A path planning method and system applied to a full-automatic cargo cleaning machine

Abstract

The present application relates to the field of wharf cleaning, especially to a path planning method and system applied to a full-automatic cleaning machine, the method comprising: obtaining environment perception data and equipment parameter data of a material taking point in a ship cabin, intelligently grouping and priority sorting, and constructing multiple efficient material taking sequences. Considering comprehensively the material volume of the material taking point, the distance between the material taking points, the position of the material stacking point, and the moving cost and obstacle avoidance demand, the current material taking cost of the possible displacement point is calculated; according to the possible displacement point, the residual material taking cost possibly existing in the future path is analyzed, the current material taking cost and the residual material taking cost are added to obtain the comprehensive material taking cost. The displacement point with the minimum comprehensive material taking cost is selected through iteration, and the material taking path is optimized. Through intelligent grouping, priority sorting and path optimization, the present application significantly improves the operation efficiency of the cleaning machine, reduces the energy consumption, and enhances the obstacle avoidance capability, ensuring safe and efficient completion of material cleaning in a complex environment.

Description

Technical Field

[0001] This invention relates to the field of dock cleaning. In particular, it relates to a path planning method and system for use in fully automated dock cleaning machines. Background Technology

[0002] The path planning technology for fully automated cargo hold cleaning machines at ports emerged driven by both the transformation towards port automation and the complex operational environment, representing a cutting-edge field integrating multidisciplinary technologies. Traditional bulk carrier hold cleaning relies on manual operation, facing harsh conditions such as high dust concentrations, variable lighting, and confined spaces, resulting in low efficiency and extremely high safety risks. With the global trade recovery, the demand for automation upgrades at China's coastal ports is projected to increase by 65% ​​year-on-year by 2025. Data from the International Association of Ports and Harbors shows that 87% of the world's top ports have already initiated robotic cargo hold cleaning upgrades, making path planning technology a core lever for overcoming efficiency bottlenecks.

[0003] In the automated cargo hold cleaning process, the path planning of the cleaning machine is of paramount importance. As a key piece of equipment for material handling within the ship's hold, the rationality of its path planning directly determines the efficiency and cost of material handling. Efficient path planning can significantly improve the working efficiency of the cleaning machine, reduce the time and energy consumption of material handling within the ship's hold, thereby optimizing the entire logistics and port clearance process and bringing significant economic benefits to enterprises.

[0004] However, in actual shipboard cleaning machine path planning, due to the dispersed distribution and significant differences in size of material piles within the ship's hold, multiple stacking points are typically required during operation. Existing technologies largely rely on the principle of selecting the nearest retrieving point from each stacking point, neglecting the crucial impact of different retrieving sequences and the cleaning machine's turning radius on the retrieving path planning. Therefore, in practice, the planned path often fails to adequately consider global optimality, resulting in low actual efficiency for the cleaning machine during retrieving and stacking, and increasing unnecessary working time and energy consumption. Summary of the Invention

[0005] To address the problem of low efficiency and high energy consumption caused by the existing cleaning machine's path planning neglecting the material picking sequence and relying solely on picking up the nearest material, this invention provides solutions in the following aspects.

[0006] In the first aspect, a path planning method for a fully automatic tank cleaning machine includes: acquiring environmental perception data and equipment parameter data for each material collection point in the tank to be cleaned; grouping the material collection points according to the material volume, calculating the material collection group selection score based on the material collection efficiency between each material collection point in the group and the distance between each material collection point, and identifying the optimal combination; constructing multiple material collection sequence sequences for the optimal combination and performing path planning, analyzing the possible displacement points and cost weights of the possible displacement points in the next step of the path planning, and obtaining the current material collection cost of the possible displacement points by combining movement cost and obstacle avoidance requirements; analyzing the remaining material collection costs that may exist in the future path based on the possible displacement points, adding the current material collection cost and the remaining material collection cost to obtain the comprehensive material collection cost, selecting the possible displacement point with the minimum comprehensive material collection cost as the initial point for the next search, and iterating until the material collection point in the group is reached to obtain the optimal material collection path for each material collection sequence sequence; constructing an objective function for path planning based on the optimal material collection path, selecting the material collection sequence with the minimum objective function value, and inputting it into the control system of the tank cleaning machine to guide the cleaning work of materials in the tank.

[0007] By grouping material collection points according to material volume, calculating the selection score of the collection group, and identifying the optimal combination, this method can optimize the material collection sequence of the cleaning machine and improve its operating efficiency. After constructing the material collection sequence, further precise path planning is performed, the cost weight of possible displacement points is analyzed, and the current material collection cost is obtained by combining movement cost and obstacle avoidance requirements. This helps the cleaning machine to perform its tasks safely and efficiently in complex ship cabin environments, while reducing the risks caused by collisions or misoperations. By comprehensively considering the current material collection cost and the remaining material collection cost, the displacement point with the minimum comprehensive material collection cost is selected as the starting point for the next search. This method not only improves the operating efficiency of the cleaning machine but also effectively reduces operating costs.

[0008] Preferably, the material selection group's scoring includes:

[0009] Select a preset number of material collection points to form a material collection group, and calculate the ratio between the average material volume of all material collection points in the material collection group and the single material collection amount of the cleaning machine to obtain the material collection efficiency. Round up the result and take the difference between the rounded result and the material collection efficiency as the material collection group efficiency deviation.

[0010] Calculate the average Euclidean distance between any two picking points within the picking group. Apply a negative exponential mapping to the product of the efficiency deviation of the picking group and the average Euclidean distance to obtain the score contribution of each picking group. The average of the accumulated score contributions of all picking groups is used as the selection score for the picking group.

[0011] By calculating the ratio of the average material volume within the material retrieving group to the single material retrieving amount of the cleaning machine and rounding it up, the material retrieving efficiency is determined, and the efficiency deviation of the material retrieving group is obtained. At the same time, considering the Euclidean distance between material retrieving points, the score contribution of each material retrieving group is calculated through negative exponential mapping. Finally, the selection score of the material retrieving group is accumulated to optimize the material retrieving sequence and operation path of the cleaning machine, improve material retrieving efficiency, reduce energy consumption and costs, enhance operational safety, and improve operational flexibility, thereby bringing significant economic benefits and operational performance improvements to the fields of automated logistics and port cleaning.

[0012] Preferably, the material taking sequence includes the priority of the cleaning machine accessing each material taking group, with the priority decreasing from left to right.

[0013] Preferably, the calculation method for the cost weight includes:

[0014] Using any material taking sequence as the target sequence, the material taking point of any material taking group in the target sequence as the target material taking point, the grid where the cleaning machine center is located as the target grid, and the neighboring grids of the target grid as possible displacement points, calculate the ratio of the sum of the diagonal radius of the first moving frame of the cleaning machine and the diagonal radius of the nearest non-target material taking point's material taking frame to the Euclidean distance between the possible displacement point and the nearest non-target material taking point of the possible displacement point, and normalize it to obtain the cost weight of obstacle avoidance cost.

[0015] By calculating the Euclidean distance between the possible displacement point of the cleaning machine and the nearest non-target material collection point, and combining the size information of the cleaning machine and its material collection frame, the obstacle avoidance cost is quantified and normalized to obtain a cost weight that reflects the obstacle avoidance risk. This improves the safety and efficiency of the cleaning machine in complex ship cabin environments, optimizes path planning, reduces unnecessary energy consumption, and enhances the adaptability to dynamic environmental changes, ultimately achieving more precise and efficient material cleaning operations.

[0016] Preferably, the calculation method for the current material extraction cost includes:

[0017] Taking the picking point of any picking group in the target sequence as the target picking point, calculate the average Euclidean distance between the possible displacement point and all target picking points. The average value of the Euclidean distance Divide by the diagonal diameter of the XOY coordinate plane of the cabin to obtain the distance cost;

[0018] Calculate the average Euclidean distance between the possible displacement point and all non-target pick-up points. Divide the diagonal diameter of the ship's cabin in the XOY coordinate plane by the average value of the Euclidean distance. Obstacle avoidance comes at a cost;

[0019] The current material collection cost of a possible displacement point is obtained by weighting and summing the cost weights as the weights of obstacle avoidance cost and distance cost (1 minus the cost weight).

[0020] Preferably, the calculation method for the remaining material extraction cost includes:

[0021] Determine the number of path points on the shortest path between the possible displacement point and the target material collection point, calculate the Euclidean distance between the path points on the shortest path and the nearest non-target material collection point, and use the sum of the reciprocals of the Euclidean distances of all path points as the remaining material collection cost.

[0022] Preferably, the objective function is calculated in the following ways:

[0023] Using any point on the optimal material picking path in the material picking sequence as a marker, calculate the comprehensive material picking cost of the marker, and sum the comprehensive material picking costs of all path points to obtain the objective function.

[0024] Preferably, the environmental perception data includes: the location of the material picking point, the material picking frame, and the material volume; the equipment parameter data includes: the location of the material stacking point, the moving frame of the cleaning machine, and the single material picking amount; and an XOY coordinate plane is established in the hull to be cleaned, and the coordinate plane is gridded for positioning and volume measurement.

[0025] Preferably, the straight-line distance between each material taking group and different material stacking points in the material taking sequence is calculated, the total distance between different material taking groups and material stacking points is evaluated, and the material stacking point with the smallest total distance is selected as the optimal material stacking point for the material taking group.

[0026] Secondly, a path planning system for a fully automated cabin cleaning machine includes a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned path planning method for a fully automated cabin cleaning machine is implemented.

[0027] The present invention has the following effects:

[0028] 1. This invention intelligently groups and prioritizes the material collection points by comprehensively considering the material volume at the collection points, the distance between the collection points, and the location of the stacking points, thereby constructing multiple efficient material collection sequence sequences. This can significantly improve the operating efficiency of the cleaning machine and reduce time waste caused by ineffective movement or repeated paths.

[0029] 2. This invention optimizes the material handling path, reduces the moving distance and number of times the cleaning machine travels, and effectively reduces the energy consumption of the cleaning machine during operation. By accurately calculating the current material handling cost and the remaining material handling cost, the material handling strategy is further optimized, thereby achieving energy conservation while ensuring operational efficiency.

[0030] 3. This invention incorporates obstacle avoidance cost calculation into path planning, enabling the cleaning machine to better avoid obstacles during operation, reducing collision risks and improving operational safety. By considering the cost weights of possible displacement points, including distance cost and obstacle avoidance cost, the cleaning machine can select the optimal path, ensuring safe operation in complex ship cabin environments. Attached Figure Description

[0031] Figure 1 This is a flowchart of steps S1-S5 in a path planning method for a fully automatic cabin cleaning machine according to an embodiment of the present invention.

[0032] Figure 2 This is a 3D scan of the cleaning machine's operating area in a path planning method for a fully automated cleaning machine, according to an embodiment of the present invention.

[0033] Figure 3 This is a schematic diagram of the material picking and stacking path planning of a fully automatic cleaning machine in an embodiment of the present invention.

[0034] Figure 4 This is a structural block diagram of a path planning system applied to a fully automatic cabin cleaning machine according to an embodiment of the present invention. Detailed Implementation

[0035] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments.

[0036] Reference Figure 1 A path planning method for a fully automated cabin cleaning machine includes steps S1-S5, as detailed below:

[0037] S1: Acquire environmental perception data and equipment parameter data for each material collection point in the ship's hold to be cleaned.

[0038] The environmental perception data includes: the location of the material collection point, the material collection frame and the material volume; the equipment parameter data includes: the location of the material stacking point, the moving frame of the cleaning machine and the amount of material collected in a single operation.

[0039] It should be noted that the environment setup process involves: establishing an XOY coordinate plane within the ship's hold and meshing this plane for positioning and volume measurement; using 3D (Three-Dimensional) radar technology and BeiDou positioning base stations to acquire the location of each material picking point, the location of each material stacking point, the material picking frame, and the material volume; determining the number of material picking points based on the actual material accumulation within the ship's hold, and constructing the material picking frame for each point, using the smallest rectangular frame of the material stack at each picking point as the picking frame; setting the movement frame and single-time material picking volume for the cleaning machine, where the movement frame includes a first movement frame and a second movement frame to determine the cleaning machine's range of movement within the ship's hold, and setting the single-time material picking volume based on the cleaning machine's bucket capacity. For example, the number and location of the material stacking points are pre-set according to operational requirements, typically 2-3 material stacking points.

[0040] Reference Figure 2 The diagram uses two green boxes to identify the key components of the cleaning machine: the first moving box represents the bucket, the component that performs material handling and transport; the second moving box represents the cockpit, the operation and control center of the cleaning machine, which may contain the operator or automated control system. This helps to clearly distinguish and consider the functions and range of motion of each part of the cleaning machine when planning routes and scheduling operations, ensuring the accuracy and safety of the operation.

[0041] In a ship's cabin environment, there are multiple material stacking and picking points, and the material volume at each picking point varies. Traditional algorithms find the path with the lowest total cost by traversing all grids and evaluating movement and obstacle avoidance costs, but they cannot simultaneously optimize obstacle avoidance performance and distance costs. Furthermore, when selecting the next grid, traditional algorithms only consider the current picking cost without fully considering subsequent picking costs, leading to an inability to make optimal choices when faced with equivalent grids, thus impacting overall planning efficiency. The specific steps are as follows:

[0042] S2: Group the material collection points according to the material volume, calculate the material collection group selection score based on the material collection efficiency between each material collection point in the group and the distance between each material collection point, and identify the optimal combination.

[0043] The material collection group selection evaluation includes:

[0044] Select a preset number of material collection points to form a material collection group, and calculate the ratio between the average material volume of all material collection points in the material collection group and the single material collection amount of the cleaning machine to obtain the material collection efficiency. Round up the result and take the difference between the rounded result and the material collection efficiency as the material collection group efficiency deviation.

[0045] Calculate the average Euclidean distance between any two picking points within the picking group. Apply a negative exponential mapping to the product of the efficiency deviation of the picking group and the average Euclidean distance to obtain the score contribution of each picking group. The average of the accumulated score contributions of all picking groups is taken as the selection score of the picking group. The picking group with the highest selection score is taken as the optimal combination.

[0046] Specifically, the selection score for the material taking group satisfies the following relationship:

[0047] ;

[0048] In the formula, This indicates the score for the material selection group. This indicates the number of material collection groups selected. Indicates the first The ratio of the average material volume of each material collection group to the material collection volume of the cleaning machine in a single operation. Indicates rounding up. Indicates the first The average value of the Euclidean distance between any two picking points within a picking group.

[0049] In other words, because the cleaning machine relies on the control principle of taking material from one point and returning it once, this principle is maintained even if the material volume is much smaller than the single taking amount. Therefore, the closer the single taking amount is to the single taking amount of the cleaning machine, the higher the taking efficiency of that time. This reflects the degree of matching between the total material volume within the reclaiming group and the maximum reclaiming capacity of the cleaning machine in a single operation. Specifically, it indicates the additional reclaiming operations the cleaning machine needs to perform to empty all material from the reclaiming group. A smaller difference indicates that the total material volume within the reclaiming group is closer to an integer multiple of the cleaning machine's single reclaiming capacity; a difference close to 0 indicates a higher contribution to the score. Conversely, when the total material volume is significantly less than or significantly greater than the single reclaiming capacity, the difference is larger, resulting in a lower contribution to the score. Multiplied by This is to avoid grouping material collection points that are far apart into the same group, as greater distances can actually lead to lower material collection efficiency.

[0050] This invention combines multiple material collection points into a single collection group. The cleaning machine continuously collects material within the group until the cumulative material collection reaches or approaches its maximum single collection capacity before returning. This ensures that the total material volume of each collection group is approximately an integer multiple of the cleaning machine's single collection capacity, thereby reducing the number of round trips and improving material collection efficiency.

[0051] Furthermore, the distance between material collection points was considered to avoid grouping distant collection points into the same collection group. This not only optimized the cleaning machine's operating path but also further reduced energy consumption and improved overall operating efficiency. By using group collection and path optimization methods, the cleaning machine's working efficiency and energy utilization efficiency when handling material piles of different volumes were significantly improved.

[0052] By comprehensively considering the material volume at the picking point, the relative position between the picking point and the stacking point, and the Euclidean distance between the grids, the operating efficiency of the cleaning machine is not only improved, but also the energy consumption caused by improper planning is reduced, which meets the needs of automated logistics and port cleaning.

[0053] S3: Construct multiple material picking sequence sequences for the optimal combination, perform path planning, analyze the possible displacement points and cost weights of the possible displacement points in the path planning, and obtain the current material picking cost of the possible displacement points by combining movement cost and obstacle avoidance requirements.

[0054] The material handling sequence of the cleaning machine includes the priority of the cleaning machine accessing each material handling group, with the priority decreasing from left to right.

[0055] The steps to obtain the optimal stockpile point include:

[0056] Determine the location of different material stacking points, calculate the straight-line distance between each material taking group and different material stacking points in the material taking sequence of the cleaning machine, evaluate the total distance between different material taking groups and material stacking points, and select the material stacking point with the smallest total distance as the optimal material stacking point for the material taking group.

[0057] By comprehensively considering the coordinates of each material handling group, the volume of the material, and the layout of the stacking points, the operation process of the cleaning machine is optimized to achieve efficient and orderly material cleaning. This not only improves the operating efficiency of the cleaning machine but also helps to reduce energy consumption and enhance overall operating performance.

[0058] The calculation methods for cost weights include:

[0059] Using any material taking sequence as the target sequence, the material taking point of any material taking group in the target sequence as the target material taking point, the grid where the cleaning machine center is located as the target grid, and the neighboring grids of the target grid as possible displacement points, calculate the ratio of the sum of the diagonal radius of the first moving frame of the cleaning machine and the diagonal radius of the nearest non-target material taking point's material taking frame to the Euclidean distance between the possible displacement point and the nearest non-target material taking point of the possible displacement point, and normalize it to obtain the cost weight of obstacle avoidance cost.

[0060] Specifically, the cost weights satisfy the following relationship:

[0061] ;

[0062] In the formula, The cost weight represents the cost of obstacle avoidance in the target mesh. This represents the Euclidean distance between a possible displacement point and its nearest non-target pick-up point. This represents the diagonal radius of the first moving frame of the cleaning machine and the diagonal radius of the picking frame nearest to the non-target picking point.

[0063] It should be noted that the cost weight reflects the degree of obstacle avoidance risk for the cleaning machine at potential displacement points. A higher cost weight indicates a higher obstacle avoidance risk and correspondingly higher obstacle avoidance cost. When the Euclidean distance between the potential displacement point and the nearest non-target material handling point is less than or equal to the sum of the diagonal radius of the cleaning machine's first moving frame and the diagonal radius of the nearest non-target material handling point's handling frame, the importance of obstacle avoidance cost is emphasized; conversely, the focus on obstacle avoidance cost is reduced. The cost weight is used in subsequent path planning optimization to ensure that the cleaning machine can effectively avoid obstacles during task execution, improving operational safety and efficiency.

[0064] The current methods for calculating material acquisition costs include:

[0065] Take the material collection point of any material collection group in the target sequence as the target material collection point (there may be one or more), calculate the average Euclidean distance between the possible displacement point and all target material collection points, divide the average Euclidean distance by the diagonal diameter of the ship's XOY coordinate plane to obtain the distance cost;

[0066] Calculate the average Euclidean distance between the possible displacement point and all non-target material collection points, and divide the diagonal diameter of the ship's XOY coordinate plane by the average Euclidean distance to obtain the obstacle avoidance cost.

[0067] The current material collection cost of a possible displacement point is obtained by weighting and summing the cost weights as the weights of obstacle avoidance cost and distance cost (1 minus the cost weight).

[0068] Specifically, the current material acquisition cost satisfies the following relationship:

[0069] ;

[0070] In the formula, This represents the current material cost at a possible displacement point. This represents the average Euclidean distance between the possible displacement point and all target material collection points. This represents the average Euclidean distance between the possible displacement point and all non-target material receiving points. This represents the diagonal diameter of the XOY coordinate plane of the ship's cabin. The cost weight represents the cost of obstacle avoidance at possible displacement points.

[0071] In other words, This represents the ratio of the average distance traveled from the possible displacement point to the target material collection point to the maximum diagonal distance of the ship's hold, reflecting the distance cost. This represents the ratio of the maximum diagonal distance of the ship's compartment to the average distance to non-target material collection points, reflecting the cost of obstacle avoidance.

[0072] To further explain, existing global path planning algorithms tend to traverse all grids within the ship's cabin network and evaluate them based on distance cost, then find a path with the minimum total cost based on the traversed grids. However, distance cost does not consider the influence of obstacles; a low distance cost does not necessarily guarantee good obstacle avoidance, and good obstacle avoidance does not necessarily result in a low distance cost. Therefore, path planning is performed by combining both distance cost and obstacle avoidance cost. The specific steps are as follows:

[0073] S4: Analyze the remaining material collection costs that may exist in the future path based on the possible displacement points, add the current material collection cost to the remaining material collection cost to obtain the comprehensive material collection cost, select the possible displacement point with the minimum comprehensive material collection cost as the starting point for the next search, and iterate until the material collection point in the material collection group is reached to obtain the optimal material collection path for each material collection sequence.

[0074] The calculation methods for the remaining material recovery cost include:

[0075] Determine the number of path points on the shortest path between the possible displacement point and the target material collection point, calculate the Euclidean distance between the path points on the shortest path and the nearest non-target material collection point, and use the sum of the reciprocals of the Euclidean distances of all path points as the remaining material collection cost.

[0076] Specifically, the remaining material extraction cost satisfies the following relationship:

[0077] ;

[0078] in, This represents the remaining material cost at the possible displacement point. This indicates the number of path points on the shortest path between the possible displacement point and the target material pick-up point. Indicates the first The Euclidean distance between each path point and its nearest non-target material pick-up point.

[0079] In other words, the remaining material collection cost reflects the contribution of each point on the path to the material collection cost. The closer a path point is to the material collection point, the larger its reciprocal is, and the greater its contribution to the total cost.

[0080] Reference Figure 3 The diagram illustrates a simulated automated material handling and stacking scenario, where a robot retrieves materials from a specific area and moves them to another area for stacking. In the diagram, 1 represents a stacking point, and 2, 3, 4, and 5 represent material handling groups. The yellow vehicle in the diagram is a cleaning machine. By combining a calculation method for selecting material handling groups, the system identifies the optimal material handling combination by evaluating the material handling efficiency and distance between handling points. This improves operational efficiency and reduces the robot's travel distance and time. It helps improve the overall performance of automated logistics and warehouse management, reduces energy consumption, and enhances operational safety and accuracy.

[0081] S5: Construct the objective function of path planning based on the optimal material picking path, select the material picking sequence with the smallest objective function value, and input it into the control system of the cleaning machine to guide the cleaning work of materials in the ship's hold.

[0082] The objective function can be calculated in the following ways:

[0083] Using any point on the optimal material picking path in the material picking sequence as a marker, calculate the comprehensive material picking cost of the marker, and sum the comprehensive material picking costs of all path points to obtain the objective function.

[0084] Specifically, the objective function satisfies the following relationship:

[0085] ;

[0086] In the formula, Describe the objective function. This represents the total number of path points included in the optimal material handling path across all material handling processes in the material handling sequence. Indicates the first The overall material collection cost at each path point This represents the minimum value function.

[0087] By iterating and evaluating the cost of all possible path points, a path is found that minimizes the total cost throughout the entire process from the starting point to the end point. The path point that minimizes the overall cost is selected, thereby optimizing the cleaning machine's operational efficiency, reducing energy consumption, and potentially improving operational safety.

[0088] This invention also provides a path planning system for a fully automated cabin cleaning machine. For example... Figure 4 As shown, the system includes a processor and a memory. The memory stores computer program instructions, which, when executed by the processor, implement a path planning method for a fully automated cabin cleaning machine according to the first aspect of the present invention. The system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface, the settings and functions of which are known in the art and will not be described further here.

[0089] It should be noted that those skilled in the art can make various modifications and improvements without departing from the inventive concept, and these all fall within the scope of protection of this invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A path planning method applied to a fully automated cabin cleaning machine, characterized in that, include: Acquire environmental perception data and equipment parameter data for each material collection point in the ship's hold to be cleaned; The material collection points are grouped according to the material volume. The material collection group selection score is calculated based on the material collection efficiency between each material collection point in the group and the distance between each material collection point to identify the optimal combination. Multiple material picking sequence sequences are constructed for the optimal combination, and path planning is performed. The possible displacement points of the next step in the path planning and the cost weight of the possible displacement points are analyzed. The current material picking cost of the possible displacement points is obtained by combining the movement cost and obstacle avoidance requirements. Based on the possible displacement points, the remaining material collection cost in the future path is analyzed. The current material collection cost is added to the remaining material collection cost to obtain the comprehensive material collection cost. The possible displacement point with the minimum comprehensive material collection cost is selected as the starting point for the next search. The search is iterated until the material collection point in the material collection group is reached, and the optimal material collection path for each material collection sequence is obtained. The objective function of path planning is constructed based on the optimal material picking path, and the material picking sequence with the smallest objective function value is selected and input into the control system of the cleaning machine to guide the cleaning work of materials in the ship's hold.

2. The path planning method for a fully automated cabin cleaning machine according to claim 1, characterized in that, The material collection group selection score includes: Select a preset number of material collection points to form a material collection group, and calculate the ratio between the average material volume of all material collection points in the material collection group and the single material collection amount of the cleaning machine to obtain the material collection efficiency. Round up the result and take the difference between the rounded result and the material collection efficiency as the material collection group efficiency deviation. Calculate the average Euclidean distance between any two picking points within the picking group. Apply a negative exponential mapping to the product of the efficiency deviation of the picking group and the average Euclidean distance to obtain the score contribution of each picking group. The average of the accumulated score contributions of all picking groups is used as the selection score for the picking group.

3. The path planning method for a fully automatic cabin cleaning machine according to claim 1, characterized in that, The material handling sequence includes the priority of the cleaning machine accessing each material handling group, with the priority decreasing from left to right.

4. The path planning method for a fully automatic cabin cleaning machine according to claim 1, characterized in that, The calculation method for the cost weight includes: Using any material taking sequence as the target sequence, the material taking point of any material taking group in the target sequence as the target material taking point, the grid where the cleaning machine center is located as the target grid, and the neighboring grids of the target grid as possible displacement points, calculate the ratio of the sum of the diagonal radius of the first moving frame of the cleaning machine and the diagonal radius of the nearest non-target material taking point's material taking frame to the Euclidean distance between the possible displacement point and the nearest non-target material taking point of the possible displacement point, and normalize it to obtain the cost weight of obstacle avoidance cost.

5. The path planning method for a fully automatic cabin cleaning machine according to claim 1, characterized in that, The calculation method for the current material extraction cost includes: Taking the picking point of any picking group in the target sequence as the target picking point, calculate the average Euclidean distance between the possible displacement point and all target picking points. The average value of the Euclidean distance Divide by the diagonal diameter of the XOY coordinate plane of the cabin to obtain the distance cost; Calculate the average Euclidean distance between the possible displacement point and all non-target pick-up points. Divide the diagonal diameter of the ship's cabin in the XOY coordinate plane by the average value of the Euclidean distance. Obstacle avoidance comes at a cost; The current material collection cost of a possible displacement point is obtained by weighting and summing the cost weights as the weights of obstacle avoidance cost and distance cost (1 minus the cost weight).

6. The path planning method for a fully automatic cabin cleaning machine according to claim 1, characterized in that, The calculation method for the remaining material extraction cost includes: Determine the number of path points on the shortest path between the possible displacement point and the target material collection point, calculate the Euclidean distance between the path points on the shortest path and the nearest non-target material collection point, and use the sum of the reciprocals of the Euclidean distances of all path points as the remaining material collection cost.

7. The path planning method for a fully automatic cabin cleaning machine according to claim 1, characterized in that, The objective function is calculated in the following ways: Using any point on the optimal material picking path in the material picking sequence as a marker, calculate the comprehensive material picking cost of the marker, and sum the comprehensive material picking costs of all path points to obtain the objective function.

8. The path planning method for a fully automatic cabin cleaning machine according to claim 1, characterized in that, The environmental perception data includes: the location of the material picking point, the material picking frame, and the material volume. The equipment parameter data includes: the location of the material stacking point, the moving frame of the cleaning machine, and the single material picking volume. An XOY coordinate plane is established in the cabin to be cleaned, and the coordinate plane is gridded for positioning and volume measurement.

9. The path planning method for a fully automatic cabin cleaning machine according to claim 1, characterized in that, Calculate the straight-line distance between each material taking group and different material stacking points in the material taking sequence, evaluate the total distance between different material taking groups and material stacking points, and select the material stacking point with the smallest total distance as the optimal material stacking point for the material taking group.

10. A path planning system for a fully automated cabin cleaning machine, characterized in that, include: A processor and a memory, the memory storing computer program instructions that, when executed by the processor, implement the path planning method for a fully automated cabin cleaning machine according to any one of claims 1-9.

Images