Serialized adaptive path planning method for automatic cabin cleaning engineering equipment in finite dynamic complex environment

Abstract

The invention belongs to the technical field of port automation, and discloses a serialized adaptive path planning method for automatic cabin cleaning engineering equipment in a finite dynamic complex environment, and the method is based on a grid map constructed by high-precision laser SLAM and an evaluation-selection-simulation cycle, a path sequence can be adaptive to topographic changes, and the path sequence can be used for planning the path in a real-time manner. And local excessive excavation or missing excavation is avoided, and the working efficiency is improved. According to the method, the path with the maximum coal removal amount is preferentially selected through each round of decision of the greedy strategy, it is ensured that main material piles are rapidly processed in the initial stage of operation, and the overall cabin clearing time is optimized. The Bezier curve is adopted for path switching, the movement continuity and stability of the automatic cabin cleaning engineering equipment are guaranteed, and the safety requirement for operation in a narrow cabin is met.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of port automation, and particularly relates to a serialization adaptive path planning method of automatic cleaning engineering equipment in a limited dynamic complex environment, namely an adaptive serialization path planning method of automatic cleaning engineering equipment for cleaning operation, which can generate an optimal path sequence according to real-time terrain. BACKGROUND

[0002] In the sea transportation logistics of bulk cargoes (such as coal and ore), the cleaning link at the end of unloading operation is the key bottleneck determining the overall efficiency. At present, this link highly depends on manual driving of automatic cleaning engineering equipment (such as a pusher) to complete, and the operation environment is closed, the dust concentration is high, the safety risk is great, and the efficiency of manual operation is limited. Realizing the autonomy and intelligence of automatic cleaning engineering equipment is an urgent demand to improve the competitiveness of ports.

[0003] The core challenge of automatic cleaning engineering equipment lies in its path planning strategy. Unlike the operation environment of conventional mobile robots, cleaning operation faces a unique environment that is limited, dynamic and complex. The terrain of bulk cargo is dynamically changed due to continuous operation, and the space is narrow, which brings great difficulty to path planning. The path planning methods of existing engineering vehicles are mostly used in open space or static environment, and the goal is the shortest movement without collision, which does not match the essential goal of "efficiently reshaping the terrain through the path itself" in cleaning operation. In addition, simple fixed sequence or random operation strategy cannot adapt to the dynamic changes of terrain, and is easy to cause local overexcavation or low operation efficiency. Therefore, there is an urgent need in the field for a planning method that can adapt to limited narrow space, cope with dynamically changing terrain, and intelligently decide and serialize the operation path, so as to realize efficient and safe operation of automatic cleaning equipment. SUMMARY

[0004] The purpose of the present application is to provide a serialization adaptive path planning method of automatic cleaning engineering equipment in a limited dynamic complex environment, which can automatically generate a group of candidate paths of optimal direction according to different coal bulk terrain, and further dynamically determine the optimal operation path execution sequence through an iterative greedy selection strategy, so as to realize efficient and intelligent cleaning operation.

[0005] To achieve the above purpose, the technical solution adopted by the present application is as follows:

[0006] The serialization adaptive path planning method of automatic cleaning engineering equipment in a limited dynamic complex environment comprises:

[0007] A 2.5D grid map of the coal part in the cargo hold is obtained, a target flattening height is determined according to the coal volume, and the grid corresponding to the coal part in the cargo hold is divided into a coal raking area and a coal spreading area according to the target flattening height.

[0008] For each coal raking area, determine the optimal coal raking path direction, and obtain the total number of coal raking paths and the uniform interval of coal raking paths according to the optimal coal raking path direction;

[0009] Taking the geometric centroid of the coal raking area as the reference, based on the total number of coal raking paths and the uniform interval of coal raking paths, a set of seed points is generated on the vertical axis of the optimal coal raking path direction;

[0010] Each seed point in the set of seed points is extended to generate a candidate coal raking path group, and simulation is performed based on the candidate coal raking path group. Through the greedy strategy, the candidate coal raking path with the largest coal removal amount is selected and removed in each round, and the final coal raking path sequence after simulation is output, thereby obtaining the coal raking path;

[0011] For each coal spreading area, the end point of the corresponding coal raking path of the coal raking area is taken as the starting point of the coal spreading path, and the coal spreading path is generated;

[0012] Between the last coal spreading path and the next coal raking path, a switching path is generated by using a Bezier curve to plan, and the path planning is completed.

[0013] The following also provides several optional modes, but not as an additional limitation to the above overall scheme, but only as a further supplement or optimization, without technical or logical contradictions, each optional mode can be combined with the above overall scheme alone, and can also be combined between multiple optional modes.

[0014] As an optimization, the target flattening height is determined according to the coal volume, and the grid corresponding to the coal in the cargo hold is divided into a coal raking area and a coal spreading area according to the target flattening height, including:

[0015] The coal volume is calculated according to the height of the coal in the 2.5D grid map, and the target flattening height after the coal is flattened is calculated based on the principle of volume conservation;

[0016] Determine whether the coal height of each grid is higher than the target flattening height, and take the grid that is higher than or equal to the target flattening height and connected as a coal raking area, and take the grid that is lower than the target flattening height and connected as a coal spreading area, to complete the division of the work area.

[0017] As an optimization, the determination of the optimal coal raking path direction includes:

[0018] The average gradient of the coal raking area is calculated by using the finite difference method, and the gradient angle is obtained;

[0019] On the basis of the gradient angle, add radian, as the angle of the optimal coal raking path direction , and the unit vector of the optimal coal raking path direction is .

[0020] Preferably, the step of obtaining the total number of coal scraping paths and the uniform spacing of the coal scraping paths based on the optimal coal scraping path direction includes:

[0021] Project the width of each grid cell in the coal-scraping zone onto the vertical axis of the optimal coal-scraping path direction.

[0022] Take the smallest projected width among all grids, add half the width of the shovel blade of the automated cleaning equipment, and use it as the projected position of the starting point of the middle area of ​​the coal scraping zone.

[0023] Take the largest projected width among all grids, and subtract half the width of the shovel blade of the automated cleaning equipment, as the projected position of the end point of the middle area of ​​the coal scraping zone;

[0024] The width of the middle area of ​​the coal-scraping zone is obtained based on the projection positions of the termination point and the starting point.

[0025] Divide the width of the middle area of ​​the coal-scraping zone by the maximum path spacing and round up to get the number of paths in the middle area. Add one to the number of paths in the middle area to get the total number of coal-scraping paths in the coal-scraping zone.

[0026] The width of the middle area of ​​the coal-scraping zone is divided by the number of paths in the middle area to determine the uniform spacing of the coal-scraping paths.

[0027] Preferably, the seed point set is represented as follows:

[0028]

[0029] in, Represents the seed point set. seed points, , This represents the geometric centroid of the coal-sinking area. This indicates the projected position of the starting point of the middle area of ​​the coal-shoveling zone. Indicates the uniform spacing of the coal rake path. The vertical axis representing the direction of the optimal coal scraping path. This indicates the total number of coal scraping paths.

[0030] Preferably, each seed point in the extended seed point set generates a group of candidate coal-harrowing paths, including:

[0031] Starting from each seed point, extend bidirectionally along the direction of the optimal coal scraping path and in the opposite direction until the boundary of the coal scraping area is encountered, thus obtaining the preliminary path.

[0032] By extending the offset length of the rake shovel of the automated cleaning equipment relative to the center of the vehicle body at the end of the initial path, the candidate coal scraping path corresponding to the seed point is obtained. By combining the candidate coal scraping paths corresponding to all seed points, a candidate coal scraping path group is obtained.

[0033] Preferably, the simulation is based on a group of candidate coal-picking paths. A greedy strategy is used to select and remove the candidate coal-picking path with the largest coal quantity in each round of decision-making. The final coal-picking path sequence is output after the simulation, including:

[0034] The initial coal-plowing path sequence is empty; take the first... The 2.5D raster map from the next iteration is used as the current map;

[0035] Traverse all candidate coal scraping paths in the candidate coal scraping path group and calculate the amount of coal to be removed for each candidate coal scraping path based on the current map;

[0036] The candidate scraping path with the largest amount of coal to be removed is selected and added to the scraping path sequence. The automated hull cleaning equipment model is then invoked to perform scraping simulation based on the selected scraping path, and the 2.5D raster map is updated as the first... The current map in the next iteration;

[0037] Determine if the iteration termination condition is met. If not, continue iterating; otherwise, output the coal scraping path sequence.

[0038] Preferably, the step of generating the coal spreading path by using the end point of the corresponding coal-scraping path in the coal-scraping zone as the starting point of the coal-spreading path includes:

[0039] If the angle between the direction of the end point of the coal scraping path and the direction of the start point of the coal spreading path exceeds the maximum turning angle constraint of the automated cargo cleaning equipment, a smooth transition strategy is first adopted to turn, and then the end point of the coal scraping path in the corresponding coal scraping area is taken as the start point of the coal spreading path, extending along a direction parallel to the longitudinal axis of the cargo hold and pointing to the center of the cargo hold until the boundary of the coal spreading area to generate the coal spreading path.

[0040] Otherwise, the starting point of the coal spreading path is taken as the end point of the corresponding coal spreading area, and the spreading path is extended in a direction parallel to the longitudinal axis of the cargo hold and pointing to the center of the cargo hold until the boundary of the coal spreading area to generate the coal spreading path.

[0041] This invention provides a sequential adaptive path planning method for automated cargo tank cleaning equipment in finite dynamic complex environments, capable of generating an efficient and safe operational path sequence for the equipment. Compared to traditional fixed paths or manual operation, this method, based on a grid map constructed using high-precision SLAM and an "evaluation-selection-simulation" loop, allows the path sequence to adapt to terrain changes, avoiding local over-excavation or under-excavation and improving operational efficiency. The method employs a greedy strategy, prioritizing the path with the largest coal removal volume in each round of decision-making, ensuring rapid processing of the main material pile in the early stages of operation and optimizing the overall cargo tank cleaning time. The use of Bézier curves for path switching ensures the continuity and stability of the automated cargo tank cleaning equipment's movement, meeting the safety requirements for operations within confined cargo tanks. Attached Figure Description

[0042] Figure 1 This is a flowchart of the serialized adaptive path planning method for automated cabin cleaning engineering equipment under finite dynamic complex environment according to the present invention.

[0043] Figure 2 This is a schematic diagram of the coal scraping path planning of the present invention;

[0044] Figure 3 This is a flowchart of the greedy strategy rotation method of the present invention;

[0045] Figure 4 This is a schematic diagram of the coal spreading path and switching path planning of the present invention. Detailed Implementation

[0046] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0047] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to limit the invention.

[0048] To achieve automated cargo hold cleaning operations using automated cargo hold cleaning equipment, this embodiment provides a complete autonomous cargo hold cleaning method, from path generation to sequence planning. The cargo hold dimensions of the involved cargo ship are approximately 50m × 40m, with a hold height of approximately 35m. The automated cargo hold cleaning equipment is 4m long and 3m wide, with a blade width of 3m, a length of 0.5m, and a depth of 1m. The blade's offset from the ship's center is 2m. The maximum turning angle of the automated cargo hold cleaning equipment is 30°, and the minimum turning radius is 3m.

[0049] like Figure 1 As shown in this embodiment, the sequential adaptive path planning method for automated cabin cleaning equipment in a finite dynamic complex environment specifically includes the following steps:

[0050] Step 1: Obtain a 2.5D grid map of the coal section in the cargo hold. Determine the target leveling height based on the coal volume. Divide the grid corresponding to the coal section in the cargo hold into a coal-scraping area and a coal-spreading area based on the target leveling height.

[0051] The automated cargo hold cleaning equipment is activated. A laser SLAM mapping system scans the cargo hold to obtain a dense point cloud map. The point cloud is then processed to separate the coal portion of the cargo hold. A raster map is constructed, with a resolution set to [resolution value missing]. In this embodiment, we take The point cloud of the coal material is projected from three-dimensional space onto a two-dimensional grid plane, resulting in a grid of [number of grid cells]. The raster map obtained in this embodiment The terrain height information is preserved within each grid cell to obtain a 2.5D grid map of the coal section inside the cargo hold. .

[0052] Based on 2.5D raster map The coal in the hold is approximated as a structure composed of multiple cubes, and the volume of coal in the hold is estimated by summing the volumes of these cubes. When the coal in the hold reaches the target state after cleaning, the terrain inside the hold is approximately flat. Based on the principle of volume conservation, the target leveling height of the cargo hold is... It can be calculated using the following formula:

[0053] (1)

[0054] in, For the first raster map Line number The height of the column grid, The total area of ​​the map represents the effective flattening area of ​​the cargo hold. After obtaining the target flattening height, for each effective grid area on the map, the work area is divided according to the height of each grid and the coal height. Grids that are higher than or equal to the target flattening height and are connected are designated as a coal scraping area, and grids that are lower than the target flattening height and are connected are designated as a coal spreading area, thus completing the work area division. This embodiment calculates... Based on the adaptive region division rules, the 2.5D raster map is... The area was divided into 8 coal-scraping zones and 8 coal-spreading zones, with no overlap between the zones.

[0055] Step 2: For each coal-scraping zone, determine the optimal coal-scraping path direction, and obtain the total number of coal-scraping paths and the uniform spacing of the coal-scraping paths based on the optimal coal-scraping path direction.

[0056] like Figure 2 As shown, for each scraping zone, the average gradient of the working block is calculated using the finite difference method. Specifically, the gradient angle is calculated based on the height of each grid cell in the scraping zone using the following formula. :

[0057] (2)

[0058] in, For the first raster map Line number The height of the column grid, For the first raster map Line number The height of the column grid, For the first raster map Line number The height of the column grid, For the first raster map Line number The height of the column grid.

[0059] Add to the gradient angle radians, as the angle of the optimal coal scraping path direction. Therefore, the unit vector of the optimal coal scraping path direction is obtained. In this embodiment, substituting the data for the coal-plowing area, its area is calculated to be 69.0 m². 2 The gradient angle is Therefore, the optimal coal scraping path direction vector is .

[0060] After determining the optimal coal scraping path direction, all grid points within the coal scraping area are aligned with the vertical axis of the optimal coal scraping path direction. Projection is performed to calculate the width of the rake zone in that direction. .

[0061] (3)

[0062] (4)

[0063] in, This is the set of all grid points within the current coal-shoveling area. Indicates the first The projection width of each grid cell on the vertical axis These are the coordinates of each grid point. Indicates the maximum projection width. This represents the minimum projected width. This embodiment calculates the projected width of the work area. .

[0064] To ensure that the automated cleaning equipment can completely cover the entire area during reciprocating operations, a coverage overlap rate is set between adjacent coal scraping paths. To avoid untreated or missed areas, this invention employs a path centerline distribution strategy of "aligned ends and evenly distributed in the middle." This strategy first determines the scraping paths at both ends of the scraping area to ensure coverage of the edge of the area, and then evenly distributes the scraping paths in the middle. Specifically, this is based on the width of the shovel blade of the automated cleaning equipment. and overlap rate Calculate the theoretical maximum path spacing Then, based on the width of the intermediate coverage area... and Determine the total number of required coal scraping paths Ultimately, according to Calculate the actual uniform spacing The specific process is as follows:

[0065] Take the minimum projected width among all grid cells. Add half the width of the shovel blade of the automated cabin cleaning equipment The projected position of the starting point of the middle area of ​​the coal-shoveling zone .

[0066] Take the largest projected width among all grid cells. Subtract half the width of the shovel blade from the automated cabin cleaning equipment. The projected position of the end point of the middle area of ​​the coal-shoveling zone .

[0067] Based on the projection position of the termination point and the projection position of the starting point The width of the middle area of ​​the coal-plowing zone is obtained. .

[0068] The width of the middle area of ​​the coal rake zone Divide by the maximum path spacing Round up The number of paths in the intermediate area is incremented by one to obtain the total number of coal scraping paths in the coal scraping area. .

[0069] Take the width of the middle area of ​​the coal-plowing zone Divide by the number of paths in the middle area As the uniform spacing of the coal rake path .

[0070] Finally, this embodiment calculates the maximum path spacing. Total number of coal-picking paths required Total length of the effective coverage area The actual spacing between the coal scraper paths .

[0071] Step 3: Using the geometric centroid of the coal-scraping zone as a reference, and based on the total number of coal-scraping paths and the uniform spacing of the coal-scraping paths, generate a set of seed points on the vertical axis of the optimal coal-scraping path direction. The seed point set is represented as follows:

[0072] (5)

[0073] in, Represents the seed point set. seed points, , This represents the geometric centroid of the coal-sinking area. This indicates the projected position of the starting point of the middle area of ​​the coal-shoveling zone. Indicates the uniform spacing of the coal rake path. The vertical axis representing the direction of the optimal coal scraping path. This indicates the total number of coal scraping paths.

[0074] Step 4: Generate a candidate coal scraping path group for each seed point in the extended seed point set. Perform simulation based on the candidate coal scraping path group. In each round, use a greedy strategy to select and remove the candidate coal scraping path with the largest amount of coal. Output the final coal scraping path sequence after the simulation is completed, which is the coal scraping path.

[0075] From each seed point Starting point, the unit vector along the optimal coal scraping path direction. The path extends bidirectionally in the opposite direction until it reaches the boundary of the scraping area, thus forming a complete preliminary path that runs through the scraping area. Because the vehicle's positioning is offset from the shovel's position, and the scraping path is closely related to the shovel's position, the scraping path needs to be extended at the end during its generation. Ensure that the shovel's coverage area covers the entire coal-shoveling area, including This represents the offset length of the rake relative to the center of the vehicle. Finally, for each rake zone, a group of candidate rake paths is generated. .

[0076] like Figure 3 As shown, after generating the candidate coal scraping path group, at each decision time, the "job value" of all candidate paths is evaluated, and a greedy strategy is used to select the candidate coal scraping path that currently yields the maximum coal removal volume for execution. A candidate coal scraping path is defined. The "value of the task" Its coverage area The total volume of all coal pieces exceeding the target leveling height. In each iteration, the algorithm selects... Path to maximize value Add the coal rake path sequence.

[0077] (6)

[0078] (7)

[0079] in Indicates the first The next iteration of the grid map The Middle Line number The volume of coal within a grid column that is higher than the target paving height. Indicates the first The next iteration of the grid map Central path coverage area The total volume of coal material exceeding the target leveling height.

[0080] After each iteration of the selection process, the automated cleaning equipment's coal scraping model is invoked to simulate the operation on the grid map. Changes, updated raster map This will serve as the basis for the next iteration's decision. This closed-loop process of "evaluation-selection-simulation" is repeated until the termination condition is met. When setting the final condition, for each path's coverage area, it is required that the coal is approximately cleared, i.e., the coal quantity adjustment and removal amount is the initial set proportion. When the termination condition is met, the loop ends, returning to the finally generated, ordered coal-picking path sequence S. In this embodiment, the coal quantity adjustment and removal proportion is set to 90% of the initial state. The results of performing the above loop on the coal-picking area are shown in Table 1.

[0081] Table 1 Results of Cyclic Execution in the Coal Shovel Area

[0082]

[0083] Step 5: For each coal spreading area, take the end point of the corresponding coal spreading path of the coal spreading area as the starting point of the coal spreading path, and generate the coal spreading path.

[0084] like Figure 4 As shown, after the coal scraping path sequence in the coal scraping area is determined, the end point of the coal scraping path is used as the starting position for coal spreading. A reversing coal spreading trajectory is generated based on the kinematic characteristics and operational requirements of the automated cleaning equipment. First, for each coal scraping path, the position information of its end point is extracted. Then, when the angle between the direction of the end point of the coal scraping path and the direction of the starting point of the coal spreading path exceeds the maximum steering angle constraint of the automated cleaning equipment, a smooth transition strategy is first adopted.

[0085] (8)

[0086] in, The current orientation angle of the automated cabin cleaning equipment. The orientation angle of the automated cleaning equipment at the starting point of the target coal spreading path. The orientation angle of the automated cleaning equipment at the end of the coal scraping path. It is an S-shaped smooth interpolation function. After turning based on the smooth transition strategy, the end point of the corresponding coal-scraping path in the coal-scraping area is taken as the starting point of the coal-spreading path, which extends along a direction parallel to the longitudinal axis of the cargo hold and pointing towards the center of the cargo hold until the boundary of the coal-spreading area, thus generating the coal-spreading path.

[0087] If the angle does not exceed the maximum turning angle constraint of the automated cargo hold cleaning equipment, the coal spreading path is generated directly from the end point of the coal scraping path, extending along a direction parallel to the longitudinal axis of the cargo hold and pointing towards the center of the cargo hold until the boundary of the coal spreading area. The path points are distributed along the ideal direction with a fixed step size. In this embodiment, a total of 158 corresponding coal spreading paths are generated according to the coal scraping path sequence of the coal scraping area.

[0088] Step 6: Between the previous coal spreading path and the next coal scraping path, use Bézier curve planning to generate a switching path to complete the path planning.

[0089] The switching path connects the coal spreading path and the coal scraping path in sequence to achieve the operation cycle. This is achieved using cubic Bézier spline interpolation, taking into account the positional and directional constraints of the start and end points. Given a starting point... and the end point Based on the corresponding direction vectors, four control points are constructed:

[0090] (9)

[0091] (10)

[0092] in, To control the distance, take 1 / 3 of the distance between the starting and ending points of the switching path. and These are the unit direction vectors of the origin and destination, respectively. The path points are generated using the parametric equations of a cubic Bézier curve as follows:

[0093] (11)

[0094] in, These are parameter variables. The generated path points are verified through kinematic constraints to ensure that the turning angle of adjacent path segments does not exceed the maximum turning capacity of the automated cleaning equipment. If the kinematic constraint verification fails, i.e., the turning angle of adjacent path segments exceeds the maximum turning angle constraint of the automated cleaning equipment, the smooth transition strategy mentioned in step 5 is used for the turning transition to ensure the executability of the trajectory. Finally, this embodiment obtains a complete operation path sequence for the coal scraping area and the coal spreading area, including 158 coal scraping paths, 158 coal spreading paths, and 157 switching paths.

[0095] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0096] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the appended claims.

Claims

1. A sequential adaptive path planning method for automated cabin cleaning equipment in finite dynamic complex environments, characterized in that, The sequential adaptive path planning method for automated cabin cleaning equipment under finite dynamic complex environments includes: Obtain a 2.5D grid map of the coal section in the cargo hold, determine the target leveling height based on the coal volume, and divide the grid corresponding to the coal section in the cargo hold into a coal scraping area and a coal spreading area based on the target leveling height. For each coal-scraping zone, determine the optimal coal-scraping path direction, and obtain the total number of coal-scraping paths and the uniform spacing of the coal-scraping paths based on the optimal coal-scraping path direction. Based on the geometric centroid of the coal-scraping area, and taking into account the total number of coal-scraping paths and the uniform spacing of the coal-scraping paths, a set of seed points is generated on the vertical axis of the optimal coal-scraping path direction. Each seed point in the extended seed point set generates a candidate coal scraping path group. Simulation is performed based on the candidate coal scraping path group. In each round, a greedy strategy is used to select and remove the candidate coal scraping path with the largest amount of coal. The final coal scraping path sequence after the simulation is completed is output to obtain the coal scraping path. For each coal spreading area, the coal spreading path is generated by taking the end point of the corresponding coal spreading path of the coal spreading area as the starting point of the coal spreading path. Between the previous coal spreading path and the next coal scraping path, a switching path is generated using Bézier curve planning to complete the path planning.

2. The sequential adaptive path planning method for automated cabin cleaning equipment under finite dynamic complex environments according to claim 1, characterized in that, The target leveling height is determined based on the coal volume. Based on this target leveling height, the corresponding grid within the cargo hold is divided into a scraping zone and a spreading zone, including: The volume of coal is calculated based on the height of the coal in the 2.5D grid map, and the target leveling height of the coal after leveling is calculated based on the principle of volume conservation. Determine whether the coal height of each grid is higher than the target leveling height. Grids that are higher than or equal to the target leveling height and are connected are designated as a coal-scraping area, and grids that are lower than the target leveling height and are connected are designated as a coal-spreading area, thus completing the division of the work area.

3. The sequential adaptive path planning method for automated cabin cleaning equipment under finite dynamic complex environments according to claim 1, characterized in that, Determining the optimal coal scraping path direction includes: The average gradient of the coal-hung zone was calculated and the gradient angle was obtained using the finite difference method. Add to the gradient angle radians, as the angle of the optimal coal scraping path direction. At the same time, the unit vector of the optimal coal scraping path direction is obtained as follows: .

4. The sequential adaptive path planning method for automated cabin cleaning equipment under finite dynamic complex environments according to claim 1, characterized in that, The process of obtaining the total number of coal scraping paths and the uniform spacing of the coal scraping paths based on the optimal coal scraping path direction includes: Project the width of each grid cell in the coal-scraping zone onto the vertical axis of the optimal coal-scraping path direction. Take the smallest projected width among all grids, add half the width of the shovel blade of the automated cleaning equipment, and use it as the projected position of the starting point of the middle area of ​​the coal scraping zone. Take the largest projected width among all grids, and subtract half the width of the shovel blade of the automated cleaning equipment, as the projected position of the end point of the middle area of ​​the coal scraping zone; The width of the middle area of ​​the coal-scraping zone is obtained based on the projection positions of the termination point and the starting point. Divide the width of the middle area of ​​the coal-scraping zone by the maximum path spacing and round up to get the number of paths in the middle area. Add one to the number of paths in the middle area to get the total number of coal-scraping paths in the coal-scraping zone. The width of the middle area of ​​the coal-scraping zone is divided by the number of paths in the middle area to determine the uniform spacing of the coal-scraping paths.

5. The sequential adaptive path planning method for automated cabin cleaning equipment under finite dynamic complex environments according to claim 1, characterized in that, The seed point set is represented as follows: in, Represents the seed point set. seed points, , This represents the geometric centroid of the coal-sinking area. This indicates the projected position of the starting point of the middle area of ​​the coal-shoveling zone. Indicates the uniform spacing of the coal rake path. The vertical axis representing the direction of the optimal coal scraping path. This indicates the total number of coal scraping paths.

6. The sequential adaptive path planning method for automated cabin cleaning equipment under finite dynamic complex environments according to claim 1, characterized in that, Each seed point in the extended seed point set generates a candidate coal scraping path group, including: Starting from each seed point, extend bidirectionally along the direction of the optimal coal scraping path and in the opposite direction until the boundary of the coal scraping area is encountered, thus obtaining the preliminary path. By extending the offset length of the rake shovel of the automated cleaning equipment relative to the center of the vehicle body at the end of the initial path, the candidate coal scraping path corresponding to the seed point is obtained. By combining the candidate coal scraping paths corresponding to all seed points, a candidate coal scraping path group is obtained.

7. The sequential adaptive path planning method for automated cabin cleaning equipment under finite dynamic complex environments according to claim 1, characterized in that, The simulation based on candidate coal-picking paths uses a greedy strategy to select and remove the candidate coal-picking path with the largest coal quantity in each round of decision-making. The final coal-picking path sequence is output after the simulation, including: The initial coal-plowing path sequence is empty; take the first... The 2.5D raster map from the next iteration is used as the current map; Traverse all candidate coal scraping paths in the candidate coal scraping path group and calculate the amount of coal to be removed for each candidate coal scraping path based on the current map; The candidate scraping path with the largest amount of coal to be removed is selected and added to the scraping path sequence. The automated hull cleaning equipment model is then invoked to perform scraping simulation based on the selected scraping path, and the 2.5D raster map is updated as the first... The current map in the next iteration; Determine if the iteration termination condition is met. If not, continue iterating; otherwise, output the coal scraping path sequence.

8. The sequential adaptive path planning method for automated cabin cleaning equipment under finite dynamic complex environments according to claim 1, characterized in that, The step of generating a coal spreading path by using the end point of the corresponding coal spreading path in the coal spreading area as the starting point of the coal spreading path includes: If the angle between the direction of the end point of the coal scraping path and the direction of the start point of the coal spreading path exceeds the maximum turning angle constraint of the automated cargo cleaning equipment, a smooth transition strategy is first adopted to turn, and then the end point of the coal scraping path in the corresponding coal scraping area is taken as the start point of the coal spreading path, extending along a direction parallel to the longitudinal axis of the cargo hold and pointing to the center of the cargo hold until the boundary of the coal spreading area to generate the coal spreading path. Otherwise, the starting point of the coal spreading path is taken as the end point of the corresponding coal spreading area, and the spreading path is extended in a direction parallel to the longitudinal axis of the cargo hold and pointing to the center of the cargo hold until the boundary of the coal spreading area to generate the coal spreading path.

Images