Navigation path planning method and device, electronic equipment and storage medium
By combining the improved FMM and DWA algorithms with the PINN model, the problems of global path optimality and high local path prediction accuracy in complex ocean environments are solved, and efficient and safe navigation in dynamic environments is achieved.
Patent Information
- Application Number
- CN202510692988.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-05-27
- Publication Date
- 2025-09-12
AI Technical Summary
Existing path planning technologies find it difficult to simultaneously balance the optimality of the global path and the high prediction accuracy of the local path in complex ocean environments. Especially when dynamic obstacles frequently appear and environmental information is highly uncertain, navigation safety and efficiency are difficult to guarantee.
Combining the improved fast marching method (FMM) and dynamic window algorithm (DWA), by introducing the physical information neural network (PINN) model based on the dynamic characteristics of the unmanned ship and the environmental factors of the target sea area, the global and local navigation paths are optimized, the trajectory is adjusted in real time to cope with complex environments, and the flexibility and smoothness of path planning are improved.
The system significantly improves the ability to handle high-density obstacle areas and multi-ship interactions in complex marine environments, ensuring that the vehicle can quickly avoid obstacles and sudden environmental changes, maintain a stable navigation state, generate accurate local trajectories to reflect the dynamic characteristics of the hull, and adapt to dynamic environmental requirements.
Smart Images
Figure CN120628094A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of autonomous surface ship path planning, and in particular to a navigation path planning method, device, electronic equipment and storage medium. Background Art
[0002] Efficient navigation of autonomous surface vessels in complex ocean environments is one of the current research hotspots. However, existing path planning technologies struggle to strike a balance between global planning and local obstacle avoidance. Single algorithms often struggle to cope with dynamic obstacles, complex terrain, and changing navigation conditions due to their inherent limitations. While the traditional FMM method can quickly generate a globally optimal path, it lacks the ability to adapt to dynamic environments and exhibits poor path smoothness. While the DWA algorithm excels in dynamic obstacle avoidance, it is limited to local path planning and struggles to optimize the overall route. Furthermore, traditional DWA relies on a simplified motion model, ignoring the impact of environmental disturbances (such as wind speed, waves, and currents) on ship motion in complex ocean environments, resulting in insufficient path prediction accuracy.
[0003] Traditional trajectory planning algorithms often struggle to balance global path optimization with the need to avoid local dynamic environments when faced with complex ocean environments. Furthermore, these algorithms suffer from insufficient local path prediction accuracy in practical applications. This makes it difficult to ensure navigation safety and efficiency, especially in complex scenarios with frequent dynamic obstacles and highly uncertain environmental information. Summary of the Invention
[0004] In view of this, it is necessary to provide a navigation path planning method, device, electronic device and storage medium to solve the technical problem that the existing technology is difficult to simultaneously take into account the optimality of the global path and the high prediction accuracy of the local path.
[0005] In order to achieve the above technical effects, in a first aspect, the present invention provides a navigation path planning method, comprising: The global navigation path of the unmanned vessel is planned based on an improved FMM algorithm and a cost field of the target sea area, wherein the improved FMM algorithm includes: increasing the search directions of the traditional FMM algorithm from four to eight and updating the weights of each search direction and the cost values of the searched nodes in real time based on path continuity, target orientation, and environmental factors of the target sea area; During the navigation of the unmanned ship along the global navigation path, the local navigation path of the unmanned ship is optimized based on an improved DWA algorithm and a PINN model constructed based on the dynamic characteristics of the unmanned ship and the environmental factors of the target sea area. The improved DWA algorithm includes setting the control input sampling granularity based on the dynamic characteristics of the unmanned ship.
[0006] In some embodiments of the present invention, the target sea area includes: a non-navigation area, a solid obstacle area, and a dynamic obstacle area; The cost field of the target sea area is generated by introducing a differentiated risk assessment strategy into the target sea area.
[0007] In some embodiments of the present invention, the global navigation path planning of the unmanned vessel based on the improved FMM algorithm and the cost field of the target sea area includes: Adjust the search weights in each direction in real time based on path continuity, target orientation, and environmental factors of the target sea area; Searching for a plurality of nodes to be evaluated in each search direction of the cost field of the target sea area based on the search weight, and storing the plurality of nodes to be evaluated in an open list; The node with the smallest cost value in the open list is used as the current node; Based on path smoothness and goal orientation, the basic cost values of all adjacent nodes of the current node are updated to obtain adjusted cost values; Mark the adjacent nodes whose base cost value is not less than the adjusted cost value as candidate nodes; Loading the candidate nodes into an open list; A plurality of candidate paths are generated based on the nodes in the open list, and the candidate path with the smallest total cost is used as the global navigation path.
[0008] In some embodiments of the present invention, before optimizing the local navigation path of the unmanned vessel based on the improved DWA algorithm and the PINN model constructed based on the dynamic characteristics of the unmanned vessel and the environmental factors of the target sea area, the method further includes: The global navigation path is smoothed based on a parameterized interpolation method.
[0009] In some embodiments of the present invention, the optimization of the local navigation path of the unmanned vessel based on the improved DWA algorithm and the PINN model constructed based on the dynamic characteristics of the unmanned vessel and the environmental factors of the target sea area includes: Sampling the value range of the control input quantity based on the improved DWA algorithm to obtain several groups of candidate control input quantities; Predict the position of moving objects within a preset range centered on the unmanned vessel based on a linear uniform velocity model; Inputting the position of the active object and several groups of candidate control inputs into the PINN model to obtain several local paths; Calculating an evaluation function value of the local path based on the path deviation, target distance, steering penalty, and obstacle avoidance penalty of the local path; The candidate control input corresponding to the local path with the smallest evaluation function value is determined as the optimal control input, and the local path optimization is completed.
[0010] In some embodiments of the present invention, after optimizing the local navigation path of the unmanned vessel based on the improved DWA algorithm and the PINN model constructed based on the dynamic characteristics of the unmanned vessel and the environmental factors of the target sea area, the method further includes: Obtaining environmental information and obstacle information of the target sea area; If the environmental information and obstacle information exceed the preset threshold, the global navigation path of the unmanned ship is replanned based on the improved FMM algorithm and the cost field of the target sea area.
[0011] In a second aspect, the present invention further provides a navigation path planning device, comprising: A global path planning module is used to plan the global navigation path of the unmanned vessel based on an improved FMM algorithm and a cost field of the target sea area. The improved FMM algorithm includes: adding exploration directions to the traditional FMM algorithm and updating the weights of each search direction and the cost values of the searched nodes in real time based on path continuity, target orientation, and environmental factors of the target sea area; A local path optimization module is used to optimize the local navigation path of the unmanned ship based on an improved DWA algorithm and a PINN model constructed based on the dynamic characteristics of the unmanned ship and the environmental factors of the target sea area while the unmanned ship is navigating along the global navigation path. The improved DWA algorithm includes: setting the control input sampling granularity based on the dynamic characteristics of the unmanned ship.
[0012] In a third aspect, the present invention further provides an electronic device, comprising: Memory, used to store programs; A processor, coupled to the memory, is used to execute the program stored in the memory to implement the steps in the navigation path planning method described in any one of the above method items.
[0013] In a fourth aspect, the present invention further provides a storage medium, comprising: Used to store computer-readable programs or instructions, which, when executed by a processor, can implement the steps of the navigation path planning method described in any one of the above method items.
[0014] The beneficial effects of the present invention are as follows: the navigation path planning method provided by the present invention integrates the fast marching method (FMM) and the dynamic windowing method (DWA), comprehensively utilizing the advantages of FMM in global trajectory generation and the advantages of DWA in real-time adjustment of local trajectories. By introducing the PINN model, the method significantly improves the ability to effectively handle high-density obstacle areas, multi-ship interactions, and uncertain environmental factors in complex marine environments. In addition, the present invention also improves the FMM algorithm and the DWA algorithm by adding movement directions, effectively improving the flexibility and smoothness of global navigation trajectory planning; through real-time trajectory updates and dynamic adjustment mechanisms, the vehicle can quickly avoid dynamically appearing obstacles and sudden environmental changes, maintaining a stable navigation state; by setting a finer control input sampling granularity based on the dynamic characteristics of the ship, it ensures that the generated local trajectory can accurately reflect the actual dynamic characteristics of the hull and can effectively respond to navigation needs in a dynamic environment in real time, effectively solving the technical problem that the existing technology is difficult to simultaneously consider the optimality of the global path and the high prediction accuracy of the local path. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.
[0016] Figure 1 A schematic diagram of a flow chart of an embodiment of a navigation path planning method provided by the present invention; Figure 2 for Figure 1 A flow chart of an embodiment of step S101; Figure 3 for Figure 1 A flow chart of an embodiment of step S102; Figure 4 A structural diagram of an embodiment of a navigation path planning device provided by the present invention Figure 5 This is a structural diagram of an embodiment of an electronic device provided by the present invention. DETAILED DESCRIPTION
[0017] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with 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 making any creative efforts shall fall within the scope of protection of the present invention.
[0018] In the description of the embodiments of the present invention, unless otherwise specified, "plurality" means two or more. "And / or" describes the association relationship between associated objects, indicating that three relationships can exist. For example, "A and / or B" can mean: A exists alone, A and B exist simultaneously, or B exists alone.
[0019] The terms "first," "second," and so on, used in the embodiments of the present invention are for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, technical features designated as "first" or "second" may explicitly or implicitly include at least one such feature.
[0020] References herein to "embodiments" mean that a particular feature, structure, or characteristic described in connection with the embodiments may be included in at least one embodiment of the present invention. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute a separate or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.
[0021] Before presenting the embodiments of the present invention, the following concepts are first explained: (1) Fast Matching Method (FMM): FMM is widely used in ship path planning to help ships find the best route in complex ocean environments. It discretizes the navigation area into grids and uses the discretized velocity field information to calculate the shortest path from the starting point to the target point.
[0022] (2) Dynamic Window Approach (DWA): An algorithm for local path planning of mobile robots, mainly used for obstacle avoidance and path planning of robots in dynamic environments. Its core idea is to dynamically generate a speed window in the current speed space, evaluate the safety, target orientation, and speed efficiency of feasible trajectories, and ultimately select the optimal motion instructions. The DWA method dynamically generates a speed window in the current speed space, considers the robot's maximum and minimum speeds, maximum acceleration and deceleration capabilities, and the position and speed of obstacles, evaluates the trajectory under different speed combinations, and selects the optimal speed and angular velocity combination, so that the robot can safely and efficiently avoid obstacles and reach the target position.
[0023] The present invention provides a navigation path planning method, device, electronic device and storage medium, which are described below respectively.
[0024] Figure 1 A schematic flow chart of an embodiment of the path planning method provided by the present invention is shown as follows: Figure 1 As shown, the navigation path planning method includes: S101. Plan the global navigation path of the unmanned ship based on the improved FMM algorithm and the cost field of the target sea area.
[0025] It should be noted that the improved FMM algorithm includes: increasing the search directions of the traditional FMM algorithm from four to eight and updating the weights of each search direction and the cost values of the searched nodes in real time based on path continuity, target orientation and environmental factors of the target sea area.
[0026] Preferably, the present invention adds diagonal movement directions, totaling eight search directions, making trajectory planning more flexible and smooth.
[0027] It should be noted that in order to further optimize trajectory quality, the present invention proposes a dynamic weight allocation mechanism that comprehensively considers environmental factors, trajectory continuity, and goal orientation. By dynamically adjusting the search weights in each direction in real time, the trajectory generation process is refined, effectively avoiding trajectory discontinuity and oversteering problems, and ensuring the efficiency and safety of the planned path.
[0028] In some embodiments of the present invention, the target sea area includes: a non-navigation area, a solid obstacle area, and a dynamic obstacle area; the cost field of the target sea area is generated by introducing a differentiated risk assessment strategy for the target sea area.
[0029] Preferably, for clear non-navigation areas, extremely high assessment values can be assigned to clearly indicate that they are inaccessible; for fixed obstacle areas, a distance-related attenuation model is used to calculate the environmental impact; and for movable obstacle areas, a multi-level superimposed attenuation model is used for simulation and real-time update, thereby forming a dynamic and accurate environmental risk assessment system, effectively avoiding potential navigation hazards.
[0030] It should be noted that in order to further ensure navigation safety, the safety buffer zone of the environmental impact area can be expanded to increase the safety margin during navigation, reduce the possibility of collisions and dangerous incidents, and greatly improve the safety and reliability of trajectory planning.
[0031] like Figure 2 In some embodiments of the present invention, step S101 includes: S201, adjusting the search weights of each direction in real time based on path continuity, target orientation, and environmental factors of the target sea area; S202, searching for a plurality of nodes to be evaluated in each search direction of the cost field of the target sea area based on the search weight, and storing the plurality of nodes to be evaluated in an open list; S203: select the node with the smallest cost in the open list as the current node; S204: Update the basic cost values of all adjacent nodes of the current node based on the path smoothness and goal orientation to obtain an adjusted cost value.
[0032] Specifically, for each neighboring node of the current node, the smoothness of the path segment formed by the neighboring node, the previous node, and the current node is calculated. Path smoothness can be measured by calculating the change in angle between the neighboring node and the previous and current nodes. For example, the angle between the neighboring node and the previous and current nodes can be calculated using the angle formula between vectors, and then multiplied by the path smoothness weight factor to calculate the smoothness cost.
[0033] For the goal-oriented cost of the adjacent nodes, the Euclidean distance or Manhattan distance from the adjacent nodes to the target node can be calculated, and then multiplied by the goal-oriented weight factor to obtain the goal-oriented cost.
[0034] When updating the base cost value of the adjacent node, the smoothness cost and the goal-oriented cost can be added together, and the movement cost from the current node to the adjacent node (such as Euclidean distance, Manhattan distance, etc.) can be added to obtain the updated base cost value of the adjacent node.
[0035] S205, marking the adjacent nodes whose basic cost value is not less than the adjusted cost value as candidate nodes; S206, loading the candidate node into the open list; S207: Generate several candidate paths based on the nodes in the open list, and use the candidate path with the smallest total cost as the global navigation path.
[0036] It's important to note that during trajectory planning, the system continuously updates the cost of each location. When a new location's cost is found to be superior to that of an adjacent location, the cost estimate for that location is automatically updated and included in the next evaluation. This continuous updating and iteration ensures that trajectory planning can always adapt to dynamically changing environmental conditions, continuously optimizing the generated trajectory and ensuring path safety, continuity, and efficiency.
[0037] Although the fast marching method can theoretically provide the optimal path, the actual generated path often contains a large number of sharp turns, which affects actual controllability and safety. In some embodiments of the present invention, before step S102, the following is also included: The global navigation path is smoothed based on the parameterized interpolation method.
[0038] It can be understood that the present invention uses an advanced parametric interpolation method to smooth the generated discrete path. By optimizing the continuity parameters, it significantly reduces the sharp turning points in the path, making the trajectory smoother, continuous and in line with actual control requirements, greatly improving the safety and operational efficiency of actual navigation.
[0039] S102. While the unmanned ship is navigating along the global navigation path, the local navigation path of the unmanned ship is optimized based on the improved DWA algorithm and the PINN model constructed based on the dynamic characteristics of the unmanned ship and the environmental factors of the target sea area.
[0040] It should be noted that the improved DWA algorithm includes: setting the control input sampling granularity based on the dynamic characteristics of the unmanned vessel.
[0041] It should also be noted that this invention introduces the PINN model to accurately predict the future trajectory of each set of control inputs. The PINN model integrates the ship's dynamic characteristics and real-time environmental data to achieve highly accurate trajectory prediction. A comprehensive evaluation mechanism is also proposed to assess the consistency of candidate trajectories with the expected trajectory, their proximity to the target position, the safety of the track, and the stability of the course, ensuring the optimization and efficiency of the selected control strategy.
[0042] Specifically, during the PINN model training process, an efficient and accurate loss function system can be constructed by integrating data-driven methods and physical law constraints. During the training process, the matching degree of actual observation data, the satisfaction of the hull dynamics equation constraints and the precise matching of the initial state are carefully considered to ensure that the model can accurately predict the future state of the ship, greatly improving the reliability and stability of the navigation trajectory prediction.
[0043] like Figure 3 In some embodiments of the present invention, step S102 includes: S301 : Sampling the value range of the control input quantity based on the improved DWA algorithm to obtain several groups of candidate control input quantities.
[0044] S302: Predict the positions of moving objects within a preset range centered on the unmanned boat based on a linear uniform velocity model.
[0045] It should be noted that using a linear uniform velocity model to predict the future position of moving objects can help prevent potential collisions during navigation. The efficient application of this prediction model significantly improves navigation safety and the timeliness of local trajectory adjustments, ensuring safe navigation of ships in complex environments.
[0046] S303: Input the position of the active object and several groups of candidate control inputs into the PINN model to obtain several local paths.
[0047] S304 : Calculate the evaluation function value of the local path based on the path deviation, target distance, steering penalty, and obstacle avoidance penalty of the local path.
[0048] Specifically, weight coefficients are first set for path deviation, target distance, steering penalty, and obstacle avoidance penalty. These weights are then adjusted based on actual needs to balance the impact of each factor on path evaluation. Then, an evaluation function formula is established, for example: Evaluation Function Value = Path Deviation Weight × Path Deviation Value + Target Distance Weight × Target Distance Value + Steering Penalty Weight × Steering Penalty + Obstacle Avoidance Penalty Weight × Obstacle Avoidance Penalty Value.
[0049] When calculating the path deviation value, the deviation distance between each point on the local path and the original planned path can be obtained, and the average or maximum value can be calculated as the path deviation value. When calculating the target distance value, the Euclidean distance or Manhattan distance from the end point of the local path to the target point can be calculated. When calculating the turning penalty value, the number of consecutive turns or the sum of angle changes in the local path can be counted as the turning penalty value. When calculating the obstacle avoidance penalty value, the distance between each point on the local path and the obstacle can be detected. If the distance is less than the safety threshold, the penalty value is calculated based on the inverse of the distance or other functional relationship, and the obstacle avoidance penalty value is accumulated. When calculating the evaluation function value, the value of each factor and its corresponding weight coefficient can be substituted into the preset evaluation function formula to calculate the evaluation function value of the local path.
[0050] S305 : Determine the candidate control input corresponding to the local path with the smallest evaluation function value as the optimal control input, and complete the local path optimization.
[0051] It is understandable that after comprehensive evaluation of all candidate control input solutions, the present invention automatically selects the best control input and implements real-time trajectory tracking and local optimization.
[0052] It should be noted that, in this embodiment, when all candidate solutions cannot meet the safety requirements, it can automatically switch to a safe low-speed, low-risk mode to ensure navigation safety.
[0053] Compared to existing technologies, the present invention provides a navigation path planning method that integrates the Fast Marching Method (FMM) and the Dynamic Windowing Approach (DWA). This method leverages the advantages of the FMM in global trajectory generation and the DWA in real-time adjustment of local trajectories. By introducing the PINN model, it significantly improves the ability to effectively handle high-density obstacle areas, multi-vessel interactions, and uncertain environmental factors in complex ocean environments. Furthermore, the present invention improves upon the FMM and DWA algorithms by adding movement directions, effectively enhancing the flexibility and smoothness of navigation trajectory planning. Through real-time trajectory updates and dynamic adjustment mechanisms, the vessel can quickly avoid dynamically emerging obstacles and sudden environmental changes, maintaining a stable navigation state. By setting a finer control input sampling granularity based on the vessel's dynamic characteristics, the generated local trajectory accurately reflects the vessel's actual dynamic characteristics and effectively addresses navigation needs in a dynamic environment in real time. It has greatly improved the safe navigation capabilities of autonomous surface vehicles in complex marine environments. It is suitable for various mission scenarios such as maritime transportation, marine resource exploration, maritime patrol and law enforcement, search and rescue operations, and scientific research. It meets the actual needs of the increasingly complex current marine activities and effectively solves the technical problem that existing technologies are difficult to simultaneously take into account the optimality of the global path and the high prediction accuracy of the local path.
[0054] In addition, in order to improve the adaptability of the unmanned ship to complex environmental changes, in some embodiments of the present invention, step S102 further includes: Obtain environmental information and obstacle information of the target sea area; If the environmental information and obstacle information exceed the preset threshold, the global navigation path of the unmanned ship is replanned based on the improved FMM algorithm and the cost field of the target sea area.
[0055] Specifically, when environmental conditions change significantly (such as when wind and waves become significantly stronger) or encounter unexpected situations (such as the sudden appearance of a large number of obstacles), the present invention automatically triggers the overall trajectory replanning mechanism, adjusts and optimizes the trajectory in real time, keeps the trajectory in the optimal state at all times, effectively responds to complex environmental changes, and further improves the overall safety of navigation.
[0056] In summary, the present invention successfully overcomes the shortcomings of traditional navigation trajectory planning methods in complex environments by innovatively integrating the improved fast marching method, dynamic window method and physical information neural network prediction model, and comprehensively improves the accuracy, stability, real-time and adaptability of navigation trajectory planning. It has outstanding technical advantages and engineering application value.
[0057] like Figure 4 In a second aspect, the present invention further provides a navigation path planning device 40, comprising: A global path planning module 410 is used to plan the global navigation path of the unmanned vessel based on an improved FMM algorithm and a cost field of the target sea area. The improved FMM algorithm includes: adding exploration directions to the traditional FMM algorithm and updating the weights of each search direction and the cost values of the searched nodes in real time based on path continuity, target orientation, and environmental factors of the target sea area; The local path optimization module 420 is used to optimize the local navigation path of the unmanned vessel based on the improved DWA algorithm and the PINN model constructed based on the dynamic characteristics of the unmanned vessel and the environmental factors of the target sea area while the unmanned vessel is navigating along the global navigation path. The improved DWA algorithm includes: setting the control input sampling granularity based on the dynamic characteristics of the unmanned vessel.
[0058] like Figure 5 In a third aspect, the present invention further provides an electronic device 50, comprising: Memory 510, used for storing programs; The processor 520 is coupled to the memory 510 and is used to execute the program stored in the memory 510 to implement the steps in the navigation path planning method described in any one of the above method items.
[0059] In a fourth aspect, the present invention further provides a storage medium, comprising: Used to store computer-readable programs or instructions, which, when executed by a processor, can implement the steps of the navigation path planning method described in any one of the above method items.
[0060] The above is a detailed introduction to the navigation path planning method, device, electronic device and storage medium provided by the present invention. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core ideas. At the same time, for those skilled in the art, according to the ideas of the present invention, there may be changes in the specific implementation methods and application scopes. In summary, the content of this specification should not be understood as limiting the present invention.
Claims
1. A navigation path planning method, characterized in that: include: The global navigation path of the unmanned vessel is planned based on an improved FMM algorithm and a cost field of the target sea area, wherein the improved FMM algorithm includes: increasing the search directions of the traditional FMM algorithm from four to eight and updating the weights of each search direction and the cost values of the searched nodes in real time based on path continuity, target orientation, and environmental factors of the target sea area; During the navigation of the unmanned ship along the global navigation path, the local navigation path of the unmanned ship is optimized based on an improved DWA algorithm and a PINN model constructed based on the dynamic characteristics of the unmanned ship and the environmental factors of the target sea area. The improved DWA algorithm includes setting the control input sampling granularity based on the dynamic characteristics of the unmanned ship.
2. The navigation path planning method according to claim 1, characterized in that: The target sea areas include: non-navigation areas, solid obstacle areas and dynamic obstacle areas; The cost field of the target sea area is generated by introducing a differentiated risk assessment strategy into the target sea area.
3. The navigation path planning method according to claim 1, characterized in that: The global navigation path planning of the unmanned ship based on the improved FMM algorithm and the cost field of the target sea area includes: Adjust the search weights in each direction in real time based on path continuity, target orientation, and environmental factors of the target sea area; Searching for a plurality of nodes to be evaluated in each search direction of the cost field of the target sea area based on the search weight, and storing the plurality of nodes to be evaluated in an open list; The node with the smallest cost value in the open list is used as the current node; Based on path smoothness and goal orientation, the basic cost values of all adjacent nodes of the current node are updated to obtain adjusted cost values; Mark the adjacent nodes whose base cost value is not less than the adjusted cost value as candidate nodes; Loading the candidate nodes into an open list; A plurality of candidate paths are generated based on the nodes in the open list, and the candidate path with the smallest total cost is used as the global navigation path.
4. The navigation path planning method according to claim 1, characterized in that: Before optimizing the local navigation path of the unmanned ship based on the improved DWA algorithm and the PINN model constructed based on the dynamic characteristics of the unmanned ship and the environmental factors of the target sea area, the method further includes: The global navigation path is smoothed based on a parameterized interpolation method.
5. The navigation path planning method according to claim 4, characterized in that: The optimization of the local navigation path of the unmanned vessel based on the improved DWA algorithm and the PINN model constructed based on the dynamic characteristics of the unmanned vessel and the environmental factors of the target sea area includes: Sampling the value range of the control input quantity based on the improved DWA algorithm to obtain several groups of candidate control input quantities; Predict the position of moving objects within a preset range centered on the unmanned vessel based on a linear uniform velocity model; Inputting the position of the active object and several groups of candidate control inputs into the PINN model to obtain several local paths; Calculating an evaluation function value of the local path based on the path deviation, target distance, steering penalty, and obstacle avoidance penalty of the local path; The candidate control input corresponding to the local path with the smallest evaluation function value is determined as the optimal control input, and the local path optimization is completed.
6. The navigation path planning method according to claim 5, characterized in that: After optimizing the local navigation path of the unmanned vessel based on the improved DWA algorithm and the PINN model constructed based on the dynamic characteristics of the unmanned vessel and the environmental factors of the target sea area, the method further includes: Obtaining environmental information and obstacle information of the target sea area; If the environmental information and obstacle information exceed the preset threshold, the global navigation path of the unmanned ship is replanned based on the improved FMM algorithm and the cost field of the target sea area.
7. A navigation path planning device, characterized in that: include: A global path planning module is used to plan the global navigation path of the unmanned vessel based on an improved FMM algorithm and a cost field of the target sea area. The improved FMM algorithm includes: adding exploration directions to the traditional FMM algorithm and updating the weights of each search direction and the cost values of the searched nodes in real time based on path continuity, target orientation, and environmental factors of the target sea area; A local path optimization module is used to optimize the local navigation path of the unmanned ship based on an improved DWA algorithm and a PINN model constructed based on the dynamic characteristics of the unmanned ship and the environmental factors of the target sea area while the unmanned ship is navigating along the global navigation path. The improved DWA algorithm includes: setting the control input sampling granularity based on the dynamic characteristics of the unmanned ship.
8. An electronic device, characterized in that: include: Memory, used to store programs; A processor, coupled to the memory, is configured to execute the program stored in the memory to implement the steps of the navigation path planning method described in any one of claims 1 to 6.
9. A storage medium, characterized in that: include: Used to store computer-readable programs or instructions, which, when executed by a processor, can implement the steps of the navigation path planning method described in any one of claims 1 to 6.