Method and device for optimal energy path planning of unmanned surface vehicle under complex marine environment

Abstract

The application provides an energy optimal path planning method and device for unmanned surface vehicle in complex marine environment, and belongs to the technical field of path planning. The method provided by the application comprises the following steps: constructing a comprehensive energy consumption model; constructing a hybrid enhanced particle swarm optimization algorithm; constructing a safety space set and an adjacent graph through a grid method, generating an initial optimal path by using an A* algorithm, and performing bounded random disturbance and safety correction on target particle waypoints; performing smoothing processing on the iteratively generated path by using a B-spline technique, resampling the smoothed optimal solution, and re-injecting the resampled optimal solution into a population; adjusting an inertia weight and a learning factor based on the number of iterations, introducing a double guide factor to perform secondary adjustment on a cognitive term of particle velocity update; embedding the comprehensive energy consumption model as a fitness function into the hybrid enhanced particle swarm optimization algorithm, performing real-time energy consumption evaluation on each candidate path in the iteration process of the algorithm, updating an individual optimal path and a global optimal path, performing path smoothing optimization, and outputting an optimal sailing path of the unmanned surface vehicle.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of path planning, in particular to an energy optimal path planning method and device for an unmanned surface vehicle in a complex marine environment. BACKGROUND

[0002] Unmanned surface vehicles (USVs) have been widely used in underwater detection, maritime security, marine surveying and other fields due to their small size, low cost, high safety and high degree of autonomy. However, energy depletion is one of the core reasons for task failure and platform loss, and the limitations of the power system have become a key shortcoming restricting the long-term reliable operation of USVs in complex marine environments. Therefore, it is of great practical significance to carry out energy optimal path planning research in complex marine environments to improve task success rate, ensure asset safety and prolong endurance.

[0003] Currently, metaheuristic optimization algorithms are the mainstream technical means for USV path planning. Particle swarm optimization (PSO) is widely used due to its simple structure and ease of implementation. In order to solve the defects of traditional algorithms, various hybrid path planning frameworks that integrate different optimization paradigms have also emerged. By balancing the exploration and exploitation capabilities of the algorithm, the convergence efficiency and navigation performance of path planning are improved. However, existing research still has significant shortcomings: most schemes focus on minimizing the geometric length of the path and rely on simplified environmental models, without fully considering the complexity, dynamics and multi-physical field characteristics of real marine environments. Even though some research includes the effects of ocean currents, they do not comprehensively integrate the three key environmental factors of wind, wave and current, and there is a gap in the precise modeling of additional resistance and propulsion efficiency decay in severe sea conditions, leading to the possibility that the geometric shortest path in simulation may become the highest energy consumption route in actual application.

[0004] Therefore, there is an urgent need for a method that integrates high-fidelity environmental coupling models to solve the problem of balancing geometric optimality and energy efficiency, and to achieve autonomous and efficient energy-saving navigation of USVs in real marine environments. SUMMARY

[0005] Therefore, the present application provides an energy optimal path planning method and device for an unmanned surface vehicle in a complex marine environment, which integrates high-fidelity environmental coupling models to solve the problem of balancing geometric optimality and energy efficiency, and to achieve autonomous and efficient energy-saving navigation of USVs in real marine environments.

[0006] Specifically, the present application is implemented by the following technical solutions:

[0007] The first aspect of the present application provides an energy optimal path planning method for an unmanned surface vehicle in a complex marine environment, comprising:

[0008] construct a comprehensive energy consumption model coupling the space-time effects of wind, wave and current, the comprehensive energy consumption model comprehensively quantifying basic energy consumption of water resistance, wind-induced additional resistance energy consumption, wave-induced additional resistance energy consumption and control system energy consumption, wherein the current effect is integrated into the quantification process of various resistance energy consumptions through a coupling mechanism;

[0009] a hybrid enhanced particle swarm optimization algorithm is constructed by improving the traditional particle swarm optimization algorithm in multiple dimensions, wherein a safety space set and an adjacency graph are constructed by a grid method, an A* algorithm is used to generate an initial optimal path and to perform bounded random disturbance and safety correction on the target particle waypoint, a B-spline technique is used to smooth the path generated by iteration, and the smoothed optimal solution is resampled and re-injected into the population, the inertia weight and the learning factor are dynamically adjusted based on the number of iterations, and a double-guiding factor is introduced to make a secondary adjustment to the cognitive term of particle speed update;

[0010] the comprehensive energy consumption model is embedded into the hybrid enhanced particle swarm optimization algorithm as a fitness function, real-time energy consumption evaluation of each candidate path, individual and global optimal path update, and path smoothing optimization are performed in the algorithm iteration process, and an optimal navigation path of the unmanned surface vehicle is output.

[0011] The second aspect of the application provides an energy optimal path planning device for an unmanned surface vehicle in a complex marine environment, the device comprising a construction module, an improvement module and a determination module;

[0012] The construction module is configured to construct an energy consumption model coupling the space-time effects of wind, wave and current, the energy consumption model comprehensively quantifying basic energy consumption of water resistance, wind-induced additional resistance energy consumption, wave-induced additional resistance energy consumption and control system energy consumption, wherein the current effect is integrated into the quantification process of various resistance energy consumptions through a coupling mechanism.

[0013] The improvement module is configured to improve the traditional particle swarm optimization algorithm in multiple dimensions to construct a hybrid enhanced particle swarm optimization algorithm, wherein a safety space set and an adjacency graph are constructed by a grid method, an A* algorithm is used to generate an initial optimal path and to perform bounded random disturbance and safety correction on the target particle waypoint, a B-spline technique is used to smooth the path generated by iteration, and the smoothed optimal solution is resampled and re-injected into the population, the inertia weight and the learning factor are dynamically adjusted based on the number of iterations, and a double-guiding factor is introduced to make a secondary adjustment to the cognitive term of particle speed update.

[0014] The determination module is configured to embed the energy consumption model into the hybrid enhanced particle swarm optimization algorithm as a fitness function, to output an optimal navigation path of the unmanned surface vehicle through real-time energy consumption evaluation of each candidate path, individual and global optimal path update, and path smoothing optimization in the algorithm iteration process.

[0015] The application provides an energy optimal path planning method and device for an unmanned surface vehicle in a complex marine environment. In a first aspect, a comprehensive energy consumption model coupled with the space-time effects of wind, wave and current is constructed, and the current effect is integrated into various resistance energy consumption through a coupling mechanism. The vector relationship is used to associate the ground speed of the unmanned surface vehicle with the water speed, so that the current effect directly acts on the water speed and accurately affects the quantification results of the basic energy consumption of water resistance, the energy consumption of wind-induced additional resistance and the energy consumption of wave-induced additional resistance. At the same time, the energy consumption of the control system is also covered, which realizes comprehensive and accurate quantification of energy consumption in a complex marine environment, and avoids the energy consumption estimation deviation caused by the traditional model due to the neglect of environmental dynamic changes or the coupling effect of the current. In a second aspect, the traditional particle swarm optimization algorithm is improved in multiple dimensions. The safety space set and the adjacent graph are constructed by using the grid method, the initial path is generated by using the A* algorithm, and the heuristic initialization mode combining the bounded random disturbance and the safety correction is used. The safety of the initial population is ensured, and the population diversity is improved, which lays a foundation for the rapid convergence of the algorithm. The application of B-spline technology realizes the continuity of the path curvature and meets the navigation constraints. The dynamic adjustment of the inertia weight and the learning factor makes the algorithm maintain strong global search ability in the early stage and enhance local search accuracy in the later stage. The secondary adjustment of the double guide factor on the cognitive term further optimizes the particle search strategy, and effectively avoids the algorithm from falling into local optimum. In a third aspect, the accurately quantified comprehensive energy consumption model is embedded into the improved hybrid enhanced particle swarm optimization algorithm as a fitness function, so that the algorithm can select the optimal path based on the real and comprehensive energy consumption evaluation in the iteration process. At the same time, through path smoothing optimization, dynamic updating of individual and global optimal path, the cooperative optimization of energy consumption optimization, path safety and smoothness is realized. The finally output unmanned surface vehicle optimal navigation path not only meets the lowest energy consumption demand in the complex marine environment, but also has good navigation feasibility and obstacle avoidance safety, solving the problems of mismatch between geometric optimization and actual energy consumption optimization, poor adaptability to complex environment and algorithm easily falling into local optimum in the traditional path planning. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 A flowchart of the energy optimal path planning method for an unmanned surface vehicle in a complex marine environment provided by Embodiment One of the application is shown in the figure.

[0017] Figure 2 A structural schematic diagram of the energy optimal path planning device for an unmanned surface vehicle in a complex marine environment provided by Embodiment Two of the application is shown in the figure. DETAILED DESCRIPTION

[0018] The exemplary embodiments will be described in detail herein with reference to the drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application.

[0019] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting. As used in this application, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "and / or," as used herein, refers to and encompasses any or all possible combinations of one or more of the associated listed items.

[0020] It is to be understood that, although the terms first, second, third, etc. can be used herein to describe various information, these terms are not intended to denote a particular order or hierarchy. These terms are used merely for the purpose of distinguishing one type of information from another. For example, a first information can be termed a second information, and similarly, a second information can be termed a first information, without departing from the scope of the present application. As used herein, the term "if' can be construed to mean "when" or "upon" or "in response to determining" terms denoting the occurrence of an action.

[0021] The specific embodiments are given below to introduce the technical solutions of the present application in detail.

[0022] Figure 1 The flow chart of the energy optimal path planning method for unmanned surface vehicle in complex marine environment provided by Embodiment One of the present application is shown in FIG. 1. Please refer to Figure 1 The method provided by the present embodiment can include:

[0023] S101, constructing a comprehensive energy consumption model coupling the space-time effects of wind, wave and current.

[0024] The comprehensive energy consumption model quantitatively integrates the basic energy consumption of water resistance, the energy consumption of wind-induced additional resistance, the energy consumption of wave-induced additional resistance and the energy consumption of the control system, wherein the current effect is integrated into the quantification process of various resistance energy consumption through a coupling mechanism.

[0025] Specifically, the comprehensive energy consumption model is a mathematical model designed for the navigation scenario of unmanned surface vehicle (USV) and capable of accurately quantifying the total energy consumption in complex marine environment. The core feature is the coupling of the space-time dynamic effects of wind, wave and current, rather than the calculation of energy consumption based on static simplified conditions. The comprehensive energy consumption model is integrated by four core energy consumption components. The basic energy consumption of water resistance refers to the energy consumption generated by the basic friction resistance and pressure difference resistance of seawater to the hull during navigation, which is the core basic energy consumption of the hull navigation and directly related to the hull speed.

[0026] The wind-induced additional resistance energy consumption refers to the energy consumption corresponding to the additional resistance of the air (wind) to the hull during navigation, which is related to the relative speed of the hull and the wind and the wind direction, and is an additional energy consumption source that cannot be ignored in complex environment.

[0027] Wave-induced additional resistance energy consumption refers to the energy consumption corresponding to the additional resistance such as impact resistance and wave resistance generated by ocean waves on the hull, which is related to the wave height, wave frequency, wave direction and the sailing state of the hull, and is the key factor of energy consumption increase in severe sea conditions.

[0028] Control system energy consumption refers to the energy consumed by the unmanned surface vehicle to maintain stable heading, execute path correction, drive the rudder / propeller response and other control behaviors, including control module operation energy consumption and actuator action energy consumption, which is necessary for ensuring the stability of navigation.

[0029] Further, the space-time effect of wind, wave and current refers to the characteristics that the intensity, direction and action mode of the three marine environmental factors of wind, wave and current will change dynamically with time and space position, rather than fixed and static parameters. From the time dimension, the wind speed / direction of the wind will change with the weather system (such as gusts, sustained wind switching), the wave height / frequency of the wave will change with the tidal period and wind duration, and the flow speed / direction of the flow will change with the tidal rise and fall and ocean current movement; From the spatial dimension, the wind speed, wave height and flow speed are different in different navigation areas (such as the wave field characteristics of nearshore and open sea are different, and the flow field intensity of strait and open sea is different).

[0030] Current effect refers to the direct or indirect influence of ocean currents (including tidal currents, alongshore currents, ocean currents, etc.) on the energy consumption of unmanned surface vehicles, the core of which is that the current changes the speed of the boat against the water (speed against water = speed against the ground - current speed) through the vector relationship, and then directly affects the basic energy consumption of water resistance (water resistance is positively correlated with the square of the speed against water); The current changes the speed of the boat against the ground, and indirectly affects the relative speed of the boat against the wind (boat wind relative speed = ground wind speed - ground speed), and then affects the wind-induced additional resistance energy consumption.

[0031] In specific implementation, a comprehensive energy consumption model coupling wind, wave and current spatiotemporal effects is constructed, including: decomposing different frequency components of wind, wave and current through a spatiotemporal dependent function to establish a dynamic environmental field model; the dynamic environmental field model outputs spatiotemporal distribution parameters of sea current, wind field and wave field; based on the hydrodynamic parameters of the dynamic environmental field model, the relationship between the speed of the unmanned surface vehicle against the ground and the speed against water is established through a vector relationship, the energy consumption corresponding to the basic water resistance in the navigation process is quantified based on the speed against water, and a water resistance energy consumption model is constructed; based on the wind field parameters of the dynamic environmental field model, the wind speed and wind direction parameters at different spatiotemporal positions are extracted, the relative speed of the unmanned surface vehicle against the wind is calculated in combination with the speed against the ground obtained through the vector relationship, the energy consumption corresponding to the wind-induced additional resistance in the wind field environment is quantified based on the relative speed, and a wind-induced additional energy consumption model is constructed; based on the wave field parameters of the dynamic environmental field model, the wave height, wave frequency and wave direction parameters at different spatiotemporal positions in the dynamic environmental field are extracted, the energy consumption corresponding to the wave-induced additional resistance in the wave field environment is quantified using wave spectrum theory, and a wave-induced additional energy consumption model is constructed; the energy consumption in the process of heading control, track keeping and actuator response of the unmanned surface vehicle is integrated to construct a control system energy consumption; the water resistance energy consumption model, the wind-induced additional energy consumption model, the wave-induced additional energy consumption model and the control system energy consumption are summarized to form a comprehensive energy consumption model.

[0032] Specifically, first, the low-frequency deterministic component and the high-frequency random component of the wind, wave, and current are decomposed by a space-time dependent function, wherein the low-frequency component reflects large-scale predictable environmental mechanisms such as tidal periods and weather scale wind types, and the high-frequency component reflects unresolved small-scale environmental processes such as turbulent fluctuations. A dynamic environmental field model is established through this decomposition method, which can output the space-time distribution parameters of the current (flow rate, flow direction), wind field (wind speed, wind direction), and wave field (wave height, wave frequency, wave direction) corresponding to different space-time positions; then, based on the hydrodynamic parameters output by the dynamic environmental field model, the relationship between the speed of the unmanned surface vehicle relative to the water and the speed of the unmanned surface vehicle relative to the ground is established through the vector relationship of "speed of the water relative to the ground - speed of the current", and then combined with the ship type parameters (length, width, depth) of the unmanned surface vehicle, the energy consumption corresponding to the basic water resistance in the process of navigation is quantified by using the water resistance calculation formula, and then a water resistance energy consumption model is constructed; subsequently, based on the wind field parameters output by the dynamic environmental field model, the wind speed and wind direction parameters of each space-time node on the navigation trajectory of the unmanned surface vehicle are extracted, combined with the ground speed obtained through the aforementioned vector relationship (ground speed = water speed + current speed), the relative speed between the unmanned surface vehicle and the wind is calculated through the vector operation of "boat wind relative speed = ground wind speed - ground speed", and then according to the wind-induced additional resistance calculation method, the energy consumption corresponding to the wind-induced additional resistance in the wind field environment is quantified by combining the wind area of the hull and other parameters, and a wind-induced additional energy consumption model is constructed; at the same time, based on the wave field parameters output by the dynamic environmental field model, the effective wave height, peak wave frequency, and main wave direction of each space-time node on the navigation trajectory are extracted, a suitable wave spectrum model (such as the JONSWAP wave spectrum model) is selected according to the actual sea area sea state type, the extracted wave field parameters are taken as input to construct a wave spectrum matching the current space-time wave field characteristics, and then combined with the ship type parameters and navigation state parameters of the unmanned surface vehicle, the additional resistance force generated by different frequency waves on the hull is calculated through wave spectrum integration, and then the energy consumption corresponding to the wave-induced additional resistance in the wave field environment is quantified, and a wave-induced additional energy consumption model is constructed; the control behavior of the unmanned surface vehicle in the process of navigation is sorted out, the energy consumption of the rudder action in the heading control, the energy consumption of the attitude adjustment in the track keeping, and the energy consumption of the driving in the actuator response are counted, and the energy consumption data is integrated to construct the control system energy consumption; finally, the quantification results of the water resistance energy consumption model, the wind-induced additional energy consumption model, the wave-induced additional energy consumption model, and the control system energy consumption are summarized through summation operation to form a comprehensive energy consumption model that can comprehensively reflect the space-time coupling effect of wind, wave, and current. The water resistance calculation formula and the wind-induced additional resistance calculation formula can be referred to the description in the related technology, which will not be repeated here.

[0033] Optionally, the wave-induced additional energy consumption model is constructed, including: extracting wave field parameters corresponding to each space-time node on the unmanned surface vehicle track from the dynamic environment field model; the wave field parameters include effective wave height, peak wave frequency, and main wave direction; according to the sea state type of the actual sailing sea area, selecting an adaptive JONSWAP wave spectrum model, taking the effective wave height and the peak wave frequency as input parameters, and constructing a wave spectrum matched with the wave field characteristics of the current space-time position; based on the constructed JONSWAP wave spectrum, combining the ship type parameters and the sailing state parameters of the unmanned surface vehicle, and through spectrum integration, the additional resistance force generated by different frequency waves on the ship body is calculated; according to the additional resistance force, combining the sailing speed and the sailing time of the unmanned surface vehicle at the corresponding space-time node, the energy consumption corresponding to the wave-induced additional resistance at the corresponding space-time node is calculated; and the wave-induced additional energy consumption quantification results of all space-time nodes on the sailing track are summarized to form a wave-induced additional energy consumption model changing with space-time dynamically.

[0034] In a specific implementation, the wave field parameters corresponding to each space-time node on the preset sailing track of the unmanned surface vehicle are accurately extracted from the dynamic environment field model, including the significant wave height, peak wave frequency, and main wave direction. Abnormal fluctuation parameters caused by environmental disturbances are synchronously removed during the extraction process to ensure the effectiveness and accuracy of the input parameters. First, the sea state type of the actual sailing sea area of the unmanned surface vehicle is determined through sea monitoring data or sea state prediction information (emphasis on determining whether it is a limited wind area sea state). According to the sea state type, the appropriate JONSWAP spectrum model is selected. The filtered significant wave height and peak wave frequency are used as the core input parameters and substituted into the expression of the JONSWAP spectrum. By adjusting the peak enhancement factor and other parameters of the spectrum, a personalized spectrum that accurately matches the energy distribution and frequency characteristics of the current space-time position wave field is constructed. Based on the constructed JONSWAP spectrum, combined with the specific ship type parameters of the unmanned surface vehicle (including ship length, ship width, draft depth, and wet surface area of the hull), as well as the real-time sailing state parameters (including the angle between the water speed and the sailing direction and the main wave direction), the frequency range for spectrum integration is determined (covering the main frequency range of the actual sea area waves). Numerical integration methods such as trapezoidal integration and Simpson integration are used to integrate and calculate the interaction terms of wave energy and the hull at different frequencies, and the additional resistance components generated by waves of different frequencies on the hull are calculated. The total additional resistance force of the hull caused by waves is obtained through vector synthesis. According to the calculated wave additional resistance force, combined with the real-time sailing speed of the unmanned surface vehicle at the corresponding space-time node, the real-time power required to overcome the wave additional resistance at this node is calculated through the power calculation formula (power = resistance × speed). Combined with the sailing time of the unmanned surface vehicle at the space-time node (calculated from the path length and sailing speed), the energy consumption corresponding to the wave additional resistance at the corresponding space-time node is calculated through the energy calculation formula (energy = power × time). Finally, the wave-induced additional energy consumption of all space-time nodes on the sailing track of the unmanned surface vehicle is summarized by using the summation method, and the mapping relationship between energy consumption and space-time node coordinates is established to form a dynamic wave-induced additional energy consumption model that can reflect the energy consumption changes in different sailing positions and different time wave field environments in real time.

[0035] For example, in an embodiment, the water resistance is represented as:

[0036] ;

[0037] wherein, is the water resistance of the unmanned surface vehicle; , , is the water resistance coefficient; is the water speed of the unmanned surface vehicle.

[0038] The wind-induced additional resistance is represented as:

[0039] ;

[0040] wherein, is the wind-induced additional resistance of the unmanned surface vehicle; is the air density; is the wind-induced resistance coefficient; is the relative attack angle of the vehicle to the wind; is the windward area of the unmanned surface vehicle; is the relative velocity of the unmanned surface vehicle to the wind.

[0041] The calculation formula of is:

[0042] ;

[0043] wherein, is the relative velocity vector of the unmanned surface vehicle to the wind; is the ground speed vector of the wind; is the ground speed vector of the unmanned surface vehicle.

[0044] Since the accurate calculation of requires a large amount of wind tunnel data, in order to make the path planning algorithm have calculation feasibility, the present application does not directly calculate , but uses a parameterized expression to quantify the energy consumption generated thereby :

[0045] ;

[0046] wherein, is the wind-induced additional energy consumption of the unmanned surface vehicle; is the energy conversion efficiency of the propulsion system of the unmanned surface vehicle; is the air density; is the wind-induced coefficient; is the windward area of the unmanned surface vehicle; is the included angle between the wind and the sailing direction of the unmanned surface vehicle; is the reference wind speed; is the sailing speed of the unmanned surface vehicle; is the sailing time of the unmanned surface vehicle in the corresponding wind field environment.

[0047] The wave-induced additional resistance is expressed as:

[0048] ;

[0049] ;

[0050] ;

[0051] wherein, is the effective wave angular frequency; is the peak wave angular frequency; is the sailing speed of the USV; is the acceleration of gravity; is the effective wave height; is the wave-induced added resistance increment; is the seawater density; is the length of the USV; is the angle between the wind and the sailing direction of the USV.

[0052] The above formula shows that the added resistance is approximately proportional to and inversely proportional to This proportional relationship is applicable to most ship types. In addition, to capture the wave motion-induced loss of propulsion efficiency, the model is established as follows:

[0053] ;

[0054] ;

[0055] wherein, is the energy conversion efficiency of the USV propulsion system; is the base efficiency of the propulsion system; is the wave-induced efficiency attenuation coefficient; is the effective wave height; is the wave-induced added energy consumption; is the wave-induced added resistance increment; is the sailing speed of the USV; is the sailing time of the USV in the corresponding wind field environment.

[0056] In addition to changing the water speed through indirect coupling, the current effect can also be quantified separately as the energy consumption caused by the current:

[0057] ;

[0058] wherein, is the current-induced added energy consumption of the USV; is the energy conversion efficiency of the USV propulsion system; is the relative speed between the USV and the current; is the angle between the current direction and the sailing direction of the USV; is the sailing speed of the USV; is the length of the USV; is the sailing time of the USV in the corresponding wind field environment.

[0059] The comprehensive energy consumption model can be expressed as:

[0060] ;

[0061] wherein, is the real-time total propulsion power of the unmanned surface vehicle; is the energy conversion efficiency of the propulsion system of the unmanned surface vehicle; is the water resistance; is the wind-induced additional resistance; is the wave-induced additional resistance; is the speed through water.

[0062] Integrating over time gives the total energy consumption:

[0063] ;

[0064] wherein, is the total energy consumption of the unmanned surface vehicle from the start time to the end time; is the start time of the voyage; is the end time of the voyage; is the real-time total propulsion power of the unmanned surface vehicle; is the time step of each time sub-interval.

[0065] To actually combine with the path planning algorithm, the equivalent modular parameterization method is adopted to discretize the energy consumption components:

[0066] ;

[0067] wherein, is the total energy consumption of the unmanned surface vehicle from the start time to the end time; is the water resistance energy consumption; is the wind-induced additional energy consumption; is the wave-induced additional energy consumption; is the control system energy consumption.

[0068] The method provided by the embodiment takes the dynamic environment field model established by decomposing the wind, wave and flow of different frequency components through the space-time dependent function as the unified basis, and the subsequent construction of the water resistance, wind-induced additional and wave-induced additional models directly calls the space-time distribution parameters of the sea current, wind field and wave field output by the model, which not only ensures the consistency and timeliness of the environment parameters relied on by each energy consumption component calculation, avoids the energy consumption estimation deviation caused by the traditional static environment model due to the neglect of the dynamic change of the environment, but also reduces the redundancy of multi-source data through centralized environment parameter modeling, and improves the overall calculation efficiency and accuracy of the model; meanwhile, the relation between the speed of the unmanned surface vehicle relative to the ground and the speed of the unmanned surface vehicle relative to the water is established through the vector relation, the core influence of the sea current effect is fully considered, the sea current not only directly changes the speed relative to the water and then affects the quantization result of the water resistance energy consumption, but also indirectly acts on the calculation of the relative speed of the vehicle wind through the speed relative to the ground, so that the quantization of the wind-induced additional energy consumption can reflect the linkage influence of the sea current, and the calculation of the wave-induced additional energy consumption also indirectly reflects the auxiliary influence of the sea current due to the coupling relation between the sea current and the wave field parameters in the dynamic environment field model, realizing the full penetration of the sea current effect in the calculation of various resistance energy consumptions. Compared with the traditional energy consumption model which ignores the sea current or only considers the direct influence of the sea current, the realness and integrity of the energy consumption quantization under the complex marine environment are greatly improved, and finally through the comprehensive energy consumption model formed by the four types of energy consumption, the energy consumption change law of the unmanned surface vehicle under the space-time coupling effect of wind, wave and flow can be accurately captured, which provides a high-fidelity fitness function support for subsequent energy optimal path planning, and effectively solves the problem that the geometric optimal path does not match the actual energy optimal path.

[0069] S102, multi-dimensional improvement is made on the traditional particle swarm optimization algorithm, and a hybrid enhanced particle swarm optimization algorithm is constructed.

[0070] Among them, the safety space set and the adjacent graph are constructed by the grid method, the A* algorithm is used to generate the initial optimal path and to perform bounded random disturbance and safety correction on the target particle waypoint; the B-spline technology is used to smooth the path generated by iteration, and the smoothed optimal solution is resampled and re-injected into the population; the inertia weight and the learning factor are dynamically adjusted based on the number of iterations, and the double-guiding factor is introduced to make a secondary adjustment to the cognitive term of the particle speed update.

[0071] In specific implementation, the safe space set and the adjacency graph are constructed by the grid method, the A* algorithm is used to generate the initial optimal path and the target particle waypoints are subjected to bounded random disturbance and safety correction, including: dividing the navigation search space of the unmanned surface vehicle into regular grids, performing obstacle collision detection on each grid point, screening out grid points that do not intersect with any obstacles, and forming a collision-free safe space set; based on the safe space set, taking the grid points as nodes and the Euclidean distance between the nodes as edge weights, an adjacency graph for path search is constructed; taking the starting point and the target point of the navigation task as input, running the A* algorithm on the adjacency graph, searching for a discrete initial path with the minimum cost between the starting point and the target point, generating a preset number of intermediate waypoints through spline interpolation to form the initial path of the first particle; for the target particle to be initialized, taking the generated initial path waypoints as a reference, applying a bounded random disturbance within a preset amplitude range to generate an initial waypoint set of the target particle; the generated target particle waypoints after disturbance are checked one by one, if a waypoint falls into an obstacle region, the nearest neighbor safe point of the waypoint is searched in the safe space set, and the nearest neighbor safe point is used to replace the original waypoint; if no effective nearest neighbor safe point is searched, the original waypoint is retained, and finally the initial path of the particle population is obtained.

[0072] Specifically, first, according to the navigation task range of the unmanned surface vehicle, the obstacle distribution density and the path planning accuracy requirement, the edge length size of the regular grid is set, the entire navigation search space is uniformly divided into a plurality of mutually adjacent regular grids, and the unique coordinate identifier (such as row and column number or rectangular coordinate system coordinates) of each grid is determined; for each grid point, a preset obstacle collision detection algorithm is called, the spatial geometric parameters (such as the vertex coordinates of the polygonal obstacle, the center coordinates and radius of the circular obstacle) of the known obstacle are input, and it is judged whether the grid point coordinates fall within the spatial coverage range of any obstacle (if it is a polygonal obstacle, the ray method is used to judge whether the point is within the polygon, and if it is a circular obstacle, it is calculated whether the distance from the point to the center is less than the radius), the grid points not intersecting with any obstacle are screened out, the coordinates and identifier information of these grid points are stored in a set, and a collision-free safe space set is formed; based on the safe space set, each grid point is taken as a node for path search, the adjacency relationship between nodes (such as only allowing adjacent grid nodes in up, down, left, right and diagonal directions to establish connection) is defined, the straight line distance between each pair of adjacent nodes is calculated by the Euclidean distance formula, the distance is taken as the weight of the corresponding edge, an adjacency graph containing the node set, edge set and edge weight information is constructed, and is stored as a data structure (such as an adjacency matrix or an adjacency list) convenient for A* algorithm calling; the start point coordinates and target point coordinates of the navigation task are converted into the identifier information of the corresponding grid points as the input parameters of the A* algorithm, the heuristic function (such as constructed based on the Euclidean distance or Manhattan distance from the node to the target point) is set, and the algorithm runs first by adding the start point to the open list, each time the node with the minimum cost function (f(n)=g(n)+h(n), wherein g(n) is the actual cost from the start point to the node n, and h(n) is the estimated cost from the node n to the target point) is selected from the open list, if the node is the target point, the search is terminated, otherwise the adjacent nodes of the node are regularly added to the open list and the cost information is updated, and the processed nodes are moved to the closed list, until the discrete initial path with the minimum cost between the start point and the target point is found; for the discrete initial path, according to the particle waypoint quantity requirement (such as presetting 20 waypoints for each particle), a cubic spline interpolation method is used to generate intermediate waypoints between adjacent nodes of the discrete path that meet the smoothness requirement, the start point, the intermediate waypoints and the target point are combined in order to form the complete initial path of the first particle; for the remaining target particles to be initialized, first read all the waypoint coordinates of the generated initial path, set the disturbance amplitude range of each waypoint (such as the maximum disturbance distance in the x-axis and y-axis directions is not more than 3 grid edge lengths), generate a random disturbance value in the range of [-disturbance amplitude, disturbance amplitude] for the x-coordinate and y-coordinate of each waypoint through a pseudo-random number generator, add the original waypoint coordinates and the corresponding disturbance value to generate the initial waypoint set of the target particle;For each target particle waypoint after disturbance, the above obstacle collision detection process is repeated one by one, if it is detected that a waypoint falls into the obstacle region, the nearest neighbor safety point search is started, the Euclidean distance between the waypoint and all grid points in the safety space set is calculated based on the k nearest neighbor algorithm (k = 1), the grid point with the smallest distance is selected as the nearest neighbor safety point, the coordinates of the safety point are used to replace the original disturbed waypoint, and the replacement information is recorded; If it is found that the distance between the waypoint and all grid points in the safety space set is greater than the preset threshold (such as 5 grid lengths) after searching, that is, no effective nearest neighbor safety point is found, the original disturbed waypoint is retained to avoid deviation from the reasonable search range due to excessive correction; After all the waypoints of the target particle are checked and replaced, the initial path of the particle is formed in the order of the waypoints, and the above target particle initialization process is repeated until the initial paths of the particle population of the preset size (such as 50 particles) are generated, and all paths meet the waypoint safety constraint and cover a reasonable search space.

[0073] For example, in an embodiment, first, a "safety point set" is constructed by scanning the map by grid: The search space is divided into regular grids, and collision detection is performed for each grid point. All grid points that do not intersect with obstacles are recorded as a free space set Based on this, an adjacency graph is constructed:

[0074] ;

[0075] wherein, is the adjacency graph used for path search; is the free space set; is the edge set of the graph; , are any two nodes in ; and is the adjacency threshold.

[0076] The Euclidean distance is used as the edge weight. For a given starting point and target point , the A* algorithm is run on the graph to determine the discrete path with the minimum cost:

[0077] ;

[0078] wherein, is the initial path obtained by A* algorithm search; is the first node of the path; is the starting point; is the node of the path; is the target point.

[0079] The path is resampled using spline interpolation. These intermediate waypoints form the initial path of the first particle:

[0080] ;

[0081] in, To represent the first particle The coordinate vectors of each waypoint; , For the first The rectangular coordinates of each waypoint in a two-dimensional plane; Waypoint number; The total number of waypoints contained in a single particle.

[0082] For the remaining particles, regarding waypoints Apply a bounded random perturbation:

[0083] ;

[0084] in, For the first Coordinate vectors after waypoint disturbance To represent the first particle The coordinate vectors of each waypoint; For the first Random disturbance coefficients for each waypoint; For the first Disturbance direction vectors for each waypoint; The disturbance coefficients are assumed to follow a uniform distribution over the interval.

[0085] Optionally, B-spline technology is used to smooth the iteratively generated paths. The smoothed optimal solution is then resampled and reinjected into the population. This includes: selecting the individual optimal path and the population-wide optimal path from the iteratively generated next-generation candidate paths; extracting the discrete waypoint coordinates corresponding to the two types of optimal paths as the original input nodes for B-spline smoothing; setting the order of the B-spline; using the extracted discrete waypoint coordinates as control vertices; calculating and determining the curve parameters through B-spline basis functions; and constructing a B-spline smoothing curve that fits the trend of the original optimal path and has continuous curvature; and based on the navigation constraints of the unmanned surface vessel, the curve is then smoothed. The feasibility of the constructed B-spline smoothed curve is verified. If there are regions where the curvature of the curve exceeds the limit, the coordinates of the corresponding control vertices are adjusted, and the B-spline curve is re-optimized until all navigation constraints are met. For the verified B-spline smoothed curve, the smoothed curve is resampled according to the preset sampling interval using the equal arc length sampling method to generate a new discrete waypoint set with the same number as the original population particle waypoints, forming the smoothed optimal path. The smoothed optimal path obtained by resampling replaces the particle paths in the population with fitness values ​​lower than the preset value, or is directly injected into the population as new particles to update the population.

[0086] In a specific implementation, from the new generation of candidate paths generated by the particle swarm optimization iteration, the individual optimal path and the population global optimal path with the optimal fitness value are selected by calculating the fitness value of each path (by integrating the energy consumption model), all discrete waypoint coordinates corresponding to the two types of optimal paths are extracted and arranged in order according to the path, and the original input nodes for B-spline smoothing processing are obtained. According to the navigation smoothing requirement of the unmanned surface vehicle, the order of the B-spline is set (for example, the 3-order B-spline is used to ensure that the curve is twice continuously derivable), the discrete waypoint coordinates extracted are directly used as the control vertices of the B-spline, the B-spline basis functions of each interval are calculated based on the de Boor algorithm or the Cox-de Boor recursive formula, the curve parameters are determined by combining the linear combination of the basis functions and the control vertices, and the B-spline smooth curve that fits the trend of the original optimal path and has continuous curvature is constructed. Based on the navigation constraints of the unmanned surface vehicle (including the maximum turning angle, the maximum curvature, the minimum turning radius, etc.), the curvature values of each point of the B-spline smooth curve are calculated by numerical differentiation, and the feasibility of the curve is checked section by section. If it is detected that there is a region with a local curvature exceeding the standard, the control vertices corresponding to the region are located, the x and y coordinates of these control vertices are adjusted (for example, translated in the direction of decreasing curvature), the B-spline curve is recalculated, and the checking and adjusting process is repeated until the curvatures of all sections of the curve meet the navigation constraints. For the B-spline smooth curve that passes the checking, the sampling point coordinates on the curve are calculated according to the preset sampling interval (consistent with the waypoint spacing of the original population particles) using the equal arc length sampling algorithm. The cumulative curve arc length is recorded when the preset interval is reached, and a new discrete waypoint set with the same number of waypoints as the original population particles is generated. The smooth optimal path is arranged in the sampling order to form the smoothed optimal path. The fitness values of all particles in the population are calculated and compared with the preset threshold. The smooth optimal path obtained by resampling is used to replace the particle path with a fitness value lower than the threshold in the population, or directly added to the population as a new particle. The population size is updated and the total number of particles remains unchanged, and the smooth path is re-injected into the population.

[0087] Optionally, the inertia weight and the learning factor are dynamically adjusted based on the number of iterations, including: determining the total number of iterations of the algorithm, and the initial value of the inertia weight, the initial value of the cognitive learning factor, and the initial value of the social learning factor; based on the ratio of the current iteration number to the total iteration number, the inertia weight is gradually reduced starting from the initial value by multiplying the initial value by 1 minus 0.5 times the ratio; the cognitive learning factor is gradually reduced starting from the initial value by multiplying the initial value by 1 minus the ratio; and the social learning factor is gradually increased starting from the initial value by multiplying the initial value by the ratio.

[0088] Specifically, first, according to the path optimization requirement of the particle swarm optimization algorithm, the total iteration number (such as 100 times, 200 times) of the algorithm is preset, and the initial value of the inertia weight (such as 0.9), the initial value of the cognitive learning factor (such as 2.0), and the initial value of the social learning factor (such as 0.5) are set; in the process of each iteration of the algorithm, the ratio of the current iteration number to the total iteration number (denoted as the iteration progress ratio) is calculated in real time; for the inertia weight, taking the set initial value as the benchmark, according to the calculation rule of "current inertia weight = initial value × [1-(iteration progress ratio × 0.5)]", the inertia weight is gradually reduced with the increase of the iteration number, so that the algorithm maintains strong global search ability in the early stage and enhances local search precision in the later stage; for the cognitive learning factor, taking the set initial value as the benchmark, according to the calculation rule of "current cognitive learning factor = initial value × (1-iteration progress ratio)", the cognitive learning factor is gradually reduced with the increase of the iteration number, reducing the dependence of the particle on its historical optimal position and avoiding falling into local optimum in the later stage; for the social learning factor, taking the set initial value as the benchmark, according to the calculation rule of "current social learning factor = initial value × iteration progress ratio", the social learning factor is gradually increased with the increase of the iteration number, enhancing the learning ability of the particle to the global optimal position of the population and guiding the population to quickly converge to the optimal solution in the later stage; the values of the above three parameters are updated in real time in each iteration, and the updated parameters are substituted into the iteration formula of the particle speed and position, until the algorithm completes the total iteration number.

[0089] For example, in an embodiment, the adjustment of the inertia weight and the learning factor can be represented as:

[0090] ;

[0091] ;

[0092] ;

[0093] wherein, is the adjusted inertia weight; is the initial value of the inertia weight; is the current iteration number; is the total iteration number; is the adjusted cognitive learning factor; is the initial value of the cognitive learning factor; is the adjusted social learning factor; is the initial value of the social learning factor.

[0094] Optionally, the introduction of the double guide factor makes a secondary adjustment to the cognitive term of the particle velocity update, including: calculating the distance between the current position of each particle and the individual optimal position of the particle; taking the ratio of the distance and the preset reference value as input to calculate the output result of the Sigmoid function; combining the ratio of the distance and the preset reference value, the preset adjustment coefficient, and the product of the first guide factor and the Sigmoid function output result, the second guide factor and the correlation value of the original calculation result of the cognitive term to obtain the scaling factor; in the cognitive term calculation link of the particle velocity update, the scaling factor is integrated into the calculation process of the cognitive term to make a secondary correction to the cognitive term; in the iteration process, the value of the first guide factor is gradually reduced and the value of the second guide factor is gradually increased with the increase of the iteration number.

[0095] Specifically, in each iteration of the particle swarm optimization algorithm, the average distance between the waypoint coordinate set corresponding to the current position of each particle and the waypoint coordinate set corresponding to the historical individual optimal position of the particle is calculated by the Euclidean distance formula to obtain the distance between the current position and the individual optimal position of each particle; the preset reference value of the distance is preset (determined based on the navigation search space range and the path optimization accuracy), the ratio of the above distance of each particle and the preset reference value is calculated, and the ratio is taken as an input parameter to substitute into the Sigmoid function to calculate the Sigmoid function output result corresponding to each particle; the adjustment coefficient (such as a fixed value between 0.8-1.2) is preset, and the initial values of the first guide factor and the second guide factor are set (such as the initial value of the first guide factor is 1.5 and the initial value of the second guide factor is 0.5), and the scaling factor of each particle is calculated by the formula; in the cognitive term calculation link of the particle velocity update (the original calculation logic of the cognitive term is “cognitive learning factor x random number x (individual optimal position-current position)”), the calculated scaling factor is multiplied by the original calculation result of the cognitive term to make a secondary correction to the cognitive term, and the adjusted cognitive term is obtained; in the algorithm iteration process, the guide factor values are adjusted according to a preset rule with the increase of the iteration number, such as the first guide factor is reduced according to the rule of multiplying the initial value by (1-iteration number / total iteration number) every iteration, and the second guide factor is increased according to the rule of multiplying the initial value by (1+iteration number / total iteration number) (or set a fixed step to gradually adjust), until the iteration is completed, and the scaling factor is calculated in real time and the cognitive term is corrected according to the above process in each iteration.

[0096] For example, in an embodiment, the distance between the current position of each particle and the individual optimal position of the particle is calculated as follows:

[0097] ;

[0098] wherein, a distance between the current position of each particle and the individual optimal position of the particle; a particle individual optimal position; a current position of each particle.

[0099] calculating a Sigmoid function:

[0100] ;

[0101] wherein, an output result of the Sigmoid function; a distance between the current position of each particle and the individual optimal position of the particle; a preset reference value.

[0102] calculating a scaling factor:

[0103] ;

[0104] wherein, a scaling factor; a distance between the current position of each particle and the individual optimal position of the particle; a preset reference value; a basic adjustment term; a first guide factor; a random disturbance term; an output result of the Sigmoid function; a second guide factor; a cognitive learning factor; a particle individual optimal position; a current position of each particle.

[0105] Optionally, the method further comprises: pre-generating a random number matrix required by algorithm iteration by processing the fitness evaluation task of each particle in parallel through multi-threading; performing preliminary screening on the candidate path through a geometric bounding box and region attribution judgment, and performing collision detection and path correction only on the path suspected to have collision; generating a random step length conforming to the Levy beta stable distribution through the Mantegna method, and adding the random step length to the velocity component of the particle pointing to the global optimal position of the population.

[0106] Specifically, firstly, a multi-threaded processing environment is configured. The number of parallel threads is set according to the population size and the number of hardware cores. Each particle in the particle population is assigned to an independent thread. A random number matrix (containing random parameters for velocity updates, perturbation calculations, etc.) required for the entire algorithm iteration process is pre-generated and stored in shared memory. Each thread reads the random numbers as needed. For all candidate paths generated in the iteration, the minimum geometric bounding box of each path is constructed (the bounding box boundary is determined by calculating the extreme values ​​of the x and y coordinates of all waypoints on the path). At the same time, based on the regional classification of obstacles (the navigation space is divided into safe areas and suspected collision areas according to the distribution of obstacles), the candidate paths are initially screened by judging whether the bounding box of the path overlaps with the bounding box of the obstacle and whether the path falls into the suspected collision area. Paths that do not overlap with any obstacle bounding box and are located in the safe area are retained (directly judged as non-collision paths, no further detection is required). Only the bounding boxes are selected. Identifying potential collision paths that overlap or fall into suspected collision areas; for each potential collision path, verifying the collision relationship between each waypoint and obstacle on the path (using ray casting or distance judgment method); if a collision is detected, correcting it by replacing the nearest neighbor safe point or replanning the path segment; employing the Mantenia method, generating two independent sets of random numbers u and v that follow a standard normal distribution, calculating a random step size that conforms to the Lévy β stable distribution based on the formula, and decomposing the random step size into corresponding components along the x and y dimensions; in the calculation stage of the social term (the velocity component pointing to the global optimal position of the population) for particle velocity updates, adding the decomposed random step size component to the social term velocity component to complete the adjustment of the social term velocity; each parallel thread synchronously completes the fitness evaluation of the assigned particles, and after the evaluation is completed, summarizes the fitness results of all particles; throughout the process, the initial screening, collision detection correction, and velocity component adjustment processes are executed cyclically until the algorithm iteration ends.

[0107] For example, in one embodiment, the Lévy β-stable random step size is generated for each dimension of each generation using the Mantenia method:

[0108] ;

[0109] in, The generated Lévy β-stable random step size; Let be a random variable that follows a normal distribution; Let be a random variable that follows a standard normal distribution; for Follows a mean of 0 and a variance of The normal distribution; for It follows a standard normal distribution; Let be the shape parameter of the Lévy distribution.

[0110] Add a random step to the velocity component pointing to the global optimum position by dimension:

[0111] ;

[0112] wherein, is the particle velocity of the i-th iteration; is the particle velocity of the i-th iteration; is the particle velocity of the i-th iteration; is the particle velocity of the i-th iteration; is the weight coefficient of the Levy step; is the generated Levy beta-stable random step; is the global optimum position of the population; is the current position of the particle.

[0113] S103, embed the comprehensive energy consumption model as a fitness function into the hybrid enhanced particle swarm optimization algorithm, perform real-time energy consumption evaluation on each candidate path, update the individual and global optimal path, and optimize the path smoothing during the algorithm iteration process, and output the optimal navigation path of the unmanned surface vehicle.

[0114] In specific implementation, the core parameters of the hybrid enhanced particle swarm optimization algorithm are set, and based on the heuristic initialization of the particle population, a plurality of initial candidate paths are obtained; the comprehensive energy consumption model is taken as a fitness function, the spatiotemporal information of the navigation trajectory and the dynamic environmental field parameters corresponding to each initial candidate path are input, and the total energy consumption of each initial candidate path is calculated in real time through the comprehensive energy consumption model to obtain the fitness value of each path; the fitness value of the current candidate path of each particle is compared with the historical optimal fitness value of the particle, and if the current value is better, the current path is updated as the individual optimal path of the particle, the individual optimal paths of all particles are summarized, the fitness values are compared, and the path with the optimal fitness value is selected as the global optimal path of the population; based on the dynamic parameter adjustment method and the double-guiding factor adjustment mechanism, the speed and position of each particle are updated based on the current individual optimal path and the global optimal path to generate a new generation of candidate paths; the B-spline technique is used for curve fitting on the new generation of candidate paths to control the continuous curvature of the path to meet the navigation constraints of the unmanned surface vehicle; at the same time, the path obstacle avoidance feasibility is checked through the fast collision detection strategy, and the path that does not meet the constraints is modified; it is judged whether the current iteration number reaches the preset maximum iteration number, and after the iteration is terminated, the global optimal path of the population finally obtained is taken as the optimal navigation path of the unmanned surface vehicle.

[0115] Specifically, first, the core parameters of the hybrid enhanced particle swarm optimization algorithm are set, including particle population size, maximum iteration number, initial value and adjustment range of inertia weight, initial value of cognitive learning factor and social learning factor, B-spline order, collision detection threshold, etc. At the same time, the safety space set and the adjacent graph are constructed based on the grid method, the initial optimal path is generated by A* algorithm, and the target particle waypoint is randomly disturbed and safety corrected, the heuristic initialization of particle population is completed, and multiple initial candidate paths are obtained. The comprehensive energy consumption model considering the spatio-temporal effects of coupled wind, wave and current is taken as the fitness function. For each initial candidate path, the corresponding spatio-temporal information of the sailing trajectory (including the coordinates of each waypoint, sailing time, ground speed, etc.) and dynamic environmental field parameters (spatio-temporal distribution parameters of sea current, wind field and wave field) are extracted and input into the comprehensive energy consumption model. Through the sub-item calculation and summary of water resistance energy consumption, wind-induced additional energy consumption, wave-induced additional energy consumption and control system energy consumption in the model, the total energy consumption of each initial candidate path is obtained in real time, which is taken as the fitness value of each path (the lower the energy consumption, the better the fitness). The fitness value of the current candidate path of each particle is compared with the historical optimal fitness value of the particle (the initial iteration historical optimal fitness value is the current path fitness value). If the current fitness value is better, the current candidate path is updated as the individual optimal path of the particle. After all particles complete the individual optimal path judgment, the individual optimal paths of all particles are summarized, the fitness values of each path are compared one by one, and the path with the optimal fitness value (the lowest total energy consumption) is selected as the global optimal path of the population. Based on the preset dynamic parameter adjustment method (dynamically adjusting inertia weight and learning factor with iteration number), double-guiding factor adjustment mechanism (calculating proportional scaling factor to modify cognitive term twice), combining the individual optimal path of the current particle and the global optimal path of the population, the particle velocity update formula (velocity update formula with Levy step length) and position update formula are substituted, the new velocity and position of each particle are calculated, and a new generation of candidate paths are generated. For the new generation of candidate paths, the individual optimal path and the global optimal path of the population are selected, the discrete waypoint coordinates are taken as the control vertices, the B-spline order is set and the B-spline smooth curve is constructed by calculating the basis function, the curvature of each point on the curve is calculated by numerical differentiation, and whether the path constraints such as maximum turning angle and minimum turning radius of the unmanned surface vehicle are met is checked. If there is a region with excessive curvature, the corresponding control vertex coordinates are adjusted to optimize the curve again to ensure the continuity of the path curvature. At the same time, a fast collision detection strategy is adopted. First, the candidate paths are preliminarily screened through geometric bounding box and region attribution judgment, and only the suspected collision paths are subjected to waypoint-by-waypoint collision detection. The paths detected to collide are modified by replacing with the nearest neighbor safety point or re-planning the path segment to ensure the feasibility of path obstacle avoidance.After each iteration, it is judged whether the current iteration number reaches the preset maximum iteration number. If not, the process of "calculating fitness value, updating individual and global optimal path, updating particle velocity and position, path smoothing and collision detection correction" is returned to continue iteration. If the maximum iteration number is reached, the iteration is terminated, and the finally obtained global optimal path of the population is taken as the optimal navigation path of the unmanned surface vehicle.

[0116] The method provided by the embodiment, in the first aspect, in the process of constructing the comprehensive energy consumption model, a dynamic environment field model is established by decomposing different frequency components of wind, wave and current through a space-time dependent function, real-time space-time distribution parameters of the current, wind field and wave field are output, and the speed of the unmanned surface vehicle relative to the ground and the speed of the unmanned surface vehicle relative to the water are associated based on a vector relationship, so that the current effect directly acts on the speed of the unmanned surface vehicle relative to the water to accurately quantify the basic energy consumption of water resistance, the relative speed of the boat wind is calculated based on the wind field parameters to quantify the wind-induced additional energy consumption, the JONSWAP wave spectrum model and wave spectrum integration are used to quantify the wave-induced additional energy consumption, and the energy consumption of the integrated control system is integrated, so that the comprehensive energy consumption model can fully capture the dynamic changes of each energy consumption component in a complex marine environment, avoid the energy consumption estimation deviation caused by the neglect of the environmental coupling effect of the traditional static model, provide a high-fidelity fitness function for path optimization, and ensure that the path selected subsequently meets the optimal demand of actual energy consumption.

[0117] In the second aspect, in the improvement of the particle swarm optimization algorithm in multiple dimensions, the safe space set and the adjacent graph are constructed by the grid method, the A* algorithm is used to generate an initial path, and heuristic initialization is realized by combining bounded random disturbance and safety correction, so that the safety of the initial population is ensured and the diversity is improved, laying a foundation for fast convergence of the algorithm. The application of B-spline technology makes the path curvature continuous and meet the navigation constraints of the unmanned surface vehicle. The inertia weight and the learning factor are dynamically adjusted based on the iteration number (the inertia weight gradually decreases, the cognitive learning factor gradually decreases, and the social learning factor gradually increases), so as to balance the global search and local search ability of the algorithm. The double-guiding factor is used for secondary adjustment of the cognitive term (a scaling factor is calculated by combining distance and a Sigmoid function output), which further optimizes the particle search strategy, effectively avoids the algorithm from falling into local optimum, and improves the path optimization precision and efficiency.

[0118] Thirdly, in generating random step sizes to optimize velocity components, a random step size conforming to the Lévy β stable distribution is generated using the Mantenia method. This random step size is then added dimensionally to the velocity component of the particle pointing to the global optimum position of the population. The "heavy-tailed" characteristic of the Lévy distribution is used to enhance the particle's exploration ability in the solution space, enabling the particle to jump out of the local optimum region to search for a better path. At the same time, multi-threaded parallel processing of fitness evaluation tasks and a fast collision detection strategy (first screening suspected collision paths through geometric bounding boxes and then accurately detecting them) are used to improve the algorithm's computational efficiency while ensuring the feasibility of path obstacle avoidance. Ultimately, the optimal navigation path output by the algorithm has the characteristics of lowest energy consumption, safe navigation, and smooth feasibility, solving the problem of mismatch between geometric optimum and actual energy consumption optimum in traditional path planning.

[0119] Corresponding to the aforementioned embodiment of the energy-optimal path planning method for unmanned surface vessels in complex marine environments, this application also provides an embodiment of an energy-optimal path planning device for unmanned surface vessels in complex marine environments.

[0120] Figure 2 This is a schematic diagram of the energy-optimal path planning device for unmanned surface vessels in complex marine environments provided in Embodiment 2 of this application. Please refer to... Figure 2 The apparatus provided in this embodiment includes a construction module 210, an improvement module 220, and a determination module 230;

[0121] The construction module 210 is used to construct an energy consumption model that couples the spatiotemporal effects of wind, waves, and currents. The energy consumption model comprehensively quantifies the basic energy consumption of water resistance, wind-induced additional resistance energy consumption, wave-induced additional resistance energy consumption, and control system energy consumption. The ocean current effect is integrated into the quantification process of various resistance energy consumptions through a coupling mechanism.

[0122] The improved module 220 is used to improve the traditional particle swarm optimization algorithm in multiple dimensions and construct a hybrid enhanced particle swarm optimization algorithm. Specifically, it constructs a safe space set and adjacency graph using a grid method, generates an initial optimal path using the A* algorithm, and applies bounded random perturbation and safety correction to the target particle waypoints. It uses B-spline technology to smooth the iteratively generated path, resamples the smoothed optimal solution, and re-injects it into the population. It dynamically adjusts the inertia weight and learning factor based on the number of iterations and introduces a dual-guiding factor to perform a secondary adjustment to the cognitive term for particle velocity updates.

[0123] The determining module 230 is used to embed the energy consumption model as a fitness function into the hybrid enhanced particle swarm optimization algorithm, and output the optimal navigation path of the unmanned surface vessel through real-time energy consumption evaluation of each candidate path, individual and global optimal path updates, and path smoothing optimization during the algorithm iteration process.

[0124] The apparatus of this embodiment can be used to perform... Figure 1The steps of the method embodiment, the specific implementation principle and the implementation process are similar, and details are not repeated here.

[0125] The implementation process of the functions and roles of each unit in the above device is specifically described in the implementation process of the corresponding steps in the above method, and details are not repeated here.

[0126] For the device embodiment, since it basically corresponds to the method embodiment, the related part can be referred to the part of the method embodiment. The device embodiment described above is only schematic, and the units described as separate components can or can not be physically separate, and the components displayed as units can or can not be physical units, that is, they can be located in one place, or distributed on multiple network units. According to actual needs, part or all of the modules can be selected to achieve the purpose of the scheme of the present application. Those skilled in the art can understand and implement without creative labor.

[0127] The above is only the preferred embodiment of the present application, and is not used to limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for energy-optimal path planning of unmanned surface vessels in complex marine environments, characterized in that, The method includes: A comprehensive energy consumption model is constructed that couples the spatiotemporal effects of wind, waves, and currents. The comprehensive energy consumption model comprehensively quantifies the basic energy consumption of water resistance, wind-induced additional resistance energy consumption, wave-induced additional resistance energy consumption, and control system energy consumption. Among them, the ocean current effect is integrated into the quantification process of various resistance energy consumptions through a coupling mechanism. This paper proposes a multi-dimensional improvement to the traditional particle swarm optimization algorithm, constructing a hybrid enhanced particle swarm optimization algorithm. Specifically, a safe space set and adjacency graph are built using a grid method; the A* algorithm is used to generate the initial optimal path and apply bounded random perturbation and safety correction to the target particle waypoints; B-spline technology is used to smooth the iteratively generated paths, and the smoothed optimal solution is resampled and re-injected into the population; the inertia weight and learning factor are dynamically adjusted based on the number of iterations, and a dual guiding factor is introduced to perform a secondary adjustment on the cognitive term for particle velocity updates. The comprehensive energy consumption model is embedded as a fitness function into the hybrid enhanced particle swarm optimization algorithm. During the algorithm iteration process, real-time energy consumption evaluation, individual and global optimal path updates, and path smoothing optimization are performed on each candidate path to output the optimal navigation path of the unmanned surface vessel. Construct a comprehensive energy consumption model that couples the spatiotemporal effects of wind, waves, and currents, including: A dynamic environmental field model is established by decomposing different frequency components of wind, waves, and currents using a spatiotemporal dependency function; the dynamic environmental field model outputs the spatiotemporal distribution parameters of ocean currents, wind fields, and wave fields. Based on the hydrodynamic parameters of the dynamic environmental field model, the relationship between the unmanned surface vessel's velocity relative to the ground and its velocity relative to the water is established through vector relationships. Based on the velocity relative to the water, the energy consumption corresponding to the basic water resistance during navigation is quantified, and a water resistance energy consumption model is constructed. Based on the wind field parameters of the dynamic environmental field model, wind speed and wind direction parameters at different spatiotemporal locations are extracted. Combined with the ground velocity obtained through the vector relationship, the relative velocity between the unmanned surface vessel and the wind is calculated. Based on the relative velocity, the energy consumption corresponding to wind-induced additional drag under the wind field environment is quantified, and a wind-induced additional energy consumption model is constructed. Based on the wave field parameters of the dynamic environmental field model, the wave height, wave frequency, and wave direction parameters at different spatiotemporal locations in the dynamic environmental field are extracted. The energy consumption corresponding to the additional resistance of waves under the wave field environment is quantified by wave spectrum theory, and a wave-induced additional energy consumption model is constructed. Integrate the energy consumption of unmanned surface vessels in the processes of course control, track maintenance, and actuator response to construct a control system energy consumption model; The energy consumption models of water resistance, wind-induced additional energy consumption, wave-induced additional energy consumption, and control system energy consumption are summarized to form a comprehensive energy consumption model.

2. The method according to claim 1, characterized in that, A safe space set and adjacency graph are constructed using a grid method. The A* algorithm is used to generate the initial optimal path, and bounded random perturbations and safety corrections are applied to the target particle waypoints, including: The navigation search space of the unmanned surface vessel is divided into regular grids. Obstacle collision detection is performed on each grid point, and grid points that do not intersect with any obstacles are selected to form a collision-free safe space set. Based on the aforementioned safe space set, an adjacency graph for path search is constructed, with grid points as nodes and the Euclidean distance between nodes as edge weights. Using the starting point and target point of the navigation mission as input, the A* algorithm is run on the adjacency graph to search for the discrete initial path with the minimum cost between the starting point and the target point. A preset number of intermediate waypoints are generated through spline interpolation to form the initial path of the first particle. For the target particle being initialized, a bounded random perturbation within a preset amplitude range is applied based on the generated initial path points to generate the initial path point set of the target particle. The waypoints of the target particles generated after the disturbance are checked one by one. If a waypoint falls into the obstacle area, the nearest safe neighbor point of the waypoint is searched in the safe space and the original waypoint is replaced by the nearest safe neighbor point. If no effective nearest safe neighbor point is found, the original waypoint is retained, and the initial path of the particle population is finally obtained.

3. The method according to claim 1, characterized in that, A second adjustment is made to the cognitive terms for particle velocity updates by introducing dual guiding factors, including: Calculate the distance between the current position and the optimal position of each particle; Using the ratio of the distance to a preset reference value as input, the output of the Sigmoid function is calculated. The scaling factor is obtained by combining the ratio of the distance to the preset reference value, the preset adjustment coefficient, the product of the first guiding factor and the output of the Sigmoid function, and the correlation value of the second guiding factor and the original calculation result of the cognitive item. In the cognitive term calculation stage of particle velocity update, the scaling factor is incorporated into the cognitive term calculation process to perform secondary correction on the cognitive term; during the iteration process, as the number of iterations increases, the value of the first guiding factor is gradually reduced and the value of the second guiding factor is increased.

4. The method according to claim 1, characterized in that, The inertia weight and learning factor are dynamically adjusted based on the number of iterations, including: Determine the total number of iterations for the algorithm, as well as the initial values ​​for the inertia weight, cognitive learning factor, and social learning factor; Based on the ratio of the current iteration number to the total iteration number, the inertia weight is gradually reduced starting from the initial value, following the rule of multiplying the initial value by 1 and subtracting 0.5 times the ratio as the number of iterations increases; For the cognitive learning factor, starting from the initial value, it gradually decreases as the number of iterations increases by multiplying the initial value by 1 and subtracting the ratio. For the social learning factor, starting from the initial value, it gradually increases with the number of iterations by multiplying the initial value by the ratio mentioned above.

5. The method according to claim 1, characterized in that, Constructing a model for the additional energy consumption caused by the wave, including: Wave field parameters corresponding to each spatiotemporal node on the unmanned surface vessel's trajectory are extracted from the dynamic environmental field model; the wave field parameters include significant wave height, peak frequency, and dominant wave direction. Based on the sea state type of the actual navigation area, a suitable JONSWAP wave spectrum model is selected, and the effective wave height and peak wave frequency are used as input parameters to construct a wave spectrum that matches the wave field characteristics of the current spatiotemporal location. Based on the constructed JONSWAP wave spectrum, combined with the hull type parameters and navigation status parameters of the unmanned surface vessel, the additional resistance force generated by waves of different frequencies on the hull is calculated by wave spectrum integration. Based on the combined additional resistance force, and combined with the unmanned surface vessel's speed and travel time at the corresponding time and space node, calculate the energy consumption corresponding to the additional resistance of the waves at the corresponding time and space node. The wave-induced additional energy consumption quantification results of all spatiotemporal nodes along the navigation trajectory are summarized to form a wave-induced additional energy consumption model that changes dynamically with time and space.

6. The method according to claim 1, characterized in that, Output the optimal navigation path for the unmanned surface vessel, including: The core parameters of the hybrid enhanced particle swarm optimization algorithm are set, and multiple initial candidate paths are obtained based on the heuristic initialization of the particle swarm. The comprehensive energy consumption model is used as the fitness function. The spatiotemporal information of the flight trajectory and the dynamic environmental field parameters corresponding to each initial candidate path are input. The total energy consumption of each initial candidate path is calculated in real time through the comprehensive energy consumption model to obtain the fitness value of each path. Compare the fitness value of each particle's current candidate path with the particle's historical best fitness value. If the current value is better, update the current path as the particle's individual best path. Summarize the individual best paths of all particles, compare their fitness values, and select the path with the best fitness value as the global best path for the population. Based on the dynamic parameter adjustment method and the dual guiding factor adjustment mechanism, the velocity and position of each particle are updated by combining the current individual optimal path and the global optimal path, and a new generation of candidate paths are generated. For the next generation of candidate paths, B-spline technology is used for curve fitting to ensure that the curvature of the path continuously meets the navigation constraints of the unmanned surface vessel. At the same time, a fast collision detection strategy is used to verify the feasibility of obstacle avoidance of the path and to correct paths that do not meet the constraints. Determine whether the current iteration count has reached the preset maximum iteration count. After the iteration terminates, use the final global optimal path obtained by the population as the optimal navigation path for the unmanned surface vessel.

7. The method according to claim 1, characterized in that, The B-spline technique is used to smooth the iteratively generated paths, and the smoothed optimal solution is resampled and re-injected into the population, including: From the iteratively generated new generation of candidate paths, the individual optimal path and the population global optimal path are selected, and the discrete waypoint coordinates corresponding to the two types of optimal paths are extracted as the original input nodes for B-spline smoothing. The order of the B-spline is set, and the extracted discrete waypoint coordinates are used as control vertices. The curve parameters are determined by calculating the B-spline basis functions to construct a smooth B-spline curve that fits the original optimal path trend and has continuous curvature. Based on the navigation constraints of the unmanned surface vessel, the feasibility of the constructed B-spline smooth curve is verified; if there are regions where the curvature of the curve exceeds the limit, the coordinates of the corresponding control vertex are adjusted and the B-spline curve is re-optimized until all navigation constraints are met. For the B-spline smoothed curve that has passed the verification, the smoothed curve is resampled according to the preset sampling interval using the equal arc length sampling method, generating a new discrete waypoint set with the same number as the original population particle waypoints, forming the smoothed optimal path; The smoothed optimal path obtained from resampling replaces the particle path in the population with a fitness value lower than the preset value, or is directly injected into the population as a new particle to update the population.

8. The method according to claim 1, characterized in that, A hybrid enhanced particle swarm optimization algorithm is constructed by making multi-dimensional improvements to the traditional particle swarm optimization algorithm, including: The fitness evaluation task of each particle is processed in parallel by multi-threading, and the random number matrix required for algorithm iteration is generated in advance. First, candidate paths are initially screened by geometric bounding boxes and region affiliation judgment. Collision detection and path correction are only performed on paths that are suspected of having collisions. A random step size conforming to the Lévy β stable distribution is generated using the Mantenia method, and this random step size is added dimensionally to the velocity component of the particle pointing to the global optimal position of the population.

9. A device for energy-optimal path planning of unmanned surface vessels in complex marine environments, characterized in that, The device includes a construction module, an improvement module, and a determination module; The construction module is used to construct an energy consumption model that couples the spatiotemporal effects of wind, waves, and currents. The energy consumption model comprehensively quantifies the basic energy consumption of water resistance, wind-induced additional resistance energy consumption, wave-induced additional resistance energy consumption, and control system energy consumption. The ocean current effect is integrated into the quantification process of various resistance energy consumptions through a coupling mechanism. The improved module is used to improve the traditional particle swarm optimization algorithm in multiple dimensions and construct a hybrid enhanced particle swarm optimization algorithm. Specifically, a safe space set and adjacency graph are constructed using a grid method; the A* algorithm is used to generate the initial optimal path and perform bounded random perturbation and safety correction on the target particle waypoints; B-spline technology is used to smooth the iteratively generated path, and the smoothed optimal solution is resampled and re-injected into the population; the inertia weight and learning factor are dynamically adjusted based on the number of iterations, and a dual guiding factor is introduced to perform a secondary adjustment on the cognitive term for particle velocity updates. The determining module is used to embed the energy consumption model as a fitness function into the hybrid enhanced particle swarm optimization algorithm, and output the optimal navigation path of the unmanned surface vessel through real-time energy consumption evaluation of each candidate path, individual and global optimal path updates, and path smoothing optimization during the algorithm iteration process. Construct a comprehensive energy consumption model that couples the spatiotemporal effects of wind, waves, and currents, including: A dynamic environmental field model is established by decomposing different frequency components of wind, waves, and currents using a spatiotemporal dependency function; the dynamic environmental field model outputs the spatiotemporal distribution parameters of ocean currents, wind fields, and wave fields. Based on the hydrodynamic parameters of the dynamic environmental field model, the relationship between the unmanned surface vessel's velocity relative to the ground and its velocity relative to the water is established through vector relationships. Based on the velocity relative to the water, the energy consumption corresponding to the basic water resistance during navigation is quantified, and a water resistance energy consumption model is constructed. Based on the wind field parameters of the dynamic environmental field model, wind speed and wind direction parameters at different spatiotemporal locations are extracted. Combined with the ground velocity obtained through the vector relationship, the relative velocity between the unmanned surface vessel and the wind is calculated. Based on the relative velocity, the energy consumption corresponding to wind-induced additional drag under the wind field environment is quantified, and a wind-induced additional energy consumption model is constructed. Based on the wave field parameters of the dynamic environmental field model, the wave height, wave frequency, and wave direction parameters at different spatiotemporal locations in the dynamic environmental field are extracted. The energy consumption corresponding to the additional resistance of waves under the wave field environment is quantified by wave spectrum theory, and a wave-induced additional energy consumption model is constructed. Integrate the energy consumption of unmanned surface vessels in the processes of course control, track maintenance, and actuator response to construct a control system energy consumption model; The energy consumption models of water resistance, wind-induced additional energy consumption, wave-induced additional energy consumption, and control system energy consumption are summarized to form a comprehensive energy consumption model.

Images