Intelligent anchor position detection algorithm based on multi-level spatial constraint

Abstract

The invention discloses an intelligent anchor position detection algorithm based on multilevel spatial constraints. The intelligent anchor position detection algorithm comprises the following steps: establishing a ship safety distance model based on a Monte Carlo stochastic simulation technology; establishing an anchoring safe water depth detection model based on KD-Tree; and screening anchor location points which simultaneously meet safety spacing and water depth constraints. According to the method, the advantages of global fast search and local fine detection are effectively integrated by constructing a layered progressive space optimization framework. The method comprises the following steps: firstly, establishing a multi-level spatial constraint model for ship anchor position detection on a theoretical level, and defining a collaborative optimization mechanism between different levels; and secondly, organic fusion of spatial index and dynamic filtering is realized in algorithm design, an anchoring position point meeting the safe distance between ships is obtained through an anchoring safety detection model based on plane distance constraint, an anchoring position point meeting the water depth constraint is obtained through water depth detection based on KD-Tree, and compared with an existing method, the calculation efficiency is remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of ship anchoring operations, and particularly to an intelligent anchor position detection algorithm based on multi-level spatial constraints. Background Art

[0002] With the development of the global shipping industry towards intelligence, significant progress has been made in aspects such as ship navigation and collision avoidance. As an essential part of ship operations, the intelligent development of ship anchoring operations urgently needs to keep up with the current development trend. However, current anchoring operations still mainly rely on traditional electronic chart systems and crew experience judgment, which can no longer meet the requirements of modern shipping industry for safety, efficiency, and accuracy. Therefore, it is particularly important to develop an anchor position detection method that takes into account both the safe distance between ships and the required safe water depth for ships.

[0003] Currently, various resources in the port system, such as the size of anchorages, channels, berths, loading and unloading equipment, and transportation equipment, have received extensive attention from researchers. In addition, existing literature related to anchorages usually conducts anchorage allocation and planning, anchorage capacity design, etc. based on optimization algorithms or models.

[0004] Regarding ship anchor position allocation, in recent years, with the rapid development of the shipping industry and continuous technological progress, significant progress has also been made in ship anchoring research. Anchorage capacity planning is one of the important fields of anchoring research. Malekipirbazari et al. introduced a heuristic algorithm to modify the maximum empty first algorithm to optimize the utilization rate of anchor position space. The research of Madadi, Bahman et al. further considered time dynamics, took into account the arrival and departure of ships, and dynamically placed them in a polygonal anchorage, and proposed a spatio-temporal method to solve the multi-objective anchorage planning problem. Huang et al. focused on studying the essence of the utilization ability of anchorages, believing that the utilization ability of anchorages depends on the selection of the actual anchor position of ships.

[0005] In terms of anchoring safety, Oz et al., on the basis of Huang's research, first considered the safety of ship anchorages when studying the anchorage allocation problem. Their algorithm greatly improved the safety of anchor position allocation while mainly maintaining a similar utilization rate level. Some researchers also studied anchorage construction and mooring planning from the perspectives of marine spatial planning and mooring systems to further optimize ship anchorage allocation. In recent years, Zhao et al. proposed a dynamic multi-objective optimization model for anchor position allocation, which not only considered the dynamic changes in anchorage occupancy rate but also took into account the dual objectives of safety and capacity maximization, and achieved the efficient and safe utilization of anchorage resources by optimizing ship position arrangements.

[0006] In addition, in terms of anchor position selection, Xie Si and Cao et al. proposed an innovative intelligent detection algorithm, which was specifically designed for the anchor area of single-anchor moored Maritime Autonomous Surface Ships (MASS). They improved the anchor radius model and the safety distance model of moored ships, greatly promoting the intelligent development of anchor position allocation technology. Based on Cao's work, Cui et al. considered the influence of water depth on anchor position selection and modeled the safety distance for different ship types, developing a three-dimensional intelligent detection algorithm for anchor positions, which improved the efficiency of ship mooring. Zhou et al. combined the Monte Carlo algorithm with the traversal algorithm on the basis of Cui's work. The Monte Carlo algorithm was used to simulate the possible anchor positions that the ship could select, and the traversal algorithm was used to optimize and screen the simulated anchor positions to obtain the optimal mooring points that met the requirements. The experimental results showed that the combination of the traversal algorithm and the Monte Carlo algorithm ensured the utilization rate of the anchorage and the accuracy of anchor position detection, effectively improving the accuracy of ship mooring position selection, making full use of the anchorage resources, and further improving the safety and efficiency of mooring operations. Qin Tingrong et al. proposed a new anchor position allocation model with a dedicated safety channel in the anchorage. The anchorage was divided into multiple blocks by the dedicated safety channel. The width of each block was determined by the ship length classification algorithm, and the width of the channel between adjacent blocks was determined by the dedicated safety channel width algorithm. The experimental results showed that the model could provide an anchor position allocation scheme that balanced the safety and utilization rate of the anchorage and could meet the automatic mooring requirements of future autonomous navigation ships. Wang Bing transferred the research on ship mooring to the level of path planning. Based on mooring practice, combined with ship dimensions, anchorage hydrological characteristics, mooring ship attitude, etc., the safe mooring waters were quantified, and a model for identifying feasible and optimal anchor positions was established. Zhang Hongchi et al. processed the marine environment information extracted from electronic chart data, classified the grid for mooring safety, and used it to train a single-anchor mooring position selection model for unmanned ships based on decision trees. The relevant data of the target waters were substituted into the model for decision-making, so as to give suggestions on the selection of anchor positions.

[0007] Researchers have studied the anchorage configuration from different perspectives. The core goal is to optimize the utilization efficiency of anchorage resources to improve the stability and economic benefits of ship operations. However, as can be seen from the above research, the research on anchor position detection mainly focuses on anchor position allocation strategies and planning problems and a few studies carried out using intelligent algorithms. Moreover, most existing studies conduct anchor position allocation from the perspective of the safe distance between planar ships based on idealized conditions or regularized algorithms, lacking comprehensive consideration of the influence of the uncertainty factors of the complex marine environment on the selection of mooring positions, and there are still obvious deficiencies in dealing with high-dimensional space search efficiency, multi-constraint coupling optimization, and dynamic environment adaptability. Summary of the Invention

[0008] In order to solve the above problems existing in the prior art, the present invention designs an intelligent anchor detection algorithm based on multi-level spatial constraints, which is modeled according to the actual anchorage environment, meets the constraints of spatial safety distance and water depth conditions, and improves the efficiency and adaptability of anchor detection.

[0009] In order to achieve the above object, the technical solution of the present invention is as follows: an intelligent anchor detection algorithm based on multi-level spatial constraints, comprising the following steps: A. Obtaining ship parameters in the anchoring area, including ship length, ship draft, and ship type; B. Establish a ship safety distance model based on Monte Carlo random simulation technology to obtain anchorage points that meet safety spacing requirements; C. Establish a safe anchoring depth detection model based on KD-Tree to obtain anchor points that meet the water depth constraints; D. Select anchor points that meet both safety spacing and water depth constraints.

[0010] Furthermore, the method for establishing the ship safety distance model in step B is as follows: To address the complex and changing nature of anchorage environments and the random distribution of ship positions, a Monte Carlo stochastic simulation technique was used to develop a ship safety distance model. First, a random distribution of existing ships and obstacles within the anchorage was generated by sampling, simulating the collision avoidance scenarios and position conflicts that occur in real anchorages. Second, the berthing positions of waiting ships were predicted based on the simulation results, and a set of candidate anchorages with spatially random characteristics was established. Subsequently, a planar distance-constrained anchorage safety detection model was constructed, consisting of an anchoring radius model, a safety distance model, and an anchoring area detection model.

[0011] Furthermore, the calculation formula of the mooring radius model is as follows: (1) After obtaining the anchoring radius of a single ship, in order to further describe the relative safety requirements between ships in the anchored state, a safety spacing model is introduced to determine the minimum safety distance maintained between two ships. This distance satisfies the following relationship: (2) (3) On the basis of establishing the anchoring radius and safety spacing model, in order to realize the detection of the feasibility of the anchoring point space, a mooring area detection model is further constructed to judge whether the actual distance between ships meets the safety spacing requirements. The formula is as follows: (4) Where, is the radius of the ship's anchor circle; and respectively represent the radius of the anchor position circle of the existing ship and the ship to be berthed; represents the length of the ship; and respectively represent the lengths of the existing ship and the ship to be berthed; is the water depth; and respectively represent the ship types of the existing ship and the ship to be berthed, and The value ranges of both are 1 - 1.2. When the ship is a general cargo ship, and take the lower limit value; when the ship is an oil, liquefied gas, or chemical ship, and take the upper limit value; represents the actual distance between ships; assume the abscissa of point A is and the ordinate is , point A represents the position of the existing ship or other objects that impede anchoring operations in the plane rectangular coordinate system of the anchorage; point B is the simulated anchor point of the ship to be berthed, with the abscissa being and the ordinate being .

[0012] Furthermore, the method for establishing the anchor safety water depth detection model described in step C is as follows: On the basis of meeting spatial safety, the feasibility of the ship anchorage area also needs to consider the limitation of water depth conditions. To achieve the rapid acquisition and detection of water depth information in a large-scale sea area, a KD-Tree spatial index structure is introduced to construct an efficient water depth query and detection mechanism.

[0013] By establishing a spatial index tree for the two-dimensional plane coordinates in the sea area terrain grid, within the anchor radius range of each ship-to-be-berthed point during the detection process, the corresponding grid point set can be quickly retrieved to achieve rapid depth search within the local area.

[0014] Based on the neighborhood water depth point set obtained by KD-Tree query, further judge whether it meets the minimum water depth condition required for ship anchoring. For this purpose, an anchor safety water depth detection model is adopted, comprehensively considering factors such as the ship's draft depth, shielding coefficient, and wave surplus water depth, and calculating the minimum safe water depth required for ship anchoring. The calculation formula is as follows: (5) Among them, is the minimum safe water depth required for the ship to be berthed; is the maximum draft of the ship during anchoring; is the shielding coefficient, taking 1.2 when there is no swell or good shielding, and taking 1.5 when there is a swell or poor shielding; is the undulation of extra depth of water, with a value range of 2 - 3m; is the minimum measured water depth of the waiting berthing area.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. By constructing a hierarchical and progressive spatial optimization framework, the present invention effectively integrates the advantages of global fast search and local fine detection. First, at the theoretical level, a multi-level spatial constraint model for ship anchor position detection is established, clarifying the collaborative optimization mechanism between different levels; second, in the algorithm design, the organic integration of spatial indexing and dynamic filtering is realized. The anchor points that meet the safe distance between ships are obtained through the anchor safety detection model based on planar distance constraints, and the anchor points that meet the water depth constraints are obtained through the water depth detection based on KD-Tree. The calculation efficiency is significantly improved compared with the existing methods; finally, according to the actual anchorage environment, a model is built, and the reliability and practicability of the present invention in complex scenarios are verified through real experiments. The research results can provide technical support for the autonomous anchoring decision-making of intelligent ships, and also provide a new solution idea for the spatial constraint optimization problems in other fields.

[0016] 2. Based on the actual environment modeling, the present invention conducts research on intelligent anchor position detection for complex environments by considering the three-dimensional angles of the safe distance between ships and the longitudinal water depth conditions. This is not only a necessary means to solve the limitations of traditional methods, but also an important direction to improve the safety of ship anchoring, optimize the allocation of port resources, and promote the development of intelligent ship technology. Description of the Drawings

[0017] Figure 1 is the flow schematic diagram of the present invention.

[0018] Figure 2 is the three-dimensional simulation diagram of the anchorage sea area.

[0019] Figure 3 is the anchor position detection result diagram of a bulk carrier with a ship length of 190m and a draft of 7.3m under the condition of wind force ≤ 7 and water depth of 20m.

[0020] Figure 4 is the anchor position detection result diagram of a bulk carrier with a ship length of 190m and a draft of 11.2m under the condition of wind force ≤ 7 and water depth of 20m.

[0021] Figure 5 is the anchor position detection result diagram of a container ship with a ship length of 190m and a draft of 11.2m under the condition of wind force ≤ 7 and water depth of 20m.

[0022] Figure 6 is the anchor position detection result diagram of a container ship with a ship length of 225m and a draft of 11.2m under the condition of wind force ≤ 7 and water depth of 20m.

[0023] Figure 7 It is a diagram of the anchor position detection results of a container ship with a length of 117 m and a draft of 9.2 m under the condition of wind force ≤ 7 and water depth of 20 m.

[0024] Figure 8 It is a diagram of the anchor position detection results of a liquefied petroleum gas ship with a length of 225 m and a draft of 11.2 m under the condition of wind force ≤ 7 and water depth of 20 m.

[0025] Figure 9 It is a diagram of the anchor position detection results of an oil tanker with a length of 250 m and a draft of 15.6 m under the condition of wind force ≤ 7 and water depth of 20 m.

[0026] Figure 10 It is a diagram of the anchor position detection results of an ultra-large container ship with a length of 400 m and a draft of 16 m under the condition of wind force ≤ 7 and water depth of 20 m. Detailed implementation manner

[0027] The present invention will be further described below in conjunction with the accompanying drawings.

[0028] As Figure 1 shown, in order to verify the effectiveness and adaptability of the multi-level space constraint intelligent anchor position detection method proposed by the present invention, an experimental platform is jointly constructed using Python and MATLAB to realize algorithm modeling, data simulation and three-dimensional visualization. Specifically, the main algorithm logic is developed in the Python 3.11 environment, and scientific computing libraries such as NumPy, SciPy, and Pandas are used to complete the Monte Carlo sampling of the ship's anchor position, the normalization processing of water depth data, and the spatial neighborhood search based on KD-Tree. After completing the data calculation and screening, the anchor position detection results are three-dimensionally visualized by MATLAB R2019a, and its powerful graphics drawing function is used to generate high-quality topographic maps of the anchorage and spatial distribution maps of anchor points.

[0029] The experiment simulates a three-dimensional anchorage sea area of 3 nautical miles × 3 nautical miles, as Figure 2 shown. In addition, to ensure the accuracy and rationality of the experiment, the ship sizes, types and related parameters used in the simulation experiment refer to the actual data of common ships in Guishan Island Anchorage. The specific parameters are shown in Table 1 and Table 2. Table 1 shows the size and type data of the existing ships in the anchorage, and Table 2 lists the relevant parameters of the ships to be berthed. To ensure the effectiveness of the simulation results, the selection of ships to be berthed comprehensively considers their types, ship lengths and draft depths, and 8 representative ships are selected as the objects to be berthed in the simulation.

[0030] Table 1 Existing ship data in the southern anchorage of Guishan Island

[0031] Table 2 Data of Ships Waiting to Berth

[0032] Under the typical mooring conditions with the set wind force ≤ 7 levels and water depth of 15 - 20 meters, combining the ship data and environmental parameters corresponding to each ship in Table 1 and Table 2, using the mooring safety detection model based on planar distance constraints, the safety distances between different ship combinations were experimentally calculated. According to the data of the ships waiting to berth provided in the table, the present invention Figure 1 adopts the Monte-Carlo stochastic simulation algorithm combined with the KD-Tree spatial index structure according to the shown process, simulates and calculates the area that meets the requirements of mooring safety distance and water depth, and visually displays it in the form of a three-dimensional image. The experiment respectively conducts mooring position detection on 8 ships waiting to berth with different types, sizes and draft depths, and obtains 8-dimensional mooring position detection images, as Figures 3 - 10 shown. In the figure, the positions of the existing ships are represented by red dots, the mooring points that meet the requirements of the two-dimensional planar safety distance are represented by blue dots, and the green dots represent the final feasible mooring points that further meet the safety water depth limit on the basis of the blue dots.

[0033] The above images show that in the ship mooring position detection experiment, when the wind force condition is the same as the water depth of the anchorage, ships of different types, sizes and draft depths show a certain degree of difference in the number of available mooring positions.

[0034] From the perspective of draft depth, under the condition of the same ship length, the bulk carrier with a shallower draft ( Figure 3 ) has more available mooring positions, while the bulk carrier with a deeper draft ( Figure 4 ) has a reduced number of available mooring positions, indicating that the increase in draft leads to a higher requirement for water depth, thus narrowing the mooring area. For very large ships such as oil tankers ( Figure 9 ) and ultra-large container ships ( Figure 10 ), no mooring positions that meet the water depth requirements are detected within the set water depth range of 15 - 20 meters, indicating that the water depth condition cannot meet their mooring needs, and large ships are more sensitive to water depth conditions.

[0035] From the perspective of ship length, under the condition of the same ship type and draft depth, the number of anchor points that meet the mooring range requirements for the 225-meter ship ( Figure 6 ) is significantly less than that of the 190-meter ship ( Figure 5 ), and the number of anchor points that meet the water depth requirements also decreases accordingly. In addition, small ships such as product oil tankers ( Figure 7 ) have more available mooring points, indicating that the increase in ship scale intensifies the safety distance constraint in the planar space, further narrowing the feasible mooring area.

[0036] From the perspective of ship type, under the conditions of the same ship length and draft, the difference in the number of available anchor positions among general cargo ships (bulk carriers and container ships) is relatively small. And among general cargo ships ( Figure 6 ), compared with dangerous goods ships ( Figure 8 ), the number of anchor points that meet the requirements of the anchoring range in dangerous goods ships decreases, which can effectively reflect that dangerous goods ships have more stringent requirements for safety distances during anchoring operations.

[0037] To sum up, the ship draft, hull size and ship type all have varying degrees of influence on the feasible distribution of the anchorage area. Among them, the draft is the main factor restricting the feasible anchorage area. The increase in the ship draft significantly improves the requirements for the water depth conditions of the anchorage, resulting in a significant reduction in the number of available anchor positions. The increase in hull size will increase the safety distance constraint in the planar space, thus further compressing the anchorage area, but the degree of its influence is slightly lower than that of the draft. In contrast, the difference in the anchor position distribution between general cargo ships and dangerous goods ships is relatively small, but dangerous goods ships have more stringent requirements for safety distances. The experimental results are basically consistent with the expected rules, verifying the applicability and effectiveness of the present invention under different ship types and different anchoring conditions.

[0038] Focusing on the problem of intelligent detection of ship anchorage areas, the present invention proposes an intelligent anchor position detection method based on multi-level space constraints. By introducing the Monte-Carlo random simulation algorithm and the KD-Tree space index structure, anchor position detection and evaluation are carried out respectively from two levels of space safety distance constraint and water depth safety constraint, effectively improving the accuracy and calculation efficiency of anchor position detection, and obtaining the following main conclusions: (1) The draft is the most significant factor affecting anchor position detection, while the hull size and ship type mainly have an indirect impact through safety distances.

[0039] (2) The anchoring safety detection model based on planar distance constraints can reasonably describe the space safety requirements between different ships, and realizes the effective modeling of the complex environment of the anchorage area by combining with the Monte Carlo random simulation algorithm.

[0040] (3) The water depth detection mechanism based on KD-Tree greatly improves the efficiency of retrieving the water depth conditions of anchor positions in the sea area and realizes the rapid detection of anchor position candidate points.

[0041] (4) Through systematic simulation experiments, it is verified that the proposed method can accurately reflect the feasible anchorage area in the anchoring scenarios of ships with different types, sizes and drafts. The experimental results are basically consistent with the requirements of the actual anchoring environment, verifying the rationality and effectiveness of the present method.

[0042] In summary, the present invention takes into account the safety, adaptability and computational efficiency of ship mooring, and can provide effective support for optimizing the layout of the mooring area and intelligent decision-making.

[0043] The present invention is not limited to this embodiment, and any equivalent concept or change within the technical scope disclosed by the present invention shall be included in the protection scope of the present invention.

Claims

1. An intelligent anchor position detection algorithm based on multi-level spatial constraints, characterized in that: It includes the following steps: A. Obtain the ship parameters of the anchorage area, where the ship parameters include ship length, ship draft and ship type; B. Establish a ship safety distance model based on the Monte Carlo random simulation technique to obtain anchor positions that meet the safety spacing; C. Establish an anchoring safety water depth detection model based on KD-Tree to obtain anchor positions that meet the water depth constraint; D. Screen the anchor positions that simultaneously meet the safety spacing and water depth constraint.

2. The intelligent anchor position detection algorithm based on multi-level spatial constraints according to claim 1, characterized in that: The method for establishing the ship safety distance model described in step B is as follows: In view of the characteristics of the complex and changeable environment in the anchorage area and the random distribution of ship positions, the Monte Carlo random simulation technique is used to establish a ship safety distance model; First, generate the random distribution of existing ships and obstacles in the anchorage through sampling to simulate the collision avoidance scenarios and position conflicts existing in the real anchorage; Second, predict the berthing positions of the ships to be berthed based on the simulation results and establish an alternative anchor position set with spatial random characteristics; Subsequently, construct an anchoring safety detection model based on plane distance constraint composed of an anchoring radius model, a safety spacing model and an anchorage area detection model.

3. The intelligent anchor position detection algorithm based on multi-level spatial constraints according to claim 2, characterized in that: The calculation formula of the anchoring radius model is as follows: (1) After obtaining the anchoring radius of a single ship, to further describe the relative safety requirements between ships in the anchored state, a safety spacing model is introduced to determine the minimum safe distance to be maintained between two ships. , and this distance satisfies the following relationship: (2) (3) On the basis of establishing the anchoring radius and safety spacing model, in order to realize the detection of the spatial feasibility of the anchoring point, further construct an anchorage area detection model to judge whether the actual distance between ships meets the requirements of the safety spacing. The formula is as follows: (4) In the formula, is the radius of the anchor position circle of the ship; and respectively represent the radii of the anchor position circles of the existing ship and the ship to be berthed; represents the length of the ship; and are respectively the lengths of the existing ship and the ship to be berthed; is the water depth; and respectively represent the ship types of the existing ship and the ship to be berthed. and The value ranges of both are 1 - 1.

2. When the ship is a general cargo ship, and take the lower limit value; when the ship is an oil product, liquefied gas and chemical ship, and take the upper limit value; represents the actual distance between ships; Suppose the abscissa of point A is , and the ordinate is . Point A represents the position of an existing ship or other object that obstructs the anchoring operation in the plane rectangular coordinate system where the anchorage is located; point B is the simulated anchoring point of the ship to be berthed, with the abscissa being and the ordinate being .

4. The intelligent anchor position detection algorithm based on multi-level spatial constraints according to claim 1, wherein: The method for establishing the anchoring safety water depth detection model described in step C is as follows: On the basis of meeting spatial safety, the feasibility of the ship anchorage area also needs to consider the limitation of water depth conditions; In order to realize the rapid acquisition and detection of water depth information in large-scale sea areas, introduce the KD-Tree spatial index structure and construct an efficient water depth query and detection mechanism; Establish a spatial index tree for the two-dimensional plane coordinates in the sea area terrain grid, so that in the process of detection, within the anchoring radius range of each berthing point to be berthed, quickly retrieve the corresponding grid point set to realize fast depth search in the local area; Based on the neighborhood water depth point set obtained by KD-Tree query, further judge whether it meets the minimum water depth condition required for ship anchoring; For this purpose, adopt an anchoring safety water depth detection model, comprehensively consider factors such as ship draft depth, shielding coefficient and wave extra water depth, and calculate the minimum safety water depth required for ship anchoring. The calculation formula is as follows: (5) Among them, is the minimum safe water depth required for the ship to be berthed; is the maximum draft of the ship when at anchor; is the shielding coefficient, taking 1.2 when there is no swell or good shielding, and 1.5 when there is swell or poor shielding; is the wave additional water depth, with a value of 2 - 3m; is the minimum measured water depth of the waiting berthing area.

Images