Intelligent ship global path planning system based on improved sparrow search algorithm
By improving the sparrow search algorithm and combining Cubic chaotic mapping and Gaussian random walk perturbation, the problems of low global optimization efficiency and robustness of unmanned ships in complex environments are solved, generating low-energy smooth paths that meet the requirements of ship navigation and achieving efficient path planning.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SHANGHAI JIAOTONG UNIV
- Filing Date
- 2026-02-26
- Publication Date
- 2026-04-10
AI Technical Summary
In complex navigation environments, traditional algorithms for unmanned vessels suffer from low global optimization efficiency and poor robustness. Furthermore, existing swarm intelligence algorithms are inadequate in terms of population initialization uniformity and local optimum avoidance, making it difficult to meet the rationality and regulatory requirements of ship navigation.
An improved sparrow search algorithm is adopted, which combines Cubic chaotic mapping and Gaussian random walk perturbation for population initialization. The electronic route map is processed by fuzzy C-means clustering and OTU segmentation. Rasterization and obstacle safety distance are designed to optimize population update and fitness calculation.
It improves the search capability of global path planning, avoids local optimum traps, generates low-energy smooth paths that meet the requirements of ship navigation, and satisfies the reliability requirements of real-time navigation.
Smart Images
Figure CN121829564A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the field of path planning, and in particular to an intelligent ship global path planning system based on an improved sparrow search algorithm. BACKGROUND
[0002] The real navigation environment of an unmanned ship has many obstacles and a large number of map grids, and using traditional A , RRT and other classic algorithms has the disadvantages of low global optimization efficiency and poor robustness, so it is necessary to introduce a global optimization algorithm to achieve faster convergence and more accurate global optimal path. In addition, considering the rationality of the ship route, regulatory requirements and other factors, many boundary conditions need to be added to the global path planning algorithm, and the traditional algorithm is complex to add conditions, so the swarm intelligence optimization algorithm can be more concise and fast, and therefore the sparrow search algorithm is used as the global optimization algorithm.
[0003] The sparrow search algorithm (SSA) is an algorithm based on the foraging behavior and anti-predation characteristics of sparrow groups, which divides the sparrows in the population into three categories: discoverers, followers and scouts. The role of the discoverer is to search for food resources for the entire population and indicate the direction of foraging for the followers; Although the sparrow search algorithm performs better than other swarm intelligence algorithms, it still has problems such as slow convergence speed, decreased population diversity in the later iterations and easy to fall into local optimal solution when solving complex optimization problems such as unmanned ship path. In view of the randomness and ergodicity characteristics of chaos, in the past, researchers have used Logistic mapping to ensure the uniformity of population distribution when initializing the sparrow population; the random population generation method has the problems of scattered initial individual distribution, weak global search ability and the need for more iteration times to approach the optimal solution; Other swarm intelligence algorithms such as particle swarm algorithm have insufficient fusion ability of navigation information in the electronic channel map scenario, high path fitness value, do not meet the requirements of ship navigation regulations and energy optimization, have not solved the core problems of population initialization uniformity and local optimal avoidance, and have not designed an adaptive scheme for the rasterization processing of the electronic channel map and the obstacle safety distance.
[0004] Therefore, an intelligent ship global path planning system based on an improved sparrow search algorithm is proposed. SUMMARY
[0005] In view of the above defects of the prior art, the present application solves the problems of population initialization uniformity and local optimal avoidance, and proposes an adaptive scheme for the rasterization processing of the electronic channel map and the obstacle safety distance.
[0006] To achieve the above objectives, this invention provides an intelligent ship global path planning system based on an improved sparrow search algorithm, comprising an electronic navigation chart preprocessing module, a rasterization modeling module, an improved SSA parameter initialization module, a Cubic chaotic population generation module, a population update and fitness calculation module, and a local optimum discrimination and perturbation module. The electronic navigation chart preprocessing module loads the electronic navigation chart of the target sea area and further generates a binarized map. The rasterization modeling module performs dilation processing on unnavigable areas. The population update and fitness calculation module updates the positions of the discoverer, followers, and scouts, calculates individual fitness values, and saves the globally optimal position. .
[0007] Furthermore, in the electronic navigation chart preprocessing module, a fuzzy C-means clustering algorithm is used to extract navigable and non-navigable areas, and a binarized map is generated by OTU image segmentation; The rasterization modeling module sets the raster side length. L is the length of the MASS ship, and grid coordinates are defined to expand the unnavigable area. The improved SSA parameter initialization module sets the population size N=100 and the maximum number of iterations iter max =500, discoverers account for 30%, followers account for 20%, scouts account for 15%, warning value R2=0.8, solution space dimension d=20; The Cubic chaotic population generation module is based on the formula , Generate chaotic sequences, through ; The local optimum discrimination and perturbation module calculates the standard deviation of the fitness of the Gaussian random walk. ,if The difference between the two values is less than the set threshold of 10. -3 ,pass right Add perturbation.
[0008] Furthermore, the principle of the expansion process of the rasterization modeling module is as follows: in order to avoid collisions with the MASS ship, the non-navigable area is expanded, obstacles that are less than 1 grid are expanded to 1 grid, and a safety distance of 0.5 grids is reserved at the edge of the obstacle.
[0009] Furthermore, if the number of iterations of the local optimal discrimination and perturbation module reaches iter max Output the globally optimal path; otherwise, repeat the iteration.
[0010] Furthermore, the navigation information in the electronic navigation chart includes virtual navigation marks, isobaths, and coordinates of no-navigation zone boundaries.
[0011] Furthermore, the grid coordinates are defined as follows: the top left corner is the origin, the horizontal direction to the right is the positive x-axis, and the vertical downward direction is the positive y-axis. The center point coordinates of the grid are , ; , These are the row and column numbers of the raster, respectively.
[0012] Furthermore, Control parameters of the Cubic chaotic population generation module initial value The range of values for a chaotic sequence is ,and .
[0013] Furthermore, the fitness standard deviation of the Gaussian random walk It follows a standard normal distribution N(0,1). Individuals randomly selected from the population. iter is the current iteration number. max This represents the maximum number of iterations.
[0014] Furthermore, the length of the MASS vessel is 10-50m.
[0015] Furthermore, the grid side length It is 10-75m.
[0016] The present invention has the following technical effects: This invention overcomes the efficiency bottleneck and robustness deficiencies of traditional path planning algorithms in complex navigation environments by combining "improved Sparrow Search Algorithm (SSA) + environmental modeling adapted to ship scenarios + accurate waterway information analysis". (1) Algorithm level: The combination of Cubic chaotic mapping and Gaussian random walk perturbation not only solves the problem of population initialization uniformity, but also avoids the local optimum trap in the later stage of iteration, making the algorithm significantly better than A in terms of global optimization ability. Traditional algorithms such as RRT and unimproved particle swarm optimization algorithms; (2) Scene adaptation level: Raster modeling, obstacle expansion design, and energy consumption-oriented fitness function accurately match the core requirements of autonomous ships (MASS) for "path smoothness, navigation safety and low energy consumption"; (3) Data processing level: The combination of fuzzy C-means clustering and OTU segmentation enables accurate analysis of navigable / unnavigable areas in the electronic navigation chart, providing a reliable environmental input basis for the algorithm.
[0017] The following will further explain the concept, specific structure, and technical effects of the present invention in conjunction with the accompanying drawings, so as to fully understand the purpose, features, and effects of the present invention. Attached Figure Description
[0018] Figure 1 This is a flowchart of a preferred embodiment of the present invention. Detailed Implementation
[0019] The following description, with reference to the accompanying drawings, illustrates several preferred embodiments of the present invention to make its technical content clearer and easier to understand. The present invention can be embodied in many different forms, and the scope of protection of the present invention is not limited to the embodiments mentioned herein.
[0020] In the accompanying drawings, components with the same structure are indicated by the same numerical designation, and components with similar structures or functions are indicated by similar numerical designations. The dimensions and thicknesses of each component shown in the drawings are arbitrary, and the present invention does not limit the dimensions and thicknesses of each component. To make the illustrations clearer, the thickness of some components has been appropriately exaggerated in the drawings.
[0021] like Figure 1 As shown, an intelligent ship global path planning system based on an improved sparrow search algorithm includes an electronic navigation chart preprocessing module, a rasterization modeling module, an improved SSA parameter initialization module, a Cubic chaotic population generation module, a population update and fitness calculation module, and a local optimum discrimination and perturbation module. The electronic navigation chart preprocessing module is used to load the electronic navigation chart of the target sea area, use the fuzzy C-means clustering algorithm to extract navigable and non-navigable areas, and generate a binarized map through OTU image segmentation. Preferably, the navigation information in the electronic navigation chart of the target sea area includes virtual navigation marks, isobaths, and coordinates of the boundaries of the restricted navigation zone; The rasterization modeling module sets the grid side length Ld=1.5×MASS ship length, defines the grid coordinates, and performs expansion processing on unnavigable areas; Preferably, the expansion process is as follows: to avoid collisions with the MASS ship, the non-navigable area is expanded: obstacles less than 1 grid are expanded to 1 grid, and a safety distance of 0.5 grids is reserved at the edge of the obstacle; The SSA parameter initialization module was improved by setting the population size N=100, the maximum number of iterations itermax=500, the proportion of discoverers 30%, the proportion of followers 20%, the proportion of scouts 15%, the warning value R2=0.8, and the solution space dimension d=20. The average time taken by the improved SSA is much lower than the time threshold for real-time navigation of the MASS ship, and the rasterization process uses a granularity of 1.5 times the ship length, balancing accuracy and computational efficiency. The fitness value variance of the improved SSA is significantly lower than that of the traditional SSA and particle swarm algorithm, and the fluctuation of 50 repeated simulations is minimal, which can meet the reliability requirements of long-term navigation of the MASS ship. The Cubic chaotic population generation module is based on the formula , Generate chaotic sequences, through By combining Cubic chaotic mapping initialization with Gaussian random walk perturbation, the global search capability of the improved SSA is increased by more than 30% compared with the traditional algorithm, which can effectively avoid local optimum traps. The population update and fitness calculation module updates the positions of discoverers, followers, and scouts, calculates individual fitness values, and saves the global best position Xbest. The local optimum discrimination and perturbation module calculates the fitness standard deviation. If the fitness standard deviation... The difference between the two values is less than the set threshold of 10. -3 ,pass right Add perturbation.
[0022] Preferably, if the number of iterations of the local optimum determination and the perturbation module reaches iter max Output the globally optimal path; otherwise, repeat the iteration.
[0023] Preferably, the grid coordinates are defined as follows: the top left corner is the origin, the horizontal direction to the right is the positive x-axis, and the vertical direction downwards is the positive y-axis. The center point coordinates of the grid are , , , These are the row and column numbers of the raster, respectively.
[0024] Preferred control parameters for Cubic chaotic mapping initial value The range of values for a chaotic sequence is ,and
[0025] Preferably, the perturbation parameter δ of the Gaussian random walk follows a standard normal distribution N(0,1). Iter is an individual randomly selected from the population, where t is the current iteration number. max This represents the maximum number of iterations.
[0026] Preferably, the length of the MASS ship is 10-50m, corresponding to the grid side length. The safe distance for obstacle expansion is 10-75m, and the safe distance for obstacle expansion is 7.5-37.5m.
[0027] The beneficial effects of this invention are as follows: This intelligent ship global path planning system based on the improved sparrow search algorithm, through the combination of Cubic chaotic mapping initialization and Gaussian random walk perturbation, improves the global search capability of the improved SSA by more than 30% compared with the traditional algorithm, and can effectively avoid local optimum traps; the fitness function integrates path length and turning energy consumption, reducing the number of path turns by 70%, which meets the requirements of low energy consumption and smoothness of ship navigation, and is suitable for the complex navigation environment of electronic navigation charts. The average time of the improved SSA is much lower than the time threshold of real-time navigation of the MASS ship, and the rasterization process adopts a granularity of 1.5 times the ship length, balancing accuracy and computational efficiency; the variance of the fitness value of the improved SSA is significantly lower than that of the traditional SSA and particle swarm algorithm, and the fluctuation of 50 repeated simulations is minimal, which can meet the reliability requirements of long-term navigation of the MASS ship.
[0028] The preferred embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make numerous modifications and variations based on the concept of the present invention without creative effort. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the present invention through logical analysis, reasoning, or limited experimentation on the basis of existing technology should be within the scope of protection defined by the claims.
Claims
1. A global path planning system for intelligent ships based on an improved sparrow search algorithm, comprising an electronic navigation chart preprocessing module, a rasterization modeling module, an improved SSA parameter initialization module, a Cubic chaotic population generation module, a population update and fitness calculation module, and a local optimum discrimination and perturbation module; the electronic navigation chart preprocessing module is used to load the electronic navigation chart of the target sea area and further generate a binary map; the rasterization modeling module performs dilation processing on unnavigable areas; the population update and fitness calculation module updates the positions of discoverers, followers, and scouts, calculates individual fitness values, and saves the globally optimal position. .
2. The intelligent ship global path planning system based on the improved sparrow search algorithm as described in claim 1, characterized in that, In the electronic navigation chart preprocessing module, fuzzy C-means clustering algorithm is used to extract navigable and non-navigable areas, and a binarized map is generated by OTU image segmentation; The rasterization modeling module has a grid side length of... L is the length of the MASS ship, and grid coordinates are defined to expand the unnavigable area. The improved SSA parameter initialization module sets the population size N=100 and the maximum number of iterations iter max =500, discoverers account for 30%, followers account for 20%, scouts account for 15%, warning value R2=0.8, solution space dimension d=20; The Cubic chaotic population generation module is based on the formula , Generate chaotic sequences, through .
3. The intelligent ship global path planning system based on the improved sparrow search algorithm as described in claim 2, characterized in that, The local optimum discrimination and perturbation module calculates the standard deviation of the fitness of the Gaussian random walk. ,if The difference between the two values is less than the set threshold of 10. -3 ,pass right Add perturbation; In the formula, the fitness standard deviation of the Gaussian random walk It follows a normal distribution N(0,1); The calculation formula is as follows: Where fa is the average fitness of the sparrow population, and f is the standard deviation. The control parameter f has the following possible values: The fitness standard deviation The difference between the two values is less than the specified value of 10. -3 If the population is considered to have fallen into a local optimum during the iteration process, then a Gaussian random walk strategy is used to adjust the fitness of the sparrow population. f i The optimal individual is perturbed to help the algorithm escape local optima. The formula for generating new sparrow individuals is as follows: In the formula, A random individual within the new sparrow population. To maximize the number of iterations, we utilize the property that convex functions decrease in the first quadrant: as the number of iterations increases... t As the disturbance increases, the disturbance is gradually reduced.
4. The intelligent ship global path planning system based on the improved sparrow search algorithm as described in claim 3, characterized in that, The principle of the expansion process of the rasterization modeling module is as follows: in order to avoid collisions with the MASS ship, the non-navigable area is expanded, and obstacles that are less than 1 grid are expanded to 1 grid, with a safety distance of 0.5 grids reserved at the edge of the obstacle.
5. The intelligent ship global path planning system based on the improved sparrow search algorithm as described in claim 4, characterized in that, If the number of iterations of the local optimum discrimination and perturbation module reaches iter max Output the globally optimal path; otherwise, repeat the iteration.
6. The intelligent ship global path planning system based on the improved sparrow search algorithm as described in claim 5, characterized in that, The navigation information in the electronic navigation chart includes virtual navigation marks, contour lines, and coordinates of the boundaries of restricted navigation zones.
7. The intelligent ship global path planning system based on the improved sparrow search algorithm as described in claim 6, characterized in that, The grid coordinates are defined as follows: the top left corner is the origin, the horizontal direction to the right is the positive x-axis, and the vertical direction downwards is the positive y-axis. The center point coordinates of the grid are , ; , These are the row and column numbers of the raster, respectively.
8. The intelligent ship global path planning system based on the improved sparrow search algorithm as described in claim 7, characterized in that, The control parameters of the Cubic chaotic population generation module initial value The range of values for a chaotic sequence is ,and .
9. The intelligent ship global path planning system based on the improved sparrow search algorithm as described in claim 8, characterized in that, The length of the MASS vessel is 10-50m.
10. The intelligent ship global path planning system based on the improved sparrow search algorithm as described in claim 9, characterized in that, The grid side length The safe distance for obstacle expansion is 10-75m, and the safe distance for obstacle expansion is 7.5-37.5m.