A rescue ship path planning method suitable for complex sea conditions
Patent Information
- Application Number
- CN202610870276.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-06-16
- Publication Date
- 2026-09-15
AI Technical Summary
[0004]（1）在恶劣海况下，船舶受到的环境风险表现出显著的各向异性特征，传统栅格空间中的代价函数存在离散跳变问题
1、本发明引入构建了非对称连续方向代价函数，有效避免了路径搜索过程中的代价值跳变，确保搜索出的路径具备更高的平滑度与执行性。
Smart Images
Figure CN122753384A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of intelligent navigation and maritime emergency search and rescue technology, and in particular to a method for planning the path of rescue vessels applicable to complex sea conditions. Background Technology
[0002] Maritime search and rescue missions face severe challenges in complex sea conditions. The core issue lies in the need for path planning technology to simultaneously address the coupling effect of the dynamic environment and the ship's motion characteristics. Traditional path planning methods are based on static environment assumptions, simplifying external disturbances such as wind and waves into fixed parameters and constructing discrete cost functions through a gridded space. However, in actual sea conditions, the wind and wave field exhibits significant spatiotemporal dynamics, and the ship's six degrees of freedom motion (roll, pitch, bow, heave, sway, and pitch) can trigger resonance effects due to the coupling between wave frequencies and the ship's natural frequencies, further exacerbating navigational risks. Furthermore, the ship's steering performance is closely related to path smoothness; frequent course adjustments can lead to dynamic problems such as stall and increased rolling, directly impacting rescue efficiency and safety.
[0003] With frequent maritime accidents, maritime search and rescue missions often face the following technical bottlenecks in adverse sea conditions:
[0004] (1) Under severe sea conditions, the environmental risks to ships exhibit significant anisotropic characteristics, and the cost function in traditional grid space suffers from discrete jump problems.
[0005] (2) Traditional path planning methods, such as static A Dijkstra viewed sea conditions as a static field, which cannot reflect the dynamic changes of wind and waves over time. (3) Existing methods only consider the single-factor risk of wind or waves and lack a comprehensive assessment of the six-degree-of-freedom response; traditional path optimization does not deeply integrate the six-degree-of-freedom motion characteristics of ships in waves, such as resonance, bottoming, and sideways movement; and cannot guarantee the physical safety of navigation at the theoretical level.
[0006] (4) Traditional 8-neighborhood A The algorithm easily generates zigzag paths with a large number of 45° and 90° abrupt changes. For a 10,000-ton professional rescue vessel, frequent large rudder angle turns can lead to extremely high induced drag and stall, and cause violent rolling; sometimes even exceeding the ship's steering performance.
[0007] (5) Traditional A The algorithm uses a fixed and singular weight during the rescue process, which cannot adapt to the different target requirements of the rescue phase and the evacuation phase. Summary of the Invention
[0008] The static environment assumptions mentioned above cannot reflect the spatiotemporal dynamics and anisotropic risk characteristics of wind and wave fields, and the traditional 8-neighborhood A... The algorithm tends to generate zigzag paths that exceed the ship's steering performance, leading to increased stall and roll. Furthermore, its fixed and singular weight parameters fail to adapt to the differentiated objectives of the rescue and evacuation phases. This invention addresses these technical problems by providing a rescue vessel path planning method suitable for complex sea conditions. This invention integrates asymmetric continuous wind and wave field directional costs, visibility risks, ship six-degree-of-freedom kinematic safety assessment, a ship natural stall model, and nonlinear steering penalties. Through Sigmoid dynamic weight allocation, it achieves time priority in rescue mode and safety priority in evacuation mode, improving the overall efficiency and safety of maritime emergency rescue.
[0009] The technical means employed in this invention are as follows: A method for route planning of rescue vessels applicable to complex sea conditions includes the following steps: Construct a three-dimensional spatiotemporal state raster base map, which includes water areas and coastal island and reef areas; Calculate the comprehensive environmental risk cost within the grid in the grid base map. The comprehensive environmental risk cost includes wind field risk cost, wave field risk cost, and visibility risk cost. The navigation safety assessment is conducted by combining the ship's encounter cycle with its six-degree-of-freedom motion characteristics, and no-passing zones are set on the grid base map based on the safety assessment results. Calculate the spatiotemporal dynamic navigation time cost of the raster base map; Based on the comprehensive environmental risk cost and the spatiotemporal dynamic navigation time cost, a multi-objective cost function is constructed, and a global spatiotemporal minimum risk value is introduced into the heuristic function; Using improved A The algorithm optimizes the path of the rescue vessel, and A... The algorithm is improved based on the multi-objective cost function, the heuristic function, and the dual-mode dynamic weight, which is used to switch between the safety-first path and the time-first path.
[0010] Furthermore, the construction of the three-dimensional spatiotemporal state raster base map includes: Construct an initial raster base map including the target waters and coastal island and reef areas, and identify and expand the coastal island and reef areas to be non-navigable areas; Meteorological raster data is overlaid on the initial raster map to expand the initial raster base map into a three-dimensional spatiotemporal raster base map.
[0011] Furthermore, the overlaying of meteorological raster data on the initial raster map includes: The meteorological raster data includes periodic angle data and meteorological data. The periodic angle data includes wind direction angle values and wave direction angle values. A time-based linear interpolation model is used for the scalar meteorological data, and a vector decomposition interpolation method is used for time continuity extrapolation of the periodic angle data. The calculation formula for the abscissa component after vector decomposition is as follows:
[0012] in, The x-coordinate components after vector decomposition. This represents the time progress for the current period. The angle value for the current time period. This is the angle value for the next time period.
[0013] The formula for calculating the ordinate component after vector decomposition is:
[0014] in, The ordinate component is the result of vector decomposition. The formula for calculating the periodic angle data obtained after vector decomposition interpolation is as follows:
[0015] in, These are periodic angle data obtained after vector decomposition and interpolation.
[0016] Furthermore, the formula for calculating the wind farm risk cost is as follows:
[0017] in, For the cost of wind farm risks, For wind field direction factor, Let i be the wind field risk level of the i-th grid. n For the nth grid cell, The formula for calculating the wind field direction factor is as follows:
[0018] in, , P , These are the first adjustment coefficient, the second adjustment coefficient, and the third adjustment coefficient, respectively. For wind angle, The formula for calculating the risk cost of the wave field is as follows:
[0019] in, For the cost of wave field risks, For wave field direction factor, Let i be the wave field risk level of the i-th grid. The formula for calculating the wave field direction factor is as follows:
[0020] in, , , These are the fourth, fifth, and sixth adjustment coefficients, representing the windward angle. The angle of the wave.
[0021] Furthermore, the navigation safety assessment, which combines the ship's encounter period with its six-degree-of-freedom motion characteristics, and the setting of no-passage zones on the grid base map based on the safety assessment results, includes: The vessel encounter cycle is calculated using the following formula:
[0022] in, For the ship encounter cycle, For wave speed, For the ship's speed, For the wave angle; The system determines whether a ship is in resonance based on the ship encounter cycle, and whether a ship is experiencing bottoming or surfing sideways based on wavelength, speed, and draft. When any one of the following conditions is met, the location where the danger occurs is designated as a no-pass zone.
[0023] Furthermore, the calculation of the spatiotemporal dynamic navigation time cost of the raster base map includes: Calculate a ship's speed in wind and waves using the ship stall formula; Extract the zonal and meridional components of ocean currents from the raster base image, and calculate the effective velocity component of the ocean current in the ship's heading direction. The formula for calculating the effective velocity component is as follows:
[0024] in, For the effective velocity component, This represents the zonal velocity component of the ocean current. C For the ship's course, This represents the meridional velocity component of the ocean current. Adding the aforementioned speed and effective speed components together yields the ship's speed over land. Based on the ship's speed over land, the spatiotemporal dynamic navigation time cost is calculated. The formula for calculating the spatiotemporal dynamic navigation time cost in a single step is as follows:
[0025] in, The time cost of a single-step spatiotemporal dynamic navigation. The speed of a ship over land in a single step. This refers to the distance traveled in a single step. This serves as the benchmark for single-step time consumption.
[0026] Furthermore, the calculation formula for the multi-objective cost function is as follows:
[0027] in, For a multi-objective cost function, For the i-th comprehensive environmental risk cost, For the i-th spatiotemporal dynamic travel time cost, For the i-th nonlinear steering penalty term, k 1 represents the weighting coefficient for the overall environmental risk cost. k 2 represents the weighting coefficient for the spatiotemporal dynamic navigation time cost. k 3 represents the weighting coefficient of the nonlinear steering penalty term. n For the nth grid cell, The formula for calculating the nonlinear steering penalty term is as follows:
[0028] in, For nonlinear steering penalty term, The deflection angle of adjacent headings, Based on the allowable offset angle, This is the ship's inertia penalty constant.
[0029] Furthermore, the formula for calculating the heuristic function is as follows:
[0030] in, For heuristic functions, This represents the minimum risk value in global spacetime. This is the weighting coefficient for the overall environmental risk cost. This represents the straight-line distance from the current node to the target node. To represent the side length of the grid, The weighting coefficients for the spatiotemporal dynamic travel time cost. The maximum speed at which the ship sails. The formula for calculating the maximum speed of the vessel is as follows:
[0031] in, For still water speed, This represents the zonal component of ocean currents throughout the entire time period. This represents the meridional component of ocean currents throughout the entire time period. The formula for calculating the global spatiotemporal minimum risk value is as follows:
[0032] in, This represents the risk value for wind, waves, and visibility at all times and in all areas. For any grid position, This is the final period.
[0033] Furthermore, the dual-mode dynamic weights define the decision boundary using the analytic hierarchy process (AHP) and establish a nonlinear mapping using the Sigmoid activation function. The dual-mode dynamic weights include weight coefficients for the comprehensive environmental risk cost and weight coefficients for the spatiotemporal dynamic navigation time cost. The formula for calculating the weight coefficients for the comprehensive environmental risk cost is as follows:
[0034] in, This is the weighting coefficient for the overall environmental risk cost. This represents the minimum value of the weight. The maximum value of the weight. The sensitivity coefficient, As a risk threshold, The sum of the weighting coefficients for the comprehensive environmental risk cost and the weighting coefficients for the spatiotemporal dynamic navigation time cost is 1.
[0035] Compared with the prior art, the present invention has the following advantages: 1. This invention introduces an asymmetric continuous directional cost function, which effectively avoids cost jumps during the path search process and ensures that the searched path has higher smoothness and executability.
[0036] 2. This invention introduces a ship kinematics safety assessment model to actively avoid high-risk dynamic grids that may cause resonance, bottoming out, and sideways movement, greatly improving the physical safety of the route in adverse sea conditions.
[0037] 3. This invention introduces a nonlinear steering penalty term independent of environmental weights, which ensures the smoothness of the global planning path from the underlying mathematical logic. While shortening the total time, it can reduce the accumulated redundant steering angle by more than 85%, avoiding the danger caused by large steering angles.
[0038] 4. This invention proposes a dynamic weighting mechanism that combines macro-level decision-making with micro-level response to achieve dual-objective path planning for rapid treatment and safe evacuation.
[0039] 5. This invention proposes a vector decomposition interpolation method, which effectively eliminates the interpolation jump error of wind and wave angles in complex sea state fields that change continuously with time.
[0040] Based on the above reasons, this invention can be widely promoted in fields such as intelligent ship navigation and maritime emergency search and rescue. Attached Figure Description
[0041] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0042] Figure 1 This is a comparison diagram showing the effect of path planning with and without a direction factor in an embodiment of the present invention.
[0043] Figure 2 This is a diagram showing the core density analysis of a ship under transverse waves in an embodiment of the present invention.
[0044] Figure 3 This is a comparison diagram of the traditional path and the improved path in an embodiment of the present invention.
[0045] Figure 4 This is a multi-dimensional performance radar chart of the traditional and improved paths in an embodiment of the present invention.
[0046] Figure 5 This is a diagram showing the wind and wave levels encountered by ships at different stages in an embodiment of the present invention.
[0047] Figure 6 This is a diagram showing the exposure time of the ship to various sea conditions at different stages in an embodiment of the present invention.
[0048] Figure 7 This is a comparison diagram of paths with and without steering penalty in an embodiment of the present invention.
[0049] Figure 8 This is a comparison chart of cumulative steering angles with and without steering penalty in an embodiment of the present invention.
[0050] Figure 9 This is a flowchart illustrating a method for planning the path of rescue vessels applicable to complex sea conditions, as proposed by the present invention. Detailed Implementation
[0051] To enable those skilled in the art to better understand the present invention, 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 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 should fall within the scope of protection of the present invention.
[0052] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.
[0053] This method is a multi-objective intelligent navigation and decision-making system specifically designed for large professional maritime rescue vessels. It first employs vector decomposition interpolation to construct a high-precision three-dimensional spatiotemporal environment field without angular jump errors. Based on this, by constructing an asymmetric continuous direction cost function and risk level, the ship's six-degree-of-freedom motion safety defense line actively avoids resonance, side-striking, and bottoming. Furthermore, by combining a ship stall model and the influence of ocean currents on ship speed, the navigation risks and actual single-step time are quantified. Finally, the core engine, in an improved A... The algorithm introduces a non-linear steering penalty term to eliminate the "zigzag" broken line, and relies on the mapping between the global minimum risk prediction heuristic function that satisfies the optimality condition and the Sigmoid activation function to realize the intelligent switching of dynamic weights between the time-priority rapid rescue and the safety-priority smooth evacuation modes.
[0054] like Figure 9 As shown, this invention provides a method for planning the path of rescue vessels applicable to complex sea conditions, comprising the following steps: S1. Construct a three-dimensional spatiotemporal raster base map, which includes water areas and coastal island and reef areas.
[0055] Identify non-navigable areas and overlay NetCDF meteorological data at 3-hour intervals. Perform time-linear interpolation on scalars such as wind speed and wave height; use vector decomposition interpolation for wind and wave direction to expand the search space from two dimensions to three-dimensional spatiotemporal (x,y,t).
[0056] S1 specifically includes: S11. Construct an initial raster base map including the target waters and coastal islands and reefs, and identify and expand the coastal islands and reefs to be non-navigable areas.
[0057] S12. Overlay meteorological raster data onto the initial raster map to expand the initial raster base map into a three-dimensional spatiotemporal raster base map.
[0058] Meteorological raster data is overlaid on the initial raster map, including: Meteorological raster data includes periodic angle data and meteorological data. The periodic angle data includes wind direction angle values and wave direction angle values. A time-based linear interpolation model is used for scalar meteorological data, and a vector decomposition interpolation method is used for time continuity extrapolation of the periodic angle data. The formula for calculating the abscissa component after vector decomposition is as follows:
[0059] in, The x-coordinate components after vector decomposition. This represents the time progress for the current period. The angle value for the current time period. This is the angle value for the next time period.
[0060] The formula for calculating the ordinate component after vector decomposition is:
[0061] in, The ordinate component is the result of vector decomposition.
[0062] The formula for calculating the periodic angle data obtained after vector decomposition interpolation is as follows:
[0063] in, These are periodic angle data obtained after vector decomposition and interpolation.
[0064] S2. Calculate the comprehensive environmental risk cost within the grid in the raster base map. The comprehensive environmental risk cost includes wind field risk cost, wave field risk cost, and visibility risk cost.
[0065] An asymmetric continuous directional cost function was constructed to quantify the wind field direction factor. and wave field direction factor This is to overcome the discrete jump problem of the cost function in the raster space.
[0066] The formula for calculating the risk cost of a wind farm is as follows:
[0067] in, For the cost of wind farm risks, For wind field direction factor, Let i be the wind field risk level of the i-th grid. n For the nth grid cell, the formula for calculating the wind direction factor is:
[0068] in, , P , These are the first adjustment coefficient, the second adjustment coefficient, and the third adjustment coefficient, respectively. The windward angle.
[0069] The formula for calculating the risk cost of wave fields is:
[0070] in, For the cost of wave field risks, For wave field direction factor, Let the wave field risk level be the i-th grid cell, and the formula for calculating the wave field direction factor is:
[0071] in, , , These are the fourth, fifth, and sixth adjustment coefficients, representing the windward angle. The angle of the wave.
[0072] The risk cost of visibility is calculated based on line-of-sight. The comprehensive environmental risk cost R is calculated by normalizing and weighting the risks of wind, waves, and visibility.
[0073] in, As a result of comprehensive environmental risks, This is a weighting coefficient for the risk cost of wind farms. The weighting coefficient for the cost of wave field risk. This is the weighting coefficient for the cost of visibility risk.
[0074] The World Meteorological Organization classifies visibility into 10 risk levels, from 0 to 9, based on visibility distance. These are normalized and used as the visibility risk cost. The wind and wave risk levels described below are also based on the magnitude of the wind and waves, but the wind and wave risk cost incorporates a direction cost.
[0075] S3. Conduct a navigation safety assessment by combining the ship's encounter period with its six-degree-of-freedom motion characteristics, and set no-passage zones on the grid base map based on the safety assessment results.
[0076] The formula for calculating the encounter period is as follows:
[0077] in, For the ship encounter cycle, For wave speed, For the ship's speed, The wave direction angle is used. Based on the vessel's encounter cycle, it is determined whether the vessel is experiencing resonance (roll, pitch, heave). Based on wavelength, speed, and draft, it is determined whether the vessel is experiencing bottoming or surfing sideways. When any one of these conditions is met, the location is designated as a no-pass zone. That is, the risk of bottoming in headwinds and surfing sideways in downwind conditions is assessed by evaluating heave resonance, pitch resonance, roll resonance, and considering wavelength, draft, and speed. When a vessel is navigating in a high-risk area that could trigger these hazards, it is directly designated as no-pass zone.
[0078] S4. Calculate the spatiotemporal dynamic navigation time cost of the raster base map.
[0079] A real-world speed calculation model is constructed that takes into account natural stall due to wind and waves as well as the effect of ocean currents on the actual trajectory drift of the ship, and the time cost of sailing in a single-step grid is quantified.
[0080] S4 specifically includes: S41. Considering the natural stall phenomenon, calculate the ship's speed V in wind and waves using the ship stall formula.
[0081] S42. Extract the zonal and meridional components of ocean currents from the raster base map, calculate the effective velocity component of the ocean current in the ship's heading direction, and calculate the impact of the ocean current on the ship's speed. The formula for calculating the effective velocity component is:
[0082] in, For the effective velocity component, This represents the zonal velocity component of the ocean current. C For the ship's course, This represents the meridional velocity component of the ocean current.
[0083] S43. Add the ship's speed and effective speed components to obtain the ship's speed relative to the ground.
[0084] S44. Based on the ship's speed over land, calculate the spatiotemporal dynamic navigation time cost. The formula for calculating the spatiotemporal dynamic navigation time cost for a single step is:
[0085] in, The time cost of a single-step spatiotemporal dynamic navigation. The speed of a ship over land in a single step. This refers to the distance traveled in a single step. This serves as the benchmark for single-step time consumption.
[0086] S5. Based on the comprehensive environmental risk cost and the spatiotemporal dynamic navigation time cost, construct a multi-objective cost function and introduce a global spatiotemporal minimum risk value into the heuristic function.
[0087] The formula for calculating the multi-objective cost function is as follows:
[0088] in, For a multi-objective cost function, For the i-th comprehensive environmental risk cost, For the i-th spatiotemporal dynamic travel time cost, For the i-th nonlinear steering penalty term, k 1 represents the weighting coefficient for the overall environmental risk cost. k 2 represents the weighting coefficient for the spatiotemporal dynamic navigation time cost. k 3 represents the weighting coefficient of the nonlinear steering penalty term. n This is the nth grid cell.
[0089] The formula for calculating the nonlinear steering penalty term is:
[0090] in, For nonlinear steering penalty term, The deflection angle of adjacent headings, Based on the allowable offset angle, This is the ship's inertia penalty constant.
[0091] The heuristic function h(n) is calculated using the global spatiotemporal minimum risk value within the prediction period. As a baseline, the search space is expanded to three-dimensional spacetime (x, y, t). The formula for calculating the heuristic function is:
[0092] in, For heuristic functions, This represents the minimum risk value in global spacetime. This is the weighting coefficient for the overall environmental risk cost. This represents the straight-line distance from the current node to the target node. To represent the side length of the grid, The weighting coefficients for the spatiotemporal dynamic travel time cost. This refers to the maximum speed at which the ship can sail.
[0093] The formula for calculating the maximum speed of the vessel is as follows:
[0094] in, For still water speed, This represents the zonal component of ocean currents throughout the entire time period. This represents the meridional component of ocean currents throughout the entire time period.
[0095] The formula for calculating the global spatiotemporal minimum risk value is:
[0096] in, This represents the risk value for wind, waves, and visibility at all times and in all areas. For any grid position, This is the final period.
[0097] S6. Utilizing Improved A The algorithm optimizes the path of the rescue vessel, A. The algorithm is improved based on the multi-objective cost function, heuristic function and dual-mode dynamic weight. The dual-mode dynamic weight is used to switch between the safety-first path and the time-first path.
[0098] The dual-mode dynamic weighting uses the analytic hierarchy process (AHP) to define the decision boundary and establish a nonlinear mapping using the Sigmoid activation function, enabling dynamic switching between time-priority in the rescue phase and safety-priority in the evacuation phase. The dual-mode dynamic weighting includes weight coefficients for the comprehensive environmental risk cost and the spatiotemporal dynamic navigation time cost. The formula for calculating the weight coefficient of the comprehensive environmental risk cost is as follows:
[0099] in, This is the weighting coefficient for the overall environmental risk cost. This represents the minimum value of the weight. The maximum value of the weight. The sensitivity coefficient, As the risk threshold, the sum of the weighting coefficients of the comprehensive environmental risk cost and the weighting coefficients of the spatiotemporal dynamic navigation time cost is 1.
[0100] Example This embodiment provides a method based on improved A Experimental verification of the algorithm for rescue vessel path planning under complex sea conditions: (1) Experiment 1: such as Figure 1 and Figure 2 Kernel density analysis shows that the traditional path without direction factor, which pursues the shortest distance, has a transverse wave exposure rate of 44.8%. After introducing an asymmetric direction factor, this invention successfully guides ships to sail in a head-on or tail-to-tail attitude, reducing the high-risk transverse wave exposure frequency to 20.7%.
[0101] (2) Experiment 2: such as Figure 3 and Figure 4 As shown, the post-physical evaluation engine recalculation indicates that the traditional route frequently triggers large rudder angles, leading to stall and resonance warnings. The improved algorithm of this invention reduces the total time by 4.5% and decreases the cumulative steering angle by over 85% while only increasing the range by 3.2%.
[0102] (3) Experiment 3: such as Figure 5 and Figure 6 As shown, by dynamically adjusting k1 and k2 using the Sigmoid function, the rescue phase path decisively traverses low- and medium-risk areas to arrive 9.47 hours ahead of schedule; while the evacuation phase path actively avoids areas with higher wave heights, ensuring high safety for the return journey.
[0103] (4) Experiment 4: such as Figure 7 and Figure 8 As shown, the introduction of a quadratic exponent penalty term eliminates the "zigzag" broken lines in the raster space, ensuring the smoothness of the global path and the executability of the project from the underlying mathematical logic.
[0104] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.
[0105] In the above embodiments of the present invention, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.
[0106] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.
Claims
1. A method for planning the path of rescue vessels applicable to complex sea conditions, characterized in that, Includes the following steps: Construct a three-dimensional spatiotemporal state raster base map, which includes water areas and coastal island and reef areas; Calculate the comprehensive environmental risk cost within the grid in the grid base map. The comprehensive environmental risk cost includes wind field risk cost, wave field risk cost, and visibility risk cost. The navigation safety assessment is conducted by combining the ship's encounter cycle with its six-degree-of-freedom motion characteristics, and no-passing zones are set on the grid base map based on the safety assessment results. Calculate the spatiotemporal dynamic navigation time cost of the raster base map; Based on the comprehensive environmental risk cost and the spatiotemporal dynamic navigation time cost, a multi-objective cost function is constructed, and a global spatiotemporal minimum risk value is introduced into the heuristic function; Using improved A The algorithm optimizes the path of the rescue vessel, and A... The algorithm is improved based on the multi-objective cost function, the heuristic function, and the dual-mode dynamic weight, which is used to switch between the safety-first path and the time-first path.
2. The rescue vessel route planning method applicable to complex sea conditions according to claim 1, characterized in that, The construction of the three-dimensional spatiotemporal state raster base map includes: Construct an initial raster base map including the target waters and coastal island and reef areas, and identify and expand the coastal island and reef areas to be non-navigable areas; Meteorological raster data is overlaid on the initial raster map to expand the initial raster base map into a three-dimensional spatiotemporal raster base map.
3. The rescue vessel route planning method applicable to complex sea conditions according to claim 2, characterized in that, The overlaying of meteorological raster data on the initial raster map includes: The meteorological raster data includes periodic angle data and meteorological data. The periodic angle data includes wind direction angle values and wave direction angle values. A time-based linear interpolation model is used for the scalar meteorological data, and a vector decomposition interpolation method is used for time continuity extrapolation of the periodic angle data. The calculation formula for the abscissa component after vector decomposition is as follows: in, The x-coordinate components after vector decomposition. This represents the time progress for the current period. This represents the angle value for the current time period. This is the angle value for the next time period. The formula for calculating the ordinate component after vector decomposition is: in, The ordinate component is the result of vector decomposition. The formula for calculating the periodic angle data obtained after vector decomposition interpolation is as follows: in, These are periodic angle data obtained after vector decomposition and interpolation.
4. The rescue vessel route planning method applicable to complex sea conditions according to claim 1, characterized in that, The formula for calculating the risk cost of the wind farm is as follows: in, For the cost of wind farm risks, For wind field direction factor, Let i be the wind field risk level of the i-th grid. n For the nth grid cell, The formula for calculating the wind field direction factor is as follows: in, , P , These are the first adjustment coefficient, the second adjustment coefficient, and the third adjustment coefficient, respectively. For wind angle, The formula for calculating the risk cost of the wave field is as follows: in, For the cost of wave field risks, For wave field direction factor, Let i be the wave field risk level of the i-th grid. The formula for calculating the wave field direction factor is as follows: in, , , These are the fourth, fifth, and sixth adjustment coefficients, representing the windward angle. The angle of the wave.
5. The rescue vessel route planning method applicable to complex sea conditions according to claim 1, characterized in that, The method of combining the ship's encounter cycle with its six-degree-of-freedom motion characteristics to conduct a navigation safety assessment, and setting no-passage zones on the grid base map based on the safety assessment results, includes: The vessel encounter cycle is calculated using the following formula: in, For the ship encounter cycle, For wave speed, For the ship's speed, For the wave angle; The system determines whether a ship is in resonance based on the ship encounter cycle, and whether a ship is experiencing bottoming or surfing sideways based on wavelength, speed, and draft. When any one of the following conditions is met, the location where the danger occurs is designated as a no-pass zone.
6. The rescue vessel route planning method applicable to complex sea conditions according to claim 1, characterized in that, The calculation of the spatiotemporal dynamic navigation time cost of the raster base map includes: Calculate a ship's speed in wind and waves using the ship stall formula; Extract the zonal and meridional components of ocean currents from the raster base image, and calculate the effective velocity component of the ocean current in the ship's heading direction. The formula for calculating the effective velocity component is as follows: in, For the effective velocity component, This represents the zonal velocity component of the ocean current. C For the ship's course, This represents the meridional velocity component of the ocean current. Adding the aforementioned speed and effective speed components together yields the ship's speed over land. Based on the ship's speed over land, the spatiotemporal dynamic navigation time cost is calculated. The formula for calculating the spatiotemporal dynamic navigation time cost in a single step is as follows: in, The time cost of a single-step spatiotemporal dynamic navigation. The speed of a ship over land in a single step. This refers to the distance traveled in a single step. This serves as the benchmark for single-step time consumption.
7. The rescue vessel route planning method applicable to complex sea conditions according to claim 1, characterized in that, The formula for calculating the multi-objective cost function is as follows: in, For a multi-objective cost function, For the i-th comprehensive environmental risk cost, For the i-th spatiotemporal dynamic travel time cost, For the i-th nonlinear steering penalty term, k 1 represents the weighting coefficient for the overall environmental risk cost. k 2 represents the weighting coefficient for the spatiotemporal dynamic navigation time cost. k 3 represents the weighting coefficient of the nonlinear steering penalty term. n For the nth grid cell, The formula for calculating the nonlinear steering penalty term is as follows: in, For nonlinear steering penalty term, The deflection angle of adjacent headings, Based on the allowable offset angle, This is the ship's inertia penalty constant.
8. The method for planning the path of rescue vessels under complex sea conditions according to claim 1, characterized in that, The formula for calculating the heuristic function is as follows: in, For heuristic functions, This represents the minimum risk value in global spacetime. This is the weighting coefficient for the overall environmental risk cost. This represents the straight-line distance from the current node to the target node. To represent the side length of the grid, The weighting coefficients for the spatiotemporal dynamic travel time cost. The maximum speed at which the ship sails. The formula for calculating the maximum speed of the vessel is as follows: in, For still water speed, This represents the zonal component of ocean currents throughout the entire time period. This represents the meridional component of ocean currents throughout the entire time period. The formula for calculating the global spatiotemporal minimum risk value is as follows: in, This represents the risk values for wind, waves, and visibility at all times and in all areas. For any grid position, This is the final period.
9. The method for planning the path of rescue vessels under complex sea conditions according to claim 1, characterized in that, The dual-mode dynamic weights are defined by the analytic hierarchy process (AHP) and a nonlinear mapping is established using the Sigmoid activation function. The dual-mode dynamic weights include weight coefficients for the comprehensive environmental risk cost and weight coefficients for the spatiotemporal dynamic flight time cost. The formula for calculating the weight coefficients for the comprehensive environmental risk cost is as follows: in, This is the weighting coefficient for the overall environmental risk cost. This represents the minimum value of the weight. The maximum value of the weight. The sensitivity coefficient, As a risk threshold, The sum of the weighting coefficients for the comprehensive environmental risk cost and the weighting coefficients for the spatiotemporal dynamic navigation time cost is 1.