated, and faces with an angle >60° with the external normal vector are defined as boundaries, resulting in 50,000 boundary triangles. Morphological feature description includes area (calculated as the sum of the areas of all boundary triangles, with an accuracy of 0.001m). 2 ), perimeter (total length of boundary sides, accuracy 0.001m), circularity (4π × area / perimeter) 2 (To three decimal places). A geometric boundary feature library for the ice-snow-covered layer region is generated, containing feature vectors for 20 regions (10 features per vector). Based on the feature library, reflection detection is performed at the ice-snow-covered interface. The reflectivity of each point is calculated (normalized value of laser echo intensity, range 0-1), and a threshold of 0.7 for ice reflectivity is set (snow cover is usually <0.7). Through local comparison (window size 0.2m×0.2m), points with abrupt changes in reflectivity (gradient >0.3 / m) are identified as candidate points for the ice-snow-covered interface (approximately 500,000). Based on the candidate points, three-dimensional spatial interpolation is performed using radial basis function interpolation (kernel function is a quadratic surface, smoothing parameter 0.05) to generate an initial ice-snow-covered interface surface with a surface accuracy of 0.01m, ensuring that the interface surface accurately reflects the geometry of the ice-snow separation interface.

[0035] Step S134: Perform voxel mesh downsampling on the three-dimensional topographic point cloud data of the ice surface to generate simplified three-dimensional point cloud data of the ice surface; based on the simplified three-dimensional point cloud data of the ice surface, perform ground point and non-ground point segmentation to obtain the basic three-dimensional point cloud data of the ice surface; In this embodiment of the invention, voxel mesh downsampling is performed on the three-dimensional topographic point cloud data of the ice surface. The voxel size is set to 0.03m × 0.03m × 0.03m, and points falling into the same voxel are replaced by their centroid points. The number of point clouds is reduced from 10 million to about 1.5 million, generating simplified three-dimensional point cloud data of the ice surface. Based on the simplified point cloud data, ground points and non-ground points are segmented using a progressive morphological filtering algorithm (window radius increases from 0.1m to 2m, step size 0.1m), with a terrain slope threshold of 15°. First, the height difference between each point and the lowest point in its local neighborhood (radius 0.5m) is calculated. Points with a height difference < 0.2m are marked as ground candidate points. Then, small obstacles are removed by morphological opening operation (structuring element radius 0.2m), finally obtaining the basic three-dimensional point cloud data of the ice surface (about 1.2 million points), including the ice surface and a small number of low obstacle points. All point clouds retain XYZ coordinates (accuracy 0.001m) and reflectivity (accuracy 0.01) to ensure that the data quality meets the requirements for subsequent interface model construction. The processing strictly follows the preset parameters without adaptive adjustment.

[0036] Step S135: The initial ice-snow-covered interface surface is optimized and fitted using the least squares method to remove the corresponding misjudged points of the ice-snow-covered layer using the Mahalanobis distance discrimination criterion, and the corresponding ice-snow-covered interface model is generated based on the remaining ice-snow-covered interface points; the thickness vertical distance between the ice-snow-covered interface models is calculated based on the three-dimensional basic point cloud data of the ice surface to generate ice-snow-covered layer thickness distribution data.

[0037] In this embodiment of the invention, the initial ice-snow-covered interface surface is optimized and fitted using the least squares method, and the 500,000 candidate points of the ice-snow-covered interface are projected onto a quadratic polynomial surface (equation z=ax). 2 +by 2The Mahalanobis distance from each point to the surface is calculated using the formula (cxy+dx+ey+f), where the covariance matrix considers the local density of the point cloud. A Mahalanobis distance threshold of 3 (corresponding to a 99.7% confidence interval) is set, and points with a distance greater than 3 (approximately 50,000) are removed, resulting in approximately 450,000 ice-snow-cover interface points. The surface is refitted based on the remaining points, iterated three times until the surface parameter change is <0.001, thus constructing the corresponding ice-snow-cover interface model with a model error <0.02m. Based on the 3D basic point cloud data of the ice surface, the snow cover thickness is inverted by calculating the vertical distance between the ice-snow-cover interface models. For each ice surface point (1.2 million), the intersection point with the interface model is found along the positive Z-axis; the distance between the two points represents the snow cover thickness. During the calculation, a bisection method is used to search for the intersection point (accuracy 0.001m), with a maximum search distance of 1m (points exceeding this distance are marked as invalid). The final data on the thickness distribution of the ice surface and snow cover layer contains 1.2 million thickness values ​​(accuracy 0.01m), with a thickness range of 0.03-0.85m. 98% of the points are valid, and invalid points (mainly obstacle areas) are filled in by nearest neighbor interpolation to ensure that the thickness data covers the entire ice surface area. All calculation steps strictly follow mathematical principles and there are no empirical parameter adjustments.

[0038] Furthermore, step S2 includes the following steps: Step S21: Perform morphological analysis on the ice surface groove contour corresponding to the ice surface-snow cover thickness correlation feature data, so as to identify the opening width, opening depth, tilt angle and cross-sectional shape parameters of the corresponding irregular groove on the ice surface, so as to obtain the basic morphological data of the ice surface groove. In this embodiment of the invention, morphological analysis is performed on the ice surface groove contour corresponding to the ice surface-snow cover thickness correlation feature data. The Canny edge detection algorithm (threshold 100-200) is used to extract the groove contour, with a contour point accuracy of ±0.5cm. The opening width is the straight-line distance between the two edge points on the top of the groove, and the average of three measurement points is taken (e.g., left edge point (2,3), right edge point (5,3), width 3.0m); the opening depth is the vertical distance between the lowest point of the groove and the line connecting the opening, which is obtained by calculating the difference in Z coordinates of the contour points (lowest point Z=1.0m, opening line Z=1.3m, depth 0.3m); the tilt angle is the angle between the groove wall and the horizontal plane, calculated using the normal vector of the contour points (30°); the cross-sectional shape parameters are obtained through contour fitting, the ratio of the base to the height of the inverted triangular cross-section is 2:1, and the side length ratio of the rectangular cross-section is 3:1. The basic morphological data of the ice surface grooves were obtained by analyzing each of the 80 grooves. Each groove includes the opening width (1.2-5.0m), opening depth (0.1-0.5m), tilt angle (20°-60°), and cross-sectional shape parameters (ratios accurate to 0.1). The data is sorted by groove ID to ensure that the morphological parameters can be directly used for subsequent classification.

[0039] Step S22: Based on the basic morphological data of the ice surface grooves, the corresponding ice surface groove areas within the target ice surface area are divided into ice surface groove morphology. The grooves are clustered into different morphology categories according to the opening size and depth ratio of different ice surface groove contours, including groove areas corresponding to inverted triangular cones and inverted quadrangular prisms, to generate different ice surface-snow cover layer groove morphology partitions. In this embodiment of the invention, the ice surface groove regions within the target ice surface area are divided into groove shapes based on the basic morphological data of the ice surface grooves. A K-means clustering algorithm (2 clusters) is used, with the opening size (width × depth) and depth ratio (depth / width) as features. The cluster center for inverted triangular pyramidal grooves is: opening size 2.5m × 0.25m, depth ratio 0.1; the cluster center for inverted quadrangular prism grooves is: opening size 4.0m × 0.4m, depth ratio 0.1. The Euclidean distance between each groove and the two cluster centers is calculated. The 35 grooves closer to the inverted triangular pyramidal center are classified into this class, with an opening width of 1.2-3.0m, a depth of 0.1-0.3m, and an inclination angle of 50°-60°; the remaining 45 grooves are classified into the inverted quadrangular prism class, with an opening width of 3.0-5.0m, a depth of 0.3-0.5m, and an inclination angle of 20°-40°. Different ice surface-snow cover layer groove morphology partitions are generated. Inverted triangular cone partitions are labeled A, and inverted quadrangular prism partitions are labeled B. Each partition contains a list of groove IDs and average morphological parameters. The partition boundary is determined by the minimum bounding rectangle of the groove profile to ensure the consistency of groove morphology within the partition. During the clustering process, the distance calculation uses standardized feature values ​​to avoid the influence of dimensions.

[0040] Step S23: Based on the different ice surface-snow cover layer groove morphology partitions, the ice surface-snow cover layer thickness distribution data is divided into different grooves to determine the snow layer thickness distribution data in the ice surface-snow cover layer grooves corresponding to different ice surface-snow cover layer groove partitions; In this embodiment of the invention, the snow layer thickness distribution data of the ice surface-snow cover layer is divided into different grooves based on different ice surface-snow cover groove morphology. For the 35 grooves in partition A, the snow thickness in each groove is extracted according to a 0.5m × 0.5m subgrid, and the average thickness within the subgrid is calculated (e.g., the average thickness of 5 points in a certain subgrid is 0.35m). The same method is used for the 45 grooves in partition B, with an average subgrid thickness of 0.45m. For each groove, a thickness value is recorded every 0.1m along the depth direction to form a thickness profile (e.g., the thickness of a certain groove in partition A from the opening to the bottom is 0.35m, 0.30m, and 0.25m). Data on snow thickness distribution within ice-snow-covered grooves corresponding to different ice-snow-covered groove partitions were obtained. Partition A data contains 1750 sub-grid thicknesses (accuracy 0.01m) for 35 grooves, and Partition B data contains 2250 sub-grid thicknesses for 45 grooves. The data is stored at the partition and groove levels to ensure that the thickness data format is consistent within the same partition, and can be directly used for comparative analysis.

[0041] Step S24: Obtain the snow layer thickness distribution data outside the grooves in the same area corresponding to different ice surface-snow cover layer groove partitions through the ice surface-snow cover layer thickness distribution data, and calculate the snow cover layer thickness difference between the snow layer thickness distribution data outside the grooves in the same area corresponding to different ice surface-snow cover layer groove partitions and the corresponding snow thickness distribution data inside the ice surface-snow cover layer grooves to obtain the ice surface-snow cover layer thickness gradient data corresponding to different ice surface-snow cover layer groove partitions.

[0042] In this embodiment of the invention, the thickness distribution data of the flat snow layer outside the grooves in different ice-snow-covered groove partitions is obtained through ice surface-snow-covered layer thickness distribution data. For partition A, the thickness data of the flat snow area outside the groove within a 1m radius of the groove is taken, and the average value of 500 sub-grids is calculated to be 0.25m. For partition B, the thickness data of the flat snow area outside the groove is taken within a 1.5m radius of the groove, and the average value of 800 sub-grids is 0.30m. Based on the difference between the flat snow layer thickness data and the corresponding snow accumulation thickness data inside the groove, the difference between the thickness of a sub-grid of 0.35m and the flat snow layer of 0.25m in a certain groove of partition A is 0.10m, and the average difference of all sub-grids is 0.12m. For partition B, the difference between the thickness of a sub-grid of 0.45m and the flat snow layer of 0.30m in a certain groove is 0.15m, and the average difference of the sub-grids is 0.18m. Ice-snow cover thickness gradient data were obtained for different ice-snow cover groove partitions. Partition A had a gradient range of 0.08-0.15m and a mean of 0.12m; partition B had a gradient range of 0.12-0.25m and a mean of 0.18m. The gradient data were stored in sub-grid coordinates, including the absolute value of the difference and its direction (positive for areas above the flat snow layer within the groove). The same regional range and statistical methods were used in the calculation process to ensure the comparability of the gradient values.

[0043] Furthermore, step S3 includes the following steps: Step S31: Obtain the corresponding irregular groove depth parameter data by partitioning the grooves of different ice surface-snow cover layers; In this embodiment of the invention, depth parameter data of irregular grooves are obtained by partitioning the grooves into different ice surface-snow cover layer groove shapes. Partition A (inverted triangular cone type) contains 35 grooves, and 500 depth sampling points (sampling interval 0.05m) are obtained for each groove through 3D scanning. The depth parameter data are the Z coordinate values ​​of these points (range 0.1-0.3m). For example, the depth sequence of groove A1 is 0.3m, 0.28m, 0.25m...0.1m (along the direction from the opening to the bottom of the groove). The statistical values ​​of the depth parameters of each groove are calculated: average depth (0.22m for A1), maximum depth (0.3m), minimum depth (0.1m), and depth standard deviation (0.05m). The depth parameter data of the 45 grooves in partition B (inverted square prism type) range from 0.3-0.5m. The average depth of groove B1 is 0.4m, the maximum depth is 0.5m, the minimum depth is 0.3m, and the standard deviation is 0.08m. All depth parameter data are stored according to groove ID and partition. Each data point is labeled with three-dimensional coordinates (X and Y accurate to 0.01m) to ensure that the depth parameters can accurately reflect the irregular shape of the groove. Data acquisition uses laser scanning equipment of the same precision, with no parameter deviation.

[0044] Step S32: Based on the irregular groove depth parameter data corresponding to different ice surface-snow cover groove partitions, perform relative thickening statistics within the grooves on the corresponding ice surface-snow cover thickness gradient data to obtain the relative thickening of the snow cover within the grooves corresponding to different ice surface-snow cover groove partitions. In this embodiment of the invention, the relative thickening within the corresponding ice-snow cover layer thickness gradient data is statistically analyzed based on the irregular groove depth parameter data corresponding to different ice-snow cover groove partitions. The thickness gradient data for partition A is the difference between the snow thickness within the groove and the thickness of the surrounding flat snow layer (range 0.08-0.15m). For groove A1, the thickness gradient values ​​corresponding to 500 depth sampling points are taken, and the average gradient for each depth interval (0.1-0.2m, 0.2-0.3m) is calculated: the average gradient for the 0.1-0.2m interval is 0.1m, and the average gradient for the 0.2-0.3m interval is 0.13m. The relative thickening is calculated cumulatively as "depth interval length × average gradient", and the relative thickening of A1 = (0.1m × 0.1) + (0.1m × 0.13) = 0.023m. The relative thickening of the 35 grooves in zone A ranged from 0.015 to 0.03 m, with an average of 0.022 m; the relative thickening of the 45 grooves in zone B ranged from 0.03 to 0.05 m, with an average of 0.04 m. The statistical results were compiled into a data table by zone, including the ID, average depth, and relative thickening of each groove. The calculation process employed strict interval division and arithmetic mean to ensure that the relative thickening accurately reflects the cumulative snow accumulation within the groove.

[0045] Step S33: Based on the irregular groove depth parameter data corresponding to different ice surface-snow cover groove partitions and combined with the corresponding relative thickening of the snow cover in the groove, construct the corresponding working depth adjustment model to determine the adjustment mathematical relationship between the working clearing depth and the groove depth and the relative thickening of the snow cover. Obtain the corresponding ice surface hardness parameter in the ice surface-snow cover groove partition. At the same time, set the corresponding working depth adjustment rate limit based on the ice surface hardness parameter, and impose adjustment limit constraints on the working depth adjustment model based on the working depth adjustment rate limit to avoid damage to the ice surface in the groove due to excessively rapid working depth adjustment. Calculate and output the corresponding depth adjustment benchmark parameter data. In this embodiment of the invention, a working depth adjustment model is constructed based on the irregular groove depth parameter data corresponding to different ice surface-snow cover groove partitions and the relative thickening of the snow cover within the grooves. The adjustment mathematical relationship of the model is: working clearing depth = groove depth + relative thickening + 0.02m (safety margin), such as the working clearing depth of groove A1 = 0.22 + 0.023 + 0.02 = 0.263m (rounded to three decimal places). The ice surface hardness parameters are obtained (measured by a hardness tester, 50MPa for partition A and 40MPa for partition B). Based on the hardness parameters, the working depth adjustment rate limit is set: adjustment rate for partition A (high hardness) ≤ 0.05m / s, and for partition B (low hardness) ≤ 0.08m / s. The working depth adjustment model is subject to adjustment constraints. When the working depth difference between adjacent clearing points is > 0.05m, a forced adjustment is performed in stages according to the adjustment rate (e.g., it takes ≥ 1 second to go from 0.2m to 0.3m). The calculated output depth adjustment reference parameters are as follows: Reference depth for zone A is 0.2-0.35m, adjustment rate is 0.05m / s; reference depth for zone B is 0.35-0.55m, adjustment rate is 0.08m / s. These parameters are correlated with the groove zones and hardness values ​​to ensure the model's output working depth can clear snow without damaging the ice surface. All mathematical relationships and limits are determined based on physical properties, without empirical adjustments.

[0046] Step S34: Divide the snow removal area based on the depth adjustment reference parameter data of the ice surface-snow cover thickness distribution data, so as to merge the areas with similar depth adjustment reference parameter data into the same removal adjustment unit, so as to generate ice surface-snow cover adaptive removal area data, including the working depth adjustment range corresponding to each removal adjustment unit. In this embodiment of the invention, snow removal areas are divided based on the ice-snow cover thickness distribution data using depth adjustment reference parameter data. Areas with a depth adjustment reference parameter difference ≤ 0.05m are merged into the same snow removal adjustment unit. In partition A, areas with a reference depth of 0.2-0.25m are merged into unit A1 (area 50m²). 2 The area of ​​0.25-0.3m is merged into unit A2 (area 60m²). 2 The area of ​​0.3-0.35m is merged into unit A3 (area 40m²). 2 In zone B, the area between 0.35 and 0.4 m is merged into unit B1 (area 70 m²). 2 The area of ​​0.4-0.45m is merged into unit B2 (area 80m²). 2 The area of ​​0.45-0.55m is merged into unit B3 (area 60m²). 2The working depth adjustment range for each clearing adjustment unit is the reference depth ±0.02m (e.g., unit A1: 0.18-0.27m). Unit boundaries are defined using a Voronoi diagram (distance between adjacent unit boundaries ≤ 0.5m). Adaptive clearing region data for the ice surface and snow cover is generated, including the IDs, areas, reference depths, adjustment ranges, and included groove IDs for the six clearing adjustment units. Region division ensures consistency of depth adjustment parameters within units, with no overlapping or omissions at boundaries.

[0047] Step S35: Based on the depth adjustment reference parameter data, pre-set the snow removal execution action for the ice surface-snow cover layer adaptive removal area data to obtain the pre-set execution data for the ice surface-snow cover layer removal area.

[0048] In this embodiment of the invention, snow removal actions are pre-defined based on depth adjustment reference parameter data for the adaptive snow removal area data of the ice surface-snow cover layer. The actions of each snow removal adjustment unit include working snow removal depth, tool pressure, and travel speed. Unit A1 has a working snow removal depth of 0.23m (reference depth 0.23m), a tool pressure of 300N (set based on ice surface hardness of 50MPa), and a travel speed of 0.5m / s (due to depth ≤ 0.3m); Unit B2 has a working snow removal depth of 0.43m, a tool pressure of 250N (hardness 40MPa), and a travel speed of 0.3m / s (depth > 0.3m). The pre-defined actions are arranged sequentially by unit, and the switching between actions of adjacent units must meet the adjustment rate limit (e.g., from unit A3 to B1, the working depth increases from 0.3m to 0.4m, taking ≥ 1.25 seconds). The system generates preset data for the ice surface-snow cover removal zone, including the action parameters, execution order, and switching time of 6 units. The data is labeled with the start / end coordinates of each action (accurate to 0.1m), ensuring that the execution action strictly matches the depth parameters of the removal unit. The action parameters are calculated based on physical formulas (pressure = hardness × contact area) and there are no subjective settings.

[0049] Furthermore, step S32 includes the following steps: The depth parameter data of irregular grooves corresponding to different ice surface-snow cover layer groove partitions are deconstructed into groove topology to convert the corresponding irregular groove depth parameter data into three-dimensional surface coordinates. A polar coordinate system is established with the lowest point of the groove bottom as the origin. At the same time, the radial distribution of the depth value at different angles is calculated to obtain the radial distribution of the irregular groove depth corresponding to different ice surface-snow cover layer groove partitions. In this embodiment of the invention, the depth parameter data of irregular grooves corresponding to different ice-snow-covered groove partitions are used to perform topological deconstruction of the groove morphology. The depth parameter data of a certain irregular groove in partition A contains 500 three-dimensional coordinate points (X, Y, Z), with the Z value being the depth (0.1-0.3m). These data are converted into three-dimensional surface coordinates, and a continuous surface (triangle side length ≤ 0.1m) is constructed through Delaunay triangulation. A polar coordinate system is established with the lowest point at the bottom of the groove (X=2.5m, Y=3.0m, Z=0.3m) as the origin, with the polar axis pointing to the widest part of the groove opening, and the polar angle θ ranging from 0° to 360° (interval of 1°). The radial distance r is the horizontal distance from the origin to the surface point (0-2.5m). The radial distribution of depth values ​​at different angles was calculated. When θ=0°, the depth at r=0m is 0.3m, at r=1m is 0.25m, and at r=2m is 0.1m; when θ=90°, the depth at r=0m is 0.3m, at r=1.5m is 0.2m, and at r=2.5m is 0.1m. Radial depth data at different angles were obtained, with 25 radial points corresponding to each angle. The data were sorted by polar angle to form a radial distribution table of irregular groove depths corresponding to partition A. All coordinate transformations and calculations were based on strict geometric formulas to ensure that the radial distribution accurately reflects the groove morphology.

[0050] Preferably, the inflection points and turning points of the irregular groove edges corresponding to different ice surface-snow cover layer groove partitions are obtained according to the radial distribution of the irregular groove depths of different ice surface-snow cover layer groove partitions. In this embodiment of the invention, edge inflection points and turning points are obtained by analyzing the radial distribution of irregular groove depths corresponding to different ice-snow-covered groove partitions. In the radial distribution data of a certain groove in partition B, the radial depth sequence in the θ=180° direction is: r=0m (0.5m), r=1m (0.45m), r=2m (0.3m), r=3m (0.15m), r=4m (0.1m). The depth change rate of adjacent points is calculated: the change rate at r=1-2m is (0.3-0.45) / (2-1)=-0.15m / m, at r=2-3m it is (0.15-0.3) / (3-2)=-0.15m / m, and at r=3-4m it is (0.1-0.15) / (4-3)=-0.05m / m. Points where the absolute value of the rate of change abruptly exceeds 0.1 m / m are identified as inflection points. For example, at r=3m, the rate of change changes from -0.15 to -0.05, a difference of 0.1, thus identifying an inflection point. Points where the second derivative changes from positive to negative are identified as inflection points. At r=2m, the second derivative is 0, and the rate of change of the first derivative changes from 0 to 0.1, thus identifying an inflection point. Partition B identified a total of 8 inflection points and 5 inflection points. Each point is labeled with polar coordinates (r, θ) and a depth value. The criteria for identifying inflection points and inflection points are fixed to ensure consistency in the edge feature extraction methods for different grooves.

[0051] Preferably, based on the irregular groove edge inflection points and irregular groove edge turning points corresponding to different ice surface-snow cover groove partitions, the corresponding ice surface-snow cover groove partitions are divided into groove sub-regions to obtain the irregular groove snow cover sub-partitions corresponding to different ice surface-snow cover groove partitions. In this embodiment of the invention, the groove partitions are subdivided based on the irregular groove edge inflection points and turning points corresponding to different ice surface-snow cover layer groove partitions. A certain groove in partition A has 3 inflection points (P1, P2, P3) and 2 turning points (Q1, Q2), arranged in polar angle order as Q1 (θ=60°), P1 (θ=120°), Q2 (θ=240°), P2 (θ=300°), and P3 (θ=360°). Radial lines are formed by connecting adjacent feature points with the origin as the center. The area between Q1 and P1 is sub-region A1, between P1 and Q2 is sub-region A2, between Q2 and P2 is sub-region A3, and between P2 and P3 is sub-region A4. Each sub-region is bounded by two radial lines and the groove edge. The sub-region boundaries are strictly defined along the lines connecting the feature points. The area is calculated using the polar coordinate sector area formula (0.5×r). 2 ×Δθ), the area of ​​subregion A1 is 0.8m. 2 A2 has an area of ​​1.2m² 2 A3 has an area of ​​0.9m². 2 A4 size, 1.1m² 2 Four irregular groove-covered sub-regions corresponding to partition A are obtained. Each sub-region is labeled with boundary feature point ID and area. The partitioning results ensure that the sub-regions do not overlap and cover the entire groove area.

[0052] Preferably, based on the irregular groove sub-regions within the irregular grooves corresponding to different ice-snow-covered groove partitions, the depth of the irregular groove sub-regions and the increase in the difference in ice-snow-covered thickness are selected from the corresponding irregular groove depth parameter data and ice-snow-covered thickness gradient data. Then, based on the depth of the irregular groove sub-regions and the increase in the difference in ice-snow-covered thickness corresponding to the same sub-region, the relative thickening within the groove is cumulatively calculated to obtain the relative thickening of the snow-covered layer within the grooves corresponding to different ice-snow-covered groove partitions.

[0053] In this embodiment of the invention, sub-region data is selected from depth parameter data and thickness gradient data by using irregular groove snow-covered sub-regions corresponding to different ice-snow-covered groove partitions. Sub-region B1 of partition B contains 100 depth parameter points with an average depth of 0.4m. In the corresponding thickness gradient data, the thickness difference increment of sub-region B1 is 0.2m (0.5m thickness in the groove minus 0.3m for the flat snow layer). The depth of the irregular groove sub-region is taken as the average depth of all points in the sub-region (0.4m), and the thickness difference increment of the ice-snow-covered layer corresponding to the same sub-region is 0.2m. The cumulative calculation of relative thickness within the groove is performed using the formula: relative thickness = sub-region depth × thickness difference increment / flat snow layer thickness (0.3m). The relative thickness of sub-region B1 = 0.4 × 0.2 / 0.3 = 0.27m. The relative thickening of the snow cover in each of the five sub-regions of zone B was calculated. The relative thickening of sub-region B2 was 0.32m, B3 was 0.25m, B4 was 0.30m, and B5 was 0.22m. The relative thickening of the snow cover within the groove corresponding to zone B was obtained. The calculated value of each sub-region was rounded to two decimal places. The data was associated with the sub-region ID and area to ensure that the thickening value could reflect the cumulative thickening of the snow cover relative to the flat snow layer within the sub-region. All parameter values ​​in the calculation process were taken from previous data without any additional adjustments.

[0054] Furthermore, step S35 includes the following steps: Step S351: By extracting the spatial coordinates and working depth adjustment range of each clearing adjustment unit from the ice surface-snow cover adaptive clearing area data, and combining the groove spatial distribution corresponding to the ice surface-snow cover thickness distribution data, basic data for ice surface-snow cover clearing area path planning is generated. In this embodiment of the invention, the spatial coordinates (X: 0-100m, Y: 0-100m, accuracy 0.01m) and working depth adjustment range (0.1-0.6m, step size 0.05m) corresponding to each clearing adjustment unit (2m×2m, a total of 2500 units) are extracted from the ice-snow cover adaptive clearing area data (range 100m×100m). Combined with the spatial distribution of grooves corresponding to the ice-snow cover thickness distribution data (80 grooves, including 40 deep grooves with depth > 0.3m and 40 shallow grooves with depth ≤ 0.3m), the minimum bounding rectangle coordinates of each groove are marked (e.g., deep groove 1: X 10-15m, Y 20-25m). Basic data for path planning of the ice-snow cover clearing area is generated, including the coordinates and working depth range of the 2500 units, as well as the position, depth, and clearing unit ID of the 80 grooves. The data is sorted by the clearing unit number, and each unit is associated with whether it contains a groove (yes / no) and the groove type (deep / shallow) to ensure that the basic data for path planning can accurately associate the clearing units with the spatial distribution of the grooves. The coordinate system adopts Gauss-Kruger projection with no data offset.

[0055] Step S352: Construct a path optimization model based on the basic data of the ice surface-snow cover removal area path planning, with the objective function being to maximize the removal coverage and minimize the path length. Introduce priority removal weight rules corresponding to the ice surface-snow cover grooves, including that the weight of deep grooves is higher than that of shallow grooves. At the same time, generate corresponding path optimization objective parameter data for the path optimization model based on the objective function and the priority removal weight rules. In this embodiment of the invention, a path optimization model is constructed based on path planning data for the ice-snow-covered layer clearing area. Objective function 1 is to maximize the clearing coverage, calculated as (number of cleared units / total number of units) × 100%. Objective function 2 is to minimize the path length, in meters. Priority clearing weight rules are introduced for the ice-snow-covered layer grooves: 1.5 for deep grooves, 1.2 for shallow grooves, and 1.0 for non-groove areas. These weights directly affect the coverage calculation (weighted coverage = Σ unit weight × clearing status). Constraints include path continuity (distance between adjacent units ≤ 2.83m, diagonal distance) and turning angle ≤ 90°. The generated path optimization target parameter data are: total number of units 2500, 300 units in deep grooves, 400 units in shallow grooves, and 1800 units in non-groove areas; maximum allowable path length 5000m, minimum coverage target 85%. Model parameters are fixed, and weight rules have no hierarchical overlap, ensuring clear objective function calculation logic and a clear priority clearing order.

[0056] Step S353: The A* algorithm is used in conjunction with the path optimization target parameter data to solve the path optimization model, so as to obtain the optimal clearing trajectory from the starting point to the end point of the adaptive clearing area of ​​ice surface-snow cover, which includes the order and turning angle of passing through each clearing adjustment unit, so as to obtain the initial path planning data. In this embodiment of the invention, the path optimization model is solved by employing the A* algorithm combined with path optimization target parameter data. The algorithm's heuristic function is the Manhattan distance (h = |X target - X current| + |Y target - Y current|), and the cost function is f = g + h (g is the length of the path already traveled, and h is the estimated remaining length). From the starting point (X0, Y0) to the ending point (X100, Y100), priority is given to clearing units containing deep groove areas. Each time, 8 adjacent units are expanded (up, down, left, right, and diagonal). The f value of each unit is calculated, and the unit with the smallest f value is selected as the next station. The optimal clearing trajectory is obtained, passing through 1200 clearing units, including 280 deep groove area units (coverage rate 93.3%), 350 shallow groove area units (coverage rate 87.5%), and 570 non-groove area units (coverage rate 31.7%). The trajectory includes the order (1-1200) of passing through each clearing adjustment unit and the turning angle (e.g., turning 30° from unit 10 to unit 11, turning 0° from unit 11 to unit 12, and going straight). The turning angle is calculated using the coordinates of adjacent units (the arctangent of ΔX and ΔY) with an accuracy of 1°. The initial path planning data is labeled with the arrival time of each unit (calculated based on a travel speed of 0.5 m / s) to ensure that the trajectory is continuous and does not pass through units repeatedly.

[0057] Step S354: Perform a smoothness check on the initial path planning data to calculate the turning radius and curvature change rate corresponding to the optimal clearing trajectory. When the curvature change rate exceeds a preset threshold, insert the corresponding transition line segment to obtain the optimized path planning data. In this embodiment of the invention, the smoothness of the initial path planning data is verified, and the turning radius (arc radius calculated based on three-point coordinates, accuracy 0.1m) and curvature change rate (unit: radians / meter, accuracy 0.01 rad / m) of each turning point in the optimal clearing trajectory are calculated. A preset curvature change rate threshold of 0.5 rad / m is set. Fifty turning points in the trajectory have curvature change rates exceeding this threshold (e.g., turning point 200: curvature change rate 0.6 rad / m). Transition segments of 1m length are inserted at these points, connecting the units before and after the original turning points, increasing the turning radius from 2m to 3m and reducing the curvature change rate to 0.4 rad / m. The optimized path length increases from 4800m to 4850m (an increase of 1.04%), while the clearing coverage remains unchanged (90.2%). The optimized path planning data is obtained, which includes 1250 unit sequences (with 50 new transition units). The corrected turning radius and curvature change rate are marked at each turning point to ensure that the curvature change rate of all turning points of the trajectory is ≤0.5rad / m, and the smoothness meets the mechanical limitations of the clearing equipment (minimum turning radius 2m).

[0058] Step S355: Based on the corresponding trajectory nodes in the optimized path planning data and combined with the depth adjustment reference parameter data, determine the working clearing depth, tool angle and travel speed corresponding to each trajectory node, and obtain the preset data for the ice surface-snow layer clearing area.

[0059] In this embodiment of the invention, based on the corresponding trajectory nodes (1250 nodes, each node being the center coordinates of the clearing unit) within the optimized path planning data, and combined with depth adjustment reference parameter data (working depth in deep groove area = groove depth + 0.05m, shallow groove area = groove depth + 0.03m, non-groove area = 0.1m), the working clearing depth corresponding to each trajectory node is determined as follows: For nodes in deep groove area (e.g., node 300), the groove depth is 0.4m and the working depth is 0.45m; for nodes in shallow groove area (e.g., node 500), the groove depth is 0.25m and the working depth is 0.28m; for nodes in non-groove area (e.g., node 800), the working depth is 0.1m. The tool angle is adjusted according to the groove inclination angle (30° in deep groove area, 20° in shallow groove area, 10° in non-groove area), and the travel speed is set based on the working depth (0.3m / s when depth > 0.3m, 0.5m / s when depth ≤ 0.3m). The system obtains preset data for the ice-snow layer removal zone. Each node includes working depth (accuracy 0.01m), tool angle (accuracy 1°), and travel speed (accuracy 0.01m / s). The data is arranged in the order of the trajectory nodes to ensure that the parameter adjustment strictly corresponds to the trajectory position and meets the adaptive removal requirements.

[0060] Furthermore, step S4 includes the following steps: Step S41: Based on the preset data for the ice-snow layer removal area, start the removal execution mechanism to perform the corresponding rapid ice-snow layer removal operation, and collect the corresponding ice-snow layer removal residue image data during the removal process through real-time feedback sensors; In this embodiment of the invention, the snow removal mechanism (including a snow milling cutter, a depth adjustment screw, and a tracked traveling device) is activated based on preset data (working depth of unit A1: 0.26m, cutter pressure: 300N, traveling speed: 0.5m / s; working depth of unit A2: 0.273m, cutter pressure: 280N, traveling speed: 0.4m / s) set for the ice-snow removal zone. The mechanism operates along a predetermined trajectory (units A1→A2 in sequence), with the snow milling cutter rotating at 1000 r / min and contacting the ice surface at a pressure of 500 Pa. Simultaneously, real-time feedback sensors (high-definition camera, resolution 4096×3072, frame rate 30fps; laser rangefinder, sampling frequency 50Hz, accuracy ±0.5mm) collect data on the snow removal process. A camera is installed 0.5m in front of the actuator, shooting vertically downwards. Each frame covers a 2m x 2m area, recording the grayscale characteristics of the remaining snow (snow grayscale value 180-255, ice surface 80-120). A laser rangefinder simultaneously measures the distance between the snow surface and the ice surface, generating thickness data. The collected image data of the ice-snow layer removal residue is stored according to timestamps (accurate to milliseconds) and unit IDs. A data block is generated every 10 seconds, containing 200 frames and 500 thickness sampling points, ensuring data coverage of the entire removal process with no blind spots.

[0061] Step S42: Perform snow cover layer residual analysis on the ice surface-snow cover layer removal residual image data to identify the areas of snow accumulation that have not been removed from the ice surface-snow cover layer using image segmentation algorithms, and calculate the corresponding residual area and snow thickness based on the areas of snow accumulation that have not been removed from the ice surface-snow cover layer to obtain snow cover layer removal residual data. In this embodiment of the invention, snow cover residual analysis is performed on the residual image data of the ice-snow-covered layer removal. The U-Net image segmentation algorithm (input image size 512×512 pixels, output binary mask) is used, with a mask value of 1 for snow-covered areas and 0 for ice-covered areas. The algorithm segments the image based on the grayscale difference (threshold 100) between snow and ice. After segmenting 200 frames of image in unit A1, the pixel percentage of the unremoved snow-covered area (15%) is calculated. Combined with the image scale (1 pixel = 0.004m), this is converted into a residual area of ​​2m × 2m × 15% = 0.6m. 2 Meanwhile, the arithmetic mean (0.03m) of the thickness data collected by the laser rangefinder was taken as the average thickness of the area with unremoved snow. The image segmentation results for unit A2 show that the percentage of pixels with unremoved snow is 12%, and the residual area is 2.5m × 2.5m × 12% = 0.75m. 2 The average thickness measured by laser was 0.02m. Residual data of snow cover removal were obtained, including the residual area of ​​each cell (accurate to 0.01m). 2The data and average residual thickness (accurate to 0.001m) correspond one-to-one with the clearing unit, and the segmentation algorithm parameters are fixed (convolution kernel size 3×3, activation function ReLU) to ensure that the residual analysis results can be repeatedly verified.

[0062] Step S43: Based on the snow cover layer removal residual data and combined with the ice surface-snow cover layer thickness correlation feature data, perform iterative optimization of the working depth adjustment model for ice surface-snow cover layer removal to output the corresponding rapid ice surface-snow cover layer removal results.

[0063] In this embodiment of the invention, residual data (residual area of ​​cell A1 is 0.6m²) is removed based on the snow cover layer. 2 Thickness 0.03m; residual area of ​​element A2 0.75m² 2 The working depth adjustment model was iteratively optimized using the ice surface-snow cover thickness correlation characteristic data (A1 groove depth 0.22m, relative thickening 0.023m; A2 groove depth 0.25m, relative thickening 0.021m). The first step involved calculating the depth correction (A1=0.03×(0.22 / 0.22)=0.03m; A2=0.02×(0.25 / 0.22)=0.023m) and updating the depth adjustment baseline parameters (A1=0.26+0.03=0.29m; A2=0.273+0.023=0.296m). The second step involves calculating a new adjustment value (A1 = 6 × 0.03 + 0.12 × 0.0036 + 0.6 × 0.018 = 0.1912m) using a PID control algorithm (Kp = 6, Ki = 0.12, Kd = 0.6), which is then converted into a control signal (12mA current). The third step involves driving the clearing actuator for secondary operation. Unit A1 has a working depth of 0.29m, and A2 has a working depth of 0.296m. The travel speed is reduced to 0.3m / s to improve clearing accuracy. After the secondary operation, residual analysis shows that the residual thickness of A1 is 0.01m and A2 is 0.008m, both meeting the preset threshold (≤0.02m). The output shows the rapid clearing results of the ice surface and snow cover layer, including the final residual area (A1 0.1m²). 2 A20.15m 2 The average thickness (A1 0.01m, A2 0.008m) and total cleaning time (45 minutes) were measured. All analysis and optimization steps were based on fixed algorithms and calculation formulas. Residual data directly drove parameter updates without human intervention.

[0064] Furthermore, step S43 includes the following steps: Based on the snow cover layer removal residual data and combined with the ice surface-snow cover layer thickness correlation feature data, the depth correction calculation is performed on the depth adjustment benchmark parameter data output by the working depth adjustment model to obtain the working depth adjustment correction amount data. In this embodiment of the invention, the depth correction calculation is performed on the depth adjustment reference parameter data (original reference depth 0.23m) output by the working depth adjustment model based on the snow cover layer removal residual data (the residual of unit A1 is 0.03m, i.e., the difference between the actual residual thickness and the expected removal thickness) and combined with the ice surface-snow cover layer thickness correlation feature data (the groove depth of A1 is 0.22m, and the relative thickness increase is 0.023m). The correction formula is: working depth adjustment correction amount = removal residual × (groove depth / average groove depth). The average groove depth of A1 is 0.22m, and the correction amount = 0.03 × (0.22 / 0.22) = 0.03m. The residual of unit A2 is 0.02m, and the average groove depth is 0.25m, and the correction amount = 0.02 × (0.25 / 0.22) = 0.023m. The working depth adjustment correction data were obtained. After correction, the reference depth of unit A1 is 0.26m and that of A2 is 0.273m. The correction amount is retained to three decimal places. The calculation process is strictly based on the numerical relationship between residuals and correlation characteristics to ensure that the correction amount can compensate for and eliminate residuals, without subjective adjustment factors.

[0065] Preferably, the working depth adjustment correction data is input into the working depth adjustment model and the adjustment amount corresponding to the clearing actuator is calculated by the PID control algorithm. Based on the adjustment amount corresponding to the clearing actuator, the corresponding control signal is output to obtain the real-time control signal data for clearing execution. In this embodiment of the invention, the working depth adjustment correction data is input into the working depth adjustment model, and the adjustment amount corresponding to the clearing actuator is calculated by the PID control algorithm. The PID parameters are set as proportional coefficient Kp=5, integral coefficient Ki=0.1, and derivative coefficient Kd=0.5. The adjustment amount calculation formula is: adjustment amount = Kp × correction amount + Ki × ∫ correction amount dt + Kd × d(correction amount) / dt. The correction amount of unit A1 is 0.03m, the integral term is 0.003 (integration time 0.1 seconds), the derivative term is 0.015 (rate of change 0.5m / s), and the adjustment amount = 5 × 0.03 + 0.1 × 0.003 + 0.5 × 0.015 = 0.15 + 0.0003 + 0.0075 = 0.1578m. The control signal corresponding to the output of the adjustment amount is a 4-20mA current signal (0.1578m corresponds to 10mA), with a frequency of 50Hz and a pulse width of 0.02 seconds. Real-time control signal data for clearing execution is obtained, including the current value, frequency, and pulse width of each clearing unit. The signal parameters are linearly correlated with the adjustment amount (0-0.5m corresponds to 4-20mA), ensuring that the control signal can accurately drive the actuator. All parameters are fixed and reproducible.

[0066] Preferably, the real-time control signal data for clearing is sent to the drive module corresponding to the clearing actuator to drive the depth adjustment component to perform the corresponding depth adaptive adjustment, thereby obtaining real-time depth adjustment data; based on the real-time depth adjustment data, the precise clearing operation corresponding to the ice surface-snow layer clearing area is completed, and the ice surface-snow layer clearing area is simultaneously scanned a second time by a laser scanning device to obtain the secondary residual thickness of the snow layer and the area of ​​ice surface damage traces corresponding to the ice surface-snow layer clearing area, thereby obtaining the original data of the clearing effect; In this embodiment of the invention, by sending real-time control signal data for the clearing operation to the corresponding drive module (stepper motor drive module, step angle 1.8°, reduction ratio 10:1) of the clearing execution mechanism, the depth adjustment component (screw and nut mechanism, pitch 5mm) is driven to perform adaptive depth adjustment. For every 1mA current signal received, the screw rotates 10 revolutions, adjusting the depth by 0.05m. Unit A1 receives a 10mA signal, the screw rotates 100 revolutions, adjusting the depth by 0.5m, obtaining real-time depth adjustment data (0.26m). After completing the clearing operation based on the real-time depth adjustment data, a laser scanning device (scanning frequency 100Hz, accuracy ±0.1mm) performs a secondary scan of the area to obtain the secondary residual thickness of the snow layer (A1 average 0.01m) and the area of ​​ice surface damage traces (A1 0.5m). 2 The original data on the removal effect were recorded by unit, including residual thickness (accurate to 0.001m) and damaged area (accurate to 0.01m). 2 The scanning range completely overlaps with the area to be cleared (50m). 2 This ensures that the data reflects the actual cleaning effect and that there are no missing data areas.

[0067] Preferably, the original thickness distribution of the snow cover layer and the total area of ​​the ice surface region corresponding to the ice-snow cover layer removal area are obtained, and the removal effect evaluation calculation is performed on the original data of the removal effect based on the original thickness distribution of the snow cover layer and the total area of ​​the ice surface region to calculate the snow residue rate and ice surface damage degree corresponding to the ice-snow cover layer removal area. The snow residue rate is the ratio of the secondary residual thickness of the snow cover layer to the original thickness distribution of the snow cover layer, and the ice surface damage degree is the ratio of the area of ​​ice surface damage traces to the total area of ​​the ice surface region, so as to obtain the removal effect evaluation data. In this embodiment of the invention, the original snow cover thickness distribution (A1 average 0.25m) and the total area of ​​the ice surface region (50m²) corresponding to the ice surface-snow cover removal area are obtained. 2The original data on snow removal effectiveness were evaluated and calculated. Snow residue rate = secondary residual snow layer thickness / original snow layer thickness distribution = 0.01 / 0.25 = 0.04 (4%); Ice surface damage degree = area of ​​ice surface damage marks / total area of ​​ice surface area = 0.5 / 50 = 0.01 (1%). In unit A2, the original snow layer thickness was 0.3m, the secondary residual thickness was 0.015m, the snow residue rate was 0.05 (5%), and the area of ​​ice surface damage marks was 0.3m². 2 The ice surface area is 60m² 2 The damage level was 0.005 (0.5%). The resulting snow removal effectiveness assessment data included the snow residue rate (retained to two decimal places) and ice surface damage level (retained to three decimal places) for each unit. The calculation process was based on a strict ratio formula to ensure that the assessment data quantifies the snow removal effect and eliminates subjective evaluation factors.

[0068] Preferably, the clearing effect evaluation data is compared with a preset effect threshold to screen out the sub-areas that have not met the clearing standards corresponding to the ice-snow-covered layer clearing area. Based on the sub-areas that have not met the clearing standards, the parameters of the working depth adjustment model are reversed to generate model optimization parameters. Based on the model optimization parameters, the corresponding working depth adjustment model is updated to perform iterative optimization of ice-snow-covered layer clearing to output the corresponding rapid clearing results of ice-snow-covered layer.

[0069] In this embodiment of the invention, by comparing the snow removal effect evaluation data with preset effect thresholds (snow residue rate ≤3%, ice surface damage ≤0.5%), unit A1 has a snow residue rate of 4% (>3%) and an ice surface damage of 1% (>0.5%), and is therefore determined to be a sub-area (area 50m²) that has not met the removal standards. 2 Based on the substandard sub-areas, the working depth adjustment model parameters were corrected in reverse: the safety margin was increased from 0.02m to 0.03m, and the PID proportional coefficient Kp was increased from 5 to 6. Model optimization parameters were generated (safety margin 0.03m, Kp=6), the working depth adjustment model was updated, and the working clearing depth was recalculated (A1=0.22+0.023+0.03=0.273m). Iterative optimization was performed, and after the second clearing, the snow residue rate of A1 was 2% (≤3%), and the ice surface damage was 0.3% (≤0.5%), meeting the standards. The rapid clearing results of the ice surface and snow cover layer were output, including the snow residue rate (average 2.5%), ice surface damage (average 0.4%), and clearing time (30 minutes) of the final cleared area. The optimization process strictly corrected parameters based on the substandard data, ensuring that the effect met the threshold after iteration and avoiding infinite loops.

[0070] Furthermore, the present invention also provides a rapid ice and snow removal device with adaptive working depth adjustment, including a processor, a memory, and a computer program stored in the memory and executable on the processor, for performing the rapid ice and snow removal method with adaptive working depth adjustment as described above.

[0071] Therefore, the embodiments should be considered as exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of the equivalents of the application are intended to be included within the invention.

[0072] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A method for rapid snow removal from ice surfaces with adaptive adjustment of working depth, characterized in that, Includes the following steps: Step S1: Obtain the corresponding three-dimensional ice surface topography point cloud data through the target ice surface area, and perform snow cover thickness inversion calculation on the three-dimensional ice surface topography point cloud data to generate ice surface-snow cover thickness distribution data. Based on the three-dimensional topographic point cloud data of the ice surface and the ice-snow cover thickness distribution data, the correlation mapping between the ice surface groove and the snow cover thickness is carried out to obtain the correlation feature data of ice surface-snow cover thickness. Step S2: Divide the target ice surface region into ice surface groove morphology based on ice surface-snow cover thickness correlation feature data to generate different ice surface-snow cover groove morphology partitions; perform snow cover difference analysis on ice surface-snow cover thickness distribution data based on different ice surface-snow cover groove morphology partitions to obtain ice surface-snow cover thickness gradient data corresponding to different ice surface-snow cover groove partitions. Step S3: Construct a working depth adjustment model based on the ice surface-snow cover layer groove morphology partition and the corresponding ice surface-snow cover layer thickness gradient data to output the corresponding depth adjustment reference parameter data, and perform snow removal execution actions on the ice surface-snow cover layer thickness distribution data based on the depth adjustment reference parameter data to obtain the pre-set data for the ice surface-snow cover layer removal area. Step S4: Start the rapid ice-snow layer removal operation according to the preset data of the ice-snow layer removal area, and collect the snow layer removal residual data corresponding to the removal process through real-time feedback sensors; perform iterative optimization of the working depth adjustment model for ice-snow layer removal based on the snow layer removal residual data and combined with the ice-snow layer thickness correlation feature data, so as to output the corresponding rapid ice-snow layer removal results.

2. The method for rapid snow removal from ice surfaces with adaptive working depth adjustment according to claim 1, characterized in that, Step S1 includes the following steps: Step S11: The target ice surface area is scanned by deploying a lidar scanner to collect the corresponding ice surface elevation data and ice surface outline point cloud data within each grid cell; Step S12: Based on the ice surface elevation data and ice surface contour range point cloud data corresponding to each grid cell, perform three-dimensional topographic coordinate registration on the target ice surface area to obtain three-dimensional topographic point cloud data of the ice surface. Step S13: Perform snow cover thickness inversion calculation on the three-dimensional topographic point cloud data of the ice surface to generate ice surface-snow cover thickness distribution data; Step S14: Perform spatial alignment processing on the three-dimensional ice surface topography point cloud data and the ice surface-snow cover thickness distribution data in the same three-dimensional coordinate system to obtain the corresponding ice surface topography point cloud data and ice surface-snow cover thickness distribution data in the same three-dimensional spatial coordinate system. Step S15: Based on the ice surface topography point cloud data and ice surface-snow cover thickness distribution data in the same three-dimensional spatial coordinate system, perform the correlation mapping between ice surface grooves and snow cover thickness, so as to obtain the correlation mapping between the geometric morphology parameters of the irregular ice surface grooves and the corresponding snow cover thickness distribution through the ice surface topography point cloud data, and obtain the ice surface-snow cover thickness correlation feature data.

3. The method for rapid snow removal from ice surfaces with adaptive working depth adjustment according to claim 2, characterized in that, Step S13 includes the following steps: Step S131: Perform ice surface snow cover point cloud density statistics on each grid cell in the ice surface three-dimensional topography point cloud data to obtain the corresponding ice surface snow cover point cloud distribution density in each grid cell; Step S132: Based on the distribution density of the snow-covered point cloud on the ice surface in each grid cell, perform spatial clustering of the snow-covered area of ​​the three-dimensional shape point cloud data of the ice surface to obtain the snow-covered spatial region of each three-dimensional shape of the ice surface. Step S133: Extract the snow cover zone boundary and describe the morphological features of each three-dimensional snow cover space region of the ice surface to generate a geometric boundary feature library of the ice surface snow cover region; perform ice surface-snow cover interface reflection detection based on the ice surface-snow cover region geometric boundary feature library to calculate the point cloud reflectivity between the ice surface-snow cover interface through local comparison and identify the corresponding ice surface-snow cover interface candidate points; at the same time, perform three-dimensional spatial interpolation based on the ice surface-snow cover interface candidate points to generate the initial ice surface-snow cover interface surface. Step S134: Perform voxel mesh downsampling on the three-dimensional topographic point cloud data of the ice surface to generate simplified three-dimensional point cloud data of the ice surface; based on the simplified three-dimensional point cloud data of the ice surface, perform ground point and non-ground point segmentation to obtain the basic three-dimensional point cloud data of the ice surface; Step S135: The initial ice-snow-covered interface surface is optimized and fitted using the least squares method to remove the corresponding misjudged points of the ice-snow-covered layer using the Mahalanobis distance discrimination criterion, and the corresponding ice-snow-covered interface model is generated based on the remaining ice-snow-covered interface points; the thickness vertical distance between the ice-snow-covered interface models is calculated based on the three-dimensional basic point cloud data of the ice surface to generate ice-snow-covered layer thickness distribution data.

4. The method for rapid snow removal from ice surfaces with adaptive working depth adjustment according to claim 1, characterized in that, Step S2 includes the following steps: Step S21: Perform morphological analysis on the ice surface groove contour corresponding to the ice surface-snow cover thickness correlation feature data, so as to identify the opening width, opening depth, tilt angle and cross-sectional shape parameters of the corresponding irregular groove on the ice surface, so as to obtain the basic morphological data of the ice surface groove. Step S22: Based on the basic morphological data of the ice surface grooves, the corresponding ice surface groove areas within the target ice surface area are divided into ice surface groove morphology. The grooves are clustered into different morphology categories according to the opening size and depth ratio of different ice surface groove contours, including groove areas corresponding to inverted triangular cones and inverted quadrangular prisms, to generate different ice surface-snow cover layer groove morphology partitions. Step S23: Based on the different ice surface-snow cover layer groove morphology partitions, the ice surface-snow cover layer thickness distribution data is divided into different grooves to determine the snow layer thickness distribution data in the ice surface-snow cover layer grooves corresponding to different ice surface-snow cover layer groove partitions; Step S24: Obtain the snow layer thickness distribution data outside the grooves in the same area corresponding to different ice surface-snow cover layer groove partitions through the ice surface-snow cover layer thickness distribution data, and calculate the snow cover layer thickness difference between the snow layer thickness distribution data outside the grooves in the same area corresponding to different ice surface-snow cover layer groove partitions and the corresponding snow thickness distribution data inside the ice surface-snow cover layer grooves to obtain the ice surface-snow cover layer thickness gradient data corresponding to different ice surface-snow cover layer groove partitions.

5. The method for rapid snow removal from ice surfaces with adaptive working depth adjustment according to claim 1, characterized in that, Step S3 includes the following steps: Step S31: Obtain the corresponding irregular groove depth parameter data by partitioning the grooves of different ice surface-snow cover layers; Step S32: Based on the irregular groove depth parameter data corresponding to different ice surface-snow cover groove partitions, perform relative thickening statistics within the grooves on the corresponding ice surface-snow cover thickness gradient data to obtain the relative thickening of the snow cover within the grooves corresponding to different ice surface-snow cover groove partitions. Step S33: Based on the irregular groove depth parameter data corresponding to different ice surface-snow cover groove partitions and combined with the corresponding relative thickening of the snow cover in the groove, construct the corresponding working depth adjustment model to determine the adjustment mathematical relationship between the working clearing depth and the groove depth and the relative thickening of the snow cover. Obtain the corresponding ice surface hardness parameter in the ice surface-snow cover groove partition. At the same time, set the corresponding working depth adjustment rate limit based on the ice surface hardness parameter, and impose adjustment limit constraints on the working depth adjustment model based on the working depth adjustment rate limit to avoid damage to the ice surface in the groove due to excessively rapid working depth adjustment. Calculate and output the corresponding depth adjustment benchmark parameter data. Step S34: Divide the snow removal area based on the depth adjustment reference parameter data of the ice surface-snow cover thickness distribution data, so as to merge the areas with similar depth adjustment reference parameter data into the same removal adjustment unit, so as to generate ice surface-snow cover adaptive removal area data, including the working depth adjustment range corresponding to each removal adjustment unit. Step S35: Based on the depth adjustment reference parameter data, pre-set the snow removal execution action for the ice surface-snow cover layer adaptive removal area data to obtain the pre-set execution data for the ice surface-snow cover layer removal area.

6. The method for rapid snow removal from ice surfaces with adaptive working depth adjustment according to claim 5, characterized in that, Step S32 includes the following steps: The depth parameter data of irregular grooves corresponding to different ice surface-snow cover layer groove partitions are deconstructed into groove topology to convert the corresponding irregular groove depth parameter data into three-dimensional surface coordinates. A polar coordinate system is established with the lowest point of the groove bottom as the origin. At the same time, the radial distribution of the depth value at different angles is calculated to obtain the radial distribution of the irregular groove depth corresponding to different ice surface-snow cover layer groove partitions. Based on the radial distribution of irregular groove depth corresponding to different ice surface-snow cover groove zones, obtain the edge inflection points and edge turning points of irregular grooves corresponding to different ice surface-snow cover groove zones; Based on the irregular groove edge inflection points and irregular groove edge turning points corresponding to different ice surface-snow cover groove partitions, the corresponding ice surface-snow cover groove partitions are divided into groove sub-regions to obtain the irregular groove snow cover sub-partitions corresponding to different ice surface-snow cover groove partitions. Based on the irregular groove sub-regions within the irregular grooves corresponding to different ice-snow-covered groove partitions, the depth of the irregular groove sub-regions and the increase in the difference in ice-snow-covered thickness are selected from the corresponding irregular groove depth parameter data and ice-snow-covered thickness gradient data. Then, based on the depth of the irregular groove sub-regions and the increase in the difference in ice-snow-covered thickness corresponding to the same sub-region, the relative thickening within the grooves is calculated cumulatively to obtain the relative thickening of the snow-covered layer within the grooves corresponding to different ice-snow-covered groove partitions.

7. The method for rapid snow removal from ice surfaces with adaptive working depth adjustment according to claim 5, characterized in that, Step S35 includes the following steps: Step S351: By extracting the spatial coordinates and working depth adjustment range of each clearing adjustment unit from the ice surface-snow cover adaptive clearing area data, and combining the groove spatial distribution corresponding to the ice surface-snow cover thickness distribution data, basic data for ice surface-snow cover clearing area path planning is generated. Step S352: Construct a path optimization model based on the basic data of the ice surface-snow cover removal area path planning, with the objective function being to maximize the removal coverage and minimize the path length. Introduce priority removal weight rules corresponding to the ice surface-snow cover grooves, including that the weight of deep grooves is higher than that of shallow grooves. At the same time, generate corresponding path optimization objective parameter data for the path optimization model based on the objective function and the priority removal weight rules. Step S353: The A* algorithm is used in conjunction with the path optimization target parameter data to solve the path optimization model, so as to obtain the optimal clearing trajectory from the starting point to the end point of the adaptive clearing area of ​​ice surface-snow cover, which includes the order and turning angle of passing through each clearing adjustment unit, so as to obtain the initial path planning data. Step S354: Perform a smoothness check on the initial path planning data to calculate the turning radius and curvature change rate corresponding to the optimal clearing trajectory. When the curvature change rate exceeds a preset threshold, insert the corresponding transition line segment to obtain the optimized path planning data. Step S355: Based on the corresponding trajectory nodes in the optimized path planning data and combined with the depth adjustment reference parameter data, determine the working clearing depth, tool angle and travel speed corresponding to each trajectory node, and obtain the preset data for the ice surface-snow layer clearing area.

8. The method for rapid snow removal from ice surfaces with adaptive working depth adjustment according to claim 1, characterized in that, Step S4 includes the following steps: Step S41: Based on the preset data for the ice-snow layer removal area, start the removal execution mechanism to perform the corresponding rapid ice-snow layer removal operation, and collect the corresponding ice-snow layer removal residue image data during the removal process through real-time feedback sensors; Step S42: Perform snow cover layer residual analysis on the ice surface-snow cover layer removal residual image data to identify the areas of snow accumulation that have not been removed from the ice surface-snow cover layer using image segmentation algorithms, and calculate the corresponding residual area and snow thickness based on the areas of snow accumulation that have not been removed from the ice surface-snow cover layer to obtain snow cover layer removal residual data. Step S43: Based on the snow cover layer removal residual data and combined with the ice surface-snow cover layer thickness correlation feature data, perform iterative optimization of the working depth adjustment model for ice surface-snow cover layer removal to output the corresponding rapid ice surface-snow cover layer removal results.

9. The method for rapid snow removal from ice surfaces with adaptive working depth adjustment according to claim 8, characterized in that, Step S43 includes the following steps: Based on the snow cover layer removal residual data and combined with the ice surface-snow cover layer thickness correlation feature data, the depth correction calculation is performed on the depth adjustment benchmark parameter data output by the working depth adjustment model to obtain the working depth adjustment correction amount data. The working depth adjustment correction data is input into the working depth adjustment model and the adjustment amount corresponding to the clearing actuator is calculated by the PID control algorithm. Based on the adjustment amount corresponding to the clearing actuator, the corresponding control signal is output to obtain the real-time control signal data for clearing execution. The real-time control signal data for clearing is sent to the corresponding drive module of the clearing actuator to drive the depth adjustment component to perform the corresponding depth adaptive adjustment, thereby obtaining real-time depth adjustment data; based on the real-time depth adjustment data, the precise clearing operation corresponding to the ice surface-snow layer clearing area is completed, and the ice surface-snow layer clearing area is simultaneously scanned a second time by a laser scanning device to obtain the secondary residual thickness of the snow layer and the area of ​​ice surface damage traces corresponding to the ice surface-snow layer clearing area, thereby obtaining the original data of the clearing effect; Obtain the original snow cover thickness distribution and the total area of ​​the ice surface corresponding to the ice-snow cover removal area. Based on the original snow cover thickness distribution and the total area of ​​the ice surface, calculate the removal effect of the original data to determine the snow residue rate and ice surface damage degree corresponding to the ice-snow cover removal area. The snow residue rate is the ratio of the secondary residual thickness of the snow cover to the original thickness distribution of the snow cover, while the ice surface damage degree is the ratio of the area of ​​ice surface damage traces to the total area of ​​the ice surface. This yields the removal effect evaluation data. The clearing effect evaluation data is compared with the preset effect threshold to filter out the sub-areas that did not meet the clearing standards for the ice-snow layer clearing zone. Based on the sub-areas that did not meet the clearing standards, the parameters of the working depth adjustment model are reversed to generate model optimization parameters. Based on the model optimization parameters, the corresponding working depth adjustment model is updated to perform iterative optimization of ice-snow layer clearing to output the corresponding rapid clearing results of ice-snow layer.

10. A rapid snow removal device for ice surfaces with adaptive working depth adjustment, characterized in that, The device includes a processor, a memory, and a computer program stored in the memory and executable on the processor, for performing the rapid snow removal method for adaptive working depth adjustment as described in any one of claims 1-9.

Images