Arctic navigation channel path planning method based on improved ant colony algorithm

Abstract

The invention discloses an improved ant colony algorithm-based arctic channel path planning method, which belongs to the technical field of navigation, and comprises the following steps of: (1) calculating a navigable area of an icebreaker in an arctic region by combining ice condition data, meteorological data and the level of the icebreaker; (2) initializing pheromone concentration for effective nodes in the navigable area; (3) finding out a next-step transferable node set naavi of the kth ant starting from the current node i at the moment t, and calculating the probability of the ant starting from the node i to the node j in the naavi according to the pheromone concentration; completing one iteration; and (4) updating the pheromone concentration on the path from the node i to the node j, and entering the next iteration until a termination condition is met. According to the north pole channel path planning method based on the improved ant colony algorithm, the initial pheromone concentration is optimized while the characteristics of the ant colony algorithm are reserved, the early blind search is remarkably reduced, the calculation amount is reduced, and the search efficiency is improved.

Description

Technical Field

[0001] This invention relates to the field of image processing technology, and in particular provides a method for Arctic shipping route planning based on an improved ant colony algorithm. Background Technology

[0002] With global warming, Arctic sea ice is melting at an accelerated pace, and the extent and density of sea ice in the Arctic region have decreased significantly. The navigability of the Arctic shipping route has been significantly improved. As the potential shortest sea route connecting Asia, Europe, and North America, it has important economic and strategic value. However, the complex ice conditions and meteorological conditions of the Arctic shipping route pose many challenges to the practical application of traditional route planning algorithms, making it difficult to meet the high efficiency and safety requirements of Arctic shipping.

[0003] The navigability of Arctic shipping routes is influenced by a variety of factors. When ships navigate in specific ice areas, sea ice conditions and the ship's icebreaking capability are the two most critical physical factors affecting the quantification of navigation risks. Sea ice conditions are mainly described by two parameters: sea ice concentration and sea ice thickness, and their derivatives such as sea ice coverage and sea ice age. In recent years, ant colony optimization (ACO) algorithms have been well-received in Arctic route planning problems due to their advantages of pheromone memory, strong positive feedback, and robustness, enabling them to continuously track paths in complex environments. However, they also suffer from slow convergence speed and susceptibility to local optima. Furthermore, the initial pheromone concentration allocation method and probabilistic selection mechanism lack specificity, and unnecessary exploration reduces the algorithm's convergence, especially considering the unique characteristics of navigation in ice-covered areas. Improved strategies are needed to enhance its performance and applicability. Summary of the Invention

[0004] To address the technical problems of slow convergence speed and susceptibility to local optima in existing Arctic shipping routes, this invention proposes an Arctic shipping route planning method based on an improved ant colony algorithm, which can solve the above problems.

[0005] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: A method for Arctic shipping route planning based on an improved ant colony algorithm includes:

[0006] (1) Calculate the navigable area of ​​the icebreaker in the Arctic region by combining ice condition data, meteorological data and the icebreaker's class;

[0007] (2) Initialize the pheromone concentration for valid nodes in the navigable area, wherein the valid nodes are those marked as navigable, including:

[0008] (21) Use a heuristic search algorithm to search for the initial optimal path between the start point and the end point from the effective nodes. All nodes on the initial optimal path form a node set Pi. Find effective nodes located near the initial optimal path. The effective nodes near Pi together form a node set Pn.

[0009] (22) Set the initial pheromone concentration of nodes in Pn to be higher than the initial pheromone concentration of other effective nodes;

[0010] (3) Find the set of nodes navi that the k-th ant can move to next from the current node i at time t. Calculate the probability that the ant will move from node i to node j in navi based on the pheromone concentration. ;

[0011] A random number r is generated in the range [0,1]. The random number r is compared with the probability threshold q(t). When r is not greater than q(t), a roulette wheel selection mechanism is used to randomly select the next node based on the probability distribution of the k-th ant moving from node i to each node in navi. When r is greater than q(t), the node with the highest probability is directly selected as the next node. This process continues until the endpoint is reached, completing one iteration.

[0012] (4) After completing one iteration, update the pheromone concentration on the path from node i to node j, and proceed to the next iteration until the termination condition is met.

[0013] Compared with existing technologies, the advantages and positive effects of this invention are as follows: The Arctic shipping route planning method based on the improved ant colony algorithm of this invention first determines the navigable areas for vessels of different ice classes by introducing key parameters such as ice condition data and meteorological data. Then, the improved ant colony algorithm performs route planning based on these navigable areas, which helps reduce the computational load and improve the computational speed. This application improves the ant colony algorithm by employing a heuristic search algorithm to search for the initial optimal path while retaining its characteristics. This initial optimal path is used to optimize the initial pheromone concentration, significantly reducing early blind searches, decreasing computational load, and improving search efficiency. By improving the probabilistic selection mechanism of the traditional ant colony algorithm, it prevents the algorithm from getting trapped in local optima too early. By enhancing path quality, it improves the convergence of the algorithm, enabling the effective planning of continuous navigable routes with shorter distances for vessels of different ice classes, providing a scientific and efficient solution for Arctic shipping route planning.

[0014] Other features and advantages of the present invention will become clearer after reading the detailed description of the embodiments of the present invention in conjunction with the accompanying drawings. Attached Figure Description

[0015] Figure 1This is a flowchart of an embodiment of the Arctic shipping route planning method based on the improved ant colony algorithm proposed in this invention; Figure 2 This is a flowchart illustrating the calculation of navigable areas for icebreakers in the Arctic region in one embodiment of the Arctic route planning method based on the improved ant colony algorithm proposed in this invention. Figure 3 This is a grid network diagram of the navigable area for icebreakers of different levels, as proposed in this invention, based on an improved ant colony algorithm for Arctic route planning. Figure 4 This is a schematic diagram of searching candidate nodes in one embodiment of the Arctic shipping route planning method based on the improved ant colony algorithm proposed in this invention; Figure 5 This is a schematic diagram of the nodes near the initial path expansion in one embodiment of the Arctic route planning method based on the improved ant colony algorithm proposed in this invention. Detailed Implementation

[0016] The specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0017] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] It should be noted that in the description of this invention, terms such as "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," indicating directional or positional relationships, are based on the directional or positional relationships shown in the accompanying drawings. These are merely for ease of description and do not indicate or imply that the device or element must have a specific orientation, or be constructed and operated in a specific orientation; therefore, they should not be construed as limitations on this invention. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0019] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0020] Example 1: This example proposes a method for Arctic shipping route planning based on an improved ant colony algorithm, such as... Figure 2 As shown, it includes: (1) Combine ice condition data, meteorological data and icebreaker class to calculate the navigable area of ​​icebreaker in the Arctic region.

[0021] According to the POLARIS Code, ice conditions must be assessed based on the Risk Value (RIV) and the Ice Navigability Index (RIO). The Risk Value (RIV) reflects the severity of the conditions faced by vessels of different ice classes (such as PC1-PC7, IA, etc.) under specific ice conditions. The RIV value can be obtained by referring to the POLARIS Risk Index table. Ice conditions depend on the development and decay stage of the ice and are classified according to ice thickness.

[0022] The POLARIS rules are primarily based on ice condition data. However, the suitability of navigation in the Arctic region is also influenced by weather conditions. Therefore, this scheme further considers multiple risk factors in the polar environment, introducing three non-ice parameters: temperature (T), wind speed (U), and ocean current speed. Critical threshold conditions are set and quantified as additional risk constraints for navigation in the Arctic waterways. This determines navigable areas for vessels of different ice classes. Subsequently, an improved ant colony algorithm is used for path planning based on these navigable areas, which helps reduce the computational load and increase the computational speed of the improved ant colony algorithm.

[0023] (2) Initialize the pheromone concentration for valid nodes in the navigable area. Valid nodes are those marked as navigable, including: (21) Use a heuristic search algorithm (such as the A* algorithm) to search for the initial optimal path between the start and end points from the valid nodes. All nodes on this initial optimal path form a node set Pi. Find valid nodes located near the initial optimal path. These valid nodes, together with Pi, form a node set Pn, such as... Figure 5 As shown.

[0024] (22) Set the initial pheromone concentration of nodes in Pn to be higher than the initial pheromone concentration of other effective nodes.

[0025] This scheme uses a heuristic search algorithm to find the initial optimal path, which is used to optimize the initial pheromone concentration. While retaining the characteristics of the ant colony algorithm, it significantly reduces early blind search, reduces computational load, and improves search efficiency.

[0026] (3) Find the set of nodes navi that the k-th ant can move to next from the current node i at time t. Calculate the probability that the ant will move from node i to node j in navi based on the pheromone concentration. .

[0027] A random number r is generated in the range [0,1]. The random number r is compared with the probability threshold q(t). When r is not greater than q(t), a roulette wheel selection mechanism is used to randomly select the next node based on the probability distribution of the k-th ant moving from node i to each node in navi. When r is greater than q(t), the node with the highest probability is directly selected as the next node. This process continues until the endpoint is reached, completing one iteration.

[0028] By improving the probabilistic selection mechanism of the traditional ant colony algorithm, which avoids the risk of prematurely trapping ants in local optima by uniformly selecting the node with the highest probability, this scheme allows ants to explore potentially optimal paths with a certain probability when r is not greater than q(t). This enhances global search capabilities and prevents the omission of potentially optimal paths. Furthermore, the algorithm's convergence is improved by enhancing path quality, balancing global exploration with convergence speed. This dynamic trade-off between exploration and utilization avoids local optima, improves path quality, and enhances the algorithm's adaptability in complex environments. This approach effectively plans continuous, navigable routes with shorter distances for vessels of different ice classes, providing a scientific and efficient solution for Arctic shipping route planning.

[0029] (4) After completing one iteration, update the pheromone concentration on the path from node i to node j, and proceed to the next iteration until the termination condition is met.

[0030] In some embodiments, Arctic shipping route planning can be performed separately for icebreakers of different classes, generating navigable environment grids for icebreakers of different classes. For example... Figure 3 The diagram shows a multi-level, tiered grid network of navigable areas. PC1, PC2, PC3…1A, 1B, 1C represent the icebreaker classes, with icebreaking capabilities decreasing sequentially from PC1 to 1C.

[0031] In some embodiments, before step (1), a data preprocessing step is also included, which performs quality control on the sea ice thickness, sea ice concentration, wind speed, ocean current speed and temperature data in the Arctic region, removes outliers and selects data suitable for the navigation period, and assimilates data from different sources into the same resolution polar grid, so as to provide a high-quality data foundation for the subsequent construction of a navigable grid.

[0032] In some embodiments, such as Figure 1 As shown, the ice condition data includes ice condition risk value and ice concentration, and the meteorological data includes temperature, wind speed and ocean current speed. In step (1), the method for calculating the navigable area of ​​the icebreaker in the Arctic region includes:

[0033] (11) Calculate the seaworthiness index RIO of the icebreaker at each grid point, and generate the navigable environment grid of the icebreaker, including: Based on the icebreaker's class, obtain the ice condition risk value (RIV) corresponding to the icebreaker in the grid map, and obtain the ice concentration (C) of each grid point in the grid map.

[0034] Calculate the seaworthiness index RIO of this class of icebreaker at each grid point in the grid map: RIO = C × RIV.

[0035] The seaworthiness index RIO is used to indicate the likelihood of a ship traversing ice conditions. It is calculated as the product of the corresponding ice risk value RIV and the ice concentration C.

[0036] The higher the RIO value, the lower the severity of the ice conditions, and the more likely the vessel is to pass. If the RIO value is negative, it indicates that the area is not operable, or the vessel needs to slow down.

[0037] Based on the POLARIS model framework, this application further considers the multiple risk factors of the polar environment, introduces three non-ice parameters: temperature T, wind speed U, and ocean current speed, sets critical condition thresholds, and quantifies them as additional risk constraint factors for navigation in the Arctic shipping route.

[0038] (12) Generate additional risk constraint factors based on meteorological data. : .

[0039] (13) Navigable area By RIO and Common constraints: .

[0040] in, A node is considered valid when its value is 1.

[0041] In some embodiments, the method for searching the initial optimal path in step (21) includes:

[0042] (211) From the valid nodes in the navigable area, find all candidate nodes that can be transferred from the starting point to each next step, such as... Figure 4 As shown.

[0043] (212) Calculate the distance estimate F(n) from the current node to the destination through each candidate node. t ):

[0044] F(n t )=g(n t )+h(n t ).

[0045] g(n t ) represents the distance from the starting point to the candidate node n t The actual distance value, h(n) t ) indicates starting from candidate node n t The estimated distance to the destination.

[0046] (212) Compare the candidate nodes with the smallest distance estimates and use them as the target nodes for the next step. All target nodes form the initial optimal path.

[0047] In some embodiments, h(n) t Using Euclidean distance as the heuristic function, the calculation method is as follows: .

[0048] For candidate node n t coordinates The coordinates of the endpoint.

[0049] This invention introduces a hybrid selection mechanism to improve the traditional ant colony algorithm. The path selection steps are described as follows: In each iteration, the probability of the k-th (1≤k≤M) ant moving from node i to node j at time t is expressed as: .

[0050] In some embodiments, in step (3) The calculation method is as follows: .

[0051] in, This represents the pheromone concentration from node i to node j in the t-th iteration. Let α represent the heuristic information from node i to node j, and β be the expected heuristic factor. These represent the influence of pheromone concentration and the magnitude of the heuristic information on the probability of the ant choosing different paths, respectively. When planning the optimal path, the heuristic factor is the reciprocal of the distance between the two nodes, as shown in the following formula: .

[0052] .

[0053] Let be the coordinates of the ant at node i. Let be the coordinates of the ant at node j.

[0054] In some embodiments, the method for calculating q(t) in step (3) is as follows: .

[0055] The initial minimum threshold can be, but is not limited to, 0. The initial maximum threshold can be, but is not limited to, 0.7. The decay constant can be, but is not limited to, 20. The probability threshold q(t) is the threshold probability used for the mixed selection in the t-th iteration.

[0056] When using the ant colony algorithm to search for paths, in some embodiments, to avoid ants backtracking or going in circles, each ant constructs a tabu list in step (3). During each iteration, the nodes that the ant has already visited are stored in the tabu list, and subsequent searches treat nodes in the list as unreachable. When constructing the node set navi, nodes that are also in the tabu list are deleted, which can avoid duplicate nodes that have already been reached and can also improve the convergence speed. The tabu list is cleared after each iteration.

[0057] In some embodiments, the pheromone concentrations from node i to node j are updated after the t-th iteration is completed: .

[0058] .

[0059] This refers to the total amount of pheromone released by the ant colony on path (i,j) in the t-th iteration. If the k-th ant fails to reach the destination, the pheromone concentration on that path is not updated. It equals 0. The pheromone evaporation factor represents the rate at which pheromones evaporate over time, and is typically set between 0 and 1. This refers to the pheromone residue factor. m refers to the total number of ants that successfully reached the endpoint in this iteration.

[0060] The pheromone concentration is updated globally: in each iteration, the pheromone concentration for the corresponding path is updated only after all artificial ants have completed their path search. Each ant will search a path completely, even if the pheromone concentration is high in some sections along the way, until the end.

[0061] In some embodiments, The calculation method is as follows: .

[0062] Where L equals the length of the search path of the k-th ant in this iteration, and Q is the pheromone intensity coefficient. The pheromone increment depends only on the path length and is independent of the specific path.

[0063] In some embodiments, in step (4), if the path distance planned in several consecutive iterations does not change, it is considered that the termination condition is met and the iteration is stopped.

[0064] After obtaining the optimal path, the route distance is estimated by accumulating the great circle distances between nodes using the Havesing formula, as shown in the following equation: .

[0065] Where R is the Earth's radius, taken as 6371 km. and These represent latitude and longitude (in radians), respectively. After obtaining the estimated route distance, based on the actual situation, low, medium, and high speeds are assigned 7 km, 10 km, and 13 km, respectively, to simplify the estimated travel time.

[0066] This embodiment uses data from the 2022 summer navigation season (July-September) to construct a risk assessment framework coupled with the POLARIS system. It comprehensively considers sea ice thickness, sea ice concentration, wind speed, air temperature, ocean current speed parameters, and the icebreaking capabilities of vessels of different icebreaking classes (PC1-1C). The Arctic Ocean is divided into navigable and non-navigable areas. Based on these regional constraints, an improved ant colony algorithm is introduced to plan routes for the Arctic shipping routes. The simulated starting points for the Northeast Passage and Northwest Passage are Murmansk and Ilulissat, respectively, with the endpoint being the Bering Strait. This embodiment compares the results of different algorithms, as shown in Table 1. As shown in Table 1, the improved ant colony algorithm significantly outperforms the traditional ant colony algorithm in both convergence efficiency and path quality. It reduces path redundancy points and iteration counts, shortening the Northeast Passage route by 12.64% and the Northwest Passage route by 13.67%, corresponding to a reduction in travel time of 2-3 days. Compared to the A* algorithm, the improved algorithm does not significantly shorten the route length, with differences within 8km. However, the A* algorithm's route closely follows navigable areas, with a minimum safe distance of less than 12.5km. In terms of path quality, the improved algorithm achieves better safety and path quality while shortening the voyage.

[0067] Table 1: Comparison of Improved Ant Colony Algorithm with Existing Algorithms

[0068] The effectiveness and feasibility of the proposed method can be verified by applying the improved algorithm to the route planning of ten different ice classes, providing a theoretical basis and practical reference for the safe, efficient and sustainable operation of the Arctic shipping route.

[0069] Of course, the above description is not intended to limit the present invention, and the present invention is not limited to the examples given above. Any changes, modifications, additions or substitutions made by those skilled in the art within the scope of the present invention should also fall within the protection scope of the present invention.

Claims

1. A method for Arctic shipping route planning based on an improved ant colony algorithm, characterized in that, include: (1) Calculate the navigable area of ​​the icebreaker in the Arctic region by combining ice condition data, meteorological data and the icebreaker's class; (2) Initialize the pheromone concentration for valid nodes in the navigable area, wherein the valid nodes are those marked as navigable, including: (21) Use a heuristic search algorithm to search for the initial optimal path between the start point and the end point from the effective nodes. All nodes on the initial optimal path form a node set Pi. Find effective nodes located near the initial optimal path. The effective nodes near Pi together form a node set Pn. (22) Set the initial pheromone concentration of nodes in Pn to be higher than the initial pheromone concentration of other effective nodes; (3) Find the set of nodes navi that the k-th ant can move to next from the current node i at time t. Calculate the probability that the ant will move from node i to node j in navi based on the pheromone concentration. ; A random number r is generated in the range [0,1]. The random number r is compared with the probability threshold q(t). When r is not greater than q(t), a roulette wheel selection mechanism is used to randomly select the next node based on the probability distribution of the k-th ant moving from node i to each node in navi. When r is greater than q(t), the node with the highest probability is directly selected as the next node. This process continues until the endpoint is reached, completing one iteration. (4) After completing one iteration, update the pheromone concentration on the path from node i to node j, and proceed to the next iteration until the termination condition is met.

2. The Arctic shipping route planning method according to claim 1, characterized in that, The ice condition data includes ice condition risk value and ice concentration, and the meteorological data includes temperature, wind speed and ocean current speed. In step (1), the method for calculating the navigable area of ​​the icebreaker in the Arctic region includes: (11) Calculate the seaworthiness index RIO of the icebreaker at each grid point, and generate the navigable environment grid of the icebreaker, including: Based on the icebreaker's class, obtain the ice condition risk value (RIV) corresponding to the icebreaker in the grid map, and obtain the ice concentration (C) of each grid point in the grid map; Calculate the seaworthiness index (RIO) of this class of icebreaker at each grid point on the grid map: RIO = C × RIV; (12) Generate additional risk constraint factors based on the meteorological data. : ； (13) Navigable area By the RIO and Common constraints: ； in, A node is considered valid when its value is 1.

3. The Arctic shipping route planning method according to claim 2, characterized in that, The search method for the initial optimal path in step (21) includes: (211) From the valid nodes in the navigable area, identify all candidate nodes that can be transferred from the starting point to each next step; (212) Calculate the distance estimate F(n) from the current node to the destination through each candidate node. t ): F(n t )=g(n t )+h(n t ); g(n t ) represents the distance from the starting point to the candidate node n t The actual distance value, h(n) t ) indicates starting from candidate node n t Estimated distance to the destination; (212) The candidate node with the smallest distance estimate is selected as the target node for the next step. All target nodes form the initial optimal path.

4. The Arctic shipping route planning method according to claim 3, characterized in that, h(n t The calculation method for ) is as follows: ； For candidate node n t coordinates The coordinates of the endpoint.

5. The Arctic shipping route planning method according to claim 1, characterized in that, In step (3) The calculation method is as follows: ； in, This represents the pheromone concentration from node i to node j in the t-th iteration. Let α represent the heuristic information from node i to node j, where α is the pheromone heuristic factor and β is the expected heuristic factor. ； ； Let be the coordinates of the ant at node i. Let be the coordinates of the ant at node j.

6. The Arctic shipping route planning method according to claim 1, characterized in that, The method for calculating q(t) in step (3) is as follows: ； As the initial minimum threshold, The initial maximum threshold, is the attenuation constant.

7. The Arctic shipping route planning method according to any one of claims 1-6, characterized in that, In step (3), each ant constructs a tabu list. During each iteration, the nodes it has visited are stored in the tabu list. When constructing the node set navi, the nodes that are simultaneously in the tabu list are deleted. After each iteration, the tabu list is cleared.

8. The Arctic shipping route planning method according to any one of claims 1-6, characterized in that, After the t-th iteration is completed, update the pheromone concentrations from node i to node j: ； ； This refers to the total amount of pheromone released by the ant colony on path (i,j) in the t-th iteration. is the pheromone evaporation factor, and m refers to the total number of ants that successfully reach the endpoint in this iteration.

9. The Arctic shipping route planning method according to claim 8, characterized in that, The calculation method is as follows: ； Where L equals the length of the search path of the k-th ant in this iteration, and Q is the pheromone intensity coefficient.

10. The Arctic shipping route planning method according to any one of claims 1-6, characterized in that, In step (4), if the path distance of the planned path does not change in several consecutive iterations, the termination condition is considered met and the iteration is stopped.

Images