A Dynamic Adaptive Amphibious Vehicle Path and Energy Consumption Optimization Method

Abstract

This application discloses a dynamic adaptive amphibious vehicle path and energy consumption optimization method, relating to the field of path and energy consumption optimization. The method includes: acquiring information data; the information data includes: environmental data and amphibious vehicle-related data; the environmental data includes: wind speed, wind direction, ocean current speed, ocean current direction, wave height, and wave frequency; the amphibious vehicle-related data includes: position, speed, and acceleration; the information data contains timestamps; inputting the information data into a path and energy consumption optimization model to obtain optimized planned path information; the optimized planned path information is the lowest energy-consuming planned path determined based on a heuristic function; the path and energy consumption optimization model employs reinforcement learning algorithms and A / B algorithms. * The algorithm is derived from an amphibious vehicle surface energy consumption model and a heuristic function, after updates and optimizations. This application enables path and energy consumption optimization, improving the amphibious vehicle's endurance and mission completion efficiency.

Description

Technical Field

[0001] This application relates to the field of path and energy consumption optimization, and in particular to a dynamic adaptive amphibious vehicle path and energy consumption optimization method. Background Technology

[0002] In modern marine applications, amphibious vehicles are widely used in environmental monitoring, resource exploration, military missions, and rescue operations. With increasing application demands, the ability of amphibious vehicles to perform tasks in complex and dynamic marine environments has become crucial. However, the complex marine environment presents numerous challenges in their path planning process.

[0003] Traditional A * Path planning algorithms such as Dijkstra's are mainly suitable for static environments and cannot cope with complex environments that change dynamically in real time. Although methods such as Model Predictive Control (MPC) can handle dynamic environments, they rely on accurate environmental and vehicle dynamics models, and their effectiveness is often limited in practical applications due to the inaccuracy of the models.

[0004] Furthermore, many existing path planning algorithms only focus on path length or time, failing to adequately consider energy consumption optimization. For energy-constrained systems like amphibious vehicles, energy consumption is a critical metric. If only path length or time is considered while energy consumption is ignored, the vehicle may run out of energy during mission execution, failing to complete its intended task. Summary of the Invention

[0005] The purpose of this application is to provide a dynamic adaptive amphibious vehicle path and energy consumption optimization method, which can optimize path and energy consumption, and improve the amphibious vehicle's range and mission completion efficiency.

[0006] To achieve the above objectives, this application provides the following solution:

[0007] This application provides a dynamic adaptive amphibious vehicle path and energy consumption optimization method, including:

[0008] Acquire information data; the information data includes: environmental data and amphibious vehicle-related data; the environmental data includes: wind speed, wind direction, ocean current speed, ocean current direction, wave height, and wave frequency; the amphibious vehicle-related data includes: position, speed, and acceleration; the information data contains timestamps;

[0009] The information data is input into the path and energy consumption optimization model to obtain optimized planned path information; the optimized planned path information is the lowest energy-consuming planned path determined based on a heuristic function; the path and energy consumption optimization model employs reinforcement learning algorithms and A... *The algorithm is obtained by updating and optimizing the amphibious vehicle surface energy consumption model and heuristic function; the amphibious vehicle surface energy consumption model is a mathematical model determined based on the dynamic characteristics of the amphibious vehicle and environmental factors.

[0010] Optionally, the amphibious vehicle surface energy consumption model includes a propulsion energy consumption model and an environmental resistance energy consumption model; the expression for the amphibious vehicle surface energy consumption model is:

[0011] E = E p +E r ;

[0012] Where E represents the amphibious vehicle's surface energy consumption; E p To promote energy consumption; E r Energy consumption due to environmental resistance.

[0013] Optionally, the expression for the propulsion energy consumption model is:

[0014]

[0015] Among them, F p η is the propulsion force; d is the travel distance; and η is the efficiency of the propulsion system.

[0016] Optionally, the expression for the environmental resistance energy consumption model is:

[0017] E r =E r_wind +E r_current +E r_wave ;

[0018] E r_wind =w c ·k c ·(v w ·cos(θ-d w ));

[0019] E r_current =w f ·k f ·(v c cos(θ-d c ));

[0020] E r_wave =w w ·k w ·(h w ·f w );

[0021] Among them, E r_wind Energy consumption due to wind resistance; E r_current Energy consumption due to ocean current resistance; E r_wave Energy consumption due to wave resistance; w ck is the weighting coefficient for wind resistance energy consumption. c The drag coefficient is v. w θ is the wind speed; θ is the heading angle; d w For wind direction; cos(θ-d) w ) is the angle between the amphibious vehicle's heading angle and the wind direction; w f k is the weighting coefficient for ocean current resistance energy consumption. f v is the ocean current drag coefficient; c d represents the ocean current velocity. c The direction of the ocean current; cos(θ-d) c ) is the angle between the amphibious vehicle's heading angle and the direction of the ocean current; h w f represents the wave height. w The wave frequency; k w The wave resistance coefficient is w. w This is the weighting coefficient for wave resistance energy consumption.

[0022] Optionally, the expression of the heuristic function is:

[0023]

[0024] Where h(n) is a heuristic function; w d The weights are the path lengths; (x) n ,y n (x) represents the coordinates of the current node; g ,y g E represents the coordinates of the target location node. r Energy consumption due to environmental resistance; E p To promote energy consumption.

[0025] Optionally, the method for determining the path and energy consumption optimization model specifically includes:

[0026] Obtain training information data;

[0027] Construct a deep Q-network; the deep Q-network includes an experience replay buffer;

[0028] Initialize the experience replay buffer and weight parameters of the deep Q-network;

[0029] The state space and action space are determined based on the training data; the expression for the state space s is: s = [v w ,d w ,v c ,d c ,h w ,f w The expression for the action space a is: a = [Δw], x, y, v, θ]. d ,Δw c,Δw f ,Δw w ,Δθ,Δv];v w For wind speed; d w For wind direction; v c d represents the ocean current velocity. c In the direction of ocean currents; h w f represents the wave height. w θ is the wave frequency; θ is the heading angle; x is the x-coordinate of the position; y is the y-coordinate of the position; v is the velocity; Δw c Adjustment of the wind resistance energy consumption weighting coefficient; Δw d Adjust the path length weight; Δw f Adjustment of the ocean current resistance energy consumption weighting coefficient; Δw w Δθ is the adjustment of the wave resistance energy consumption weighting coefficient; Δθ is the heading angle adjustment; Δv is the speed adjustment.

[0030] For any time step, the corresponding action space is selected based on the current state space, and the next state space is determined according to the corresponding reward function; the time step is determined after time point alignment processing based on the timestamp; the reward function is determined based on the state space and the action space.

[0031] Randomly sample data from the stored experience replay buffer, and base the sample data on A. * The algorithm performs path search and, based on the heuristic function, trains and updates the weight parameters of the deep Q network with network convergence as the goal, to obtain the trained deep Q network; the stored experience replay buffer stores the state space, action space, reward function, and next state space corresponding to each time step;

[0032] The trained deep Q-network was determined as the path and energy consumption optimization model.

[0033] Optionally, the expression for the reward function is:

[0034]

[0035] Where R(s,a) is the reward function; E is the amphibious vehicle's surface energy consumption; α, β, and γ are all weighting parameters; a i For the action at the i-th time step; δ j This is the result of the j-th training update adjustment; n is the total number of time steps; m is the total number of training updates; a i+1 This refers to the action at the (i+1)th time step.

[0036] Optionally, the information data to be acquired specifically includes:

[0037] Obtain initial information data;

[0038] The initial information data is aligned according to timestamps;

[0039] A Kalman filter is used to remove noise from the aligned initial information data to obtain the processed data.

[0040] The processed data is fused using a weighted average method to obtain the information data.

[0041] According to the specific embodiments provided in this application, this application has the following technical effects:

[0042] This application provides a dynamic adaptive amphibious vehicle path and energy consumption optimization method, which combines reinforcement learning algorithms and A... * The algorithm optimizes the heuristic function parameters, improving the efficiency and accuracy of path search. The heuristic function comprehensively considers factors such as path length, propulsion energy consumption, and environmental drag energy consumption to ensure the lowest energy-consuming path planning. Furthermore, the reinforcement learning algorithm not only optimizes the heuristic function parameters but also dynamically adjusts the path planning based on real-time environmental data and the amphibious vehicle's status during path execution, ensuring the path planning remains optimal. This application significantly improves the amphibious vehicle's autonomous operation capability and mission completion efficiency, enabling it to complete more tasks under limited energy conditions through real-time data processing and dynamic path adjustment. Therefore, this application achieves path and energy consumption optimization, improving the amphibious vehicle's endurance and mission completion efficiency. Attached Figure Description

[0043] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0044] Figure 1 This is a flowchart of the dynamic adaptive amphibious vehicle path and energy consumption optimization method of this application;

[0045] Figure 2 To improve A * Overall flowchart of the algorithm's dynamic adaptive amphibious vehicle path and energy consumption optimization method;

[0046] Figure 3 A schematic diagram illustrating sensor data collection and environmental modeling;

[0047] Figure 4 This is a flowchart of reinforcement learning, experience playback, and model update based on deep Q-networks. Detailed Implementation

[0048] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0049] The applicant is aware of Chinese patent document CN116560363A, which describes a low-energy path planning method for robots using an improved ant colony algorithm in a multi-attribute grid environment. This method establishes a multi-attribute grid map of robot motion, considers various energy consumption criteria such as path length, number of turns, slope, and ground friction coefficient, and improves the heuristic function of the traditional ant colony algorithm while optimizing the pheromone model to enhance the energy efficiency of path planning. However, this method is mainly applicable to pre-known grid environments and may be insufficient for practical applications in complex marine environments.

[0050] Chinese patent document CN116243701A discloses an optimal energy consumption path planning method for underwater robots based on optimal propulsion speed. This method generates candidate paths in the form of B-spline curves using a quantum particle swarm optimization (PSO) algorithm, calculates the optimal propulsion speed and energy consumption of each path segment using vector analysis, and sums the energy consumption of each path segment to obtain the total energy consumption of the candidate path. This method improves the computational efficiency of optimal energy consumption path planning. However, this method is mainly applicable to static or quasi-static environments and has weak real-time response capability to dynamic environmental changes. Furthermore, although the particle swarm optimization algorithm has advantages in searching for the global optimum, its computational complexity is high in high-dimensional complex environments, and it has certain shortcomings in real-time performance.

[0051] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0052] In an exemplary embodiment, a dynamic adaptive amphibious vehicle path and energy consumption optimization method is provided. This method is executed by a computer device, specifically, it can be executed by a computer device such as a terminal or a server alone, or it can be executed by both a terminal and a server. In this embodiment, the method is described using an application to a server as an example. Figure 1 As shown, it includes the following steps 100 to 200.

[0053] Specifically, a dynamic adaptive amphibious vehicle path and energy consumption optimization method includes:

[0054] Step 100: Acquire information data. The information data includes: environmental data and data related to the amphibious vehicle itself; environmental data includes: wind speed, wind direction, ocean current speed, ocean current direction, wave height, and wave frequency; data related to the amphibious vehicle itself includes: location, speed, and acceleration; the information data contains timestamps.

[0055] Step 200: Input the information data into the path and energy consumption optimization model to obtain the optimized planned path information. The optimized planned path information is the lowest energy-consuming planned path determined based on a heuristic function; the path and energy consumption optimization model uses reinforcement learning algorithms and A... * The algorithm is obtained by updating and optimizing the amphibious vehicle surface energy consumption model and heuristic function; the amphibious vehicle surface energy consumption model is a mathematical model determined based on the dynamic characteristics of the amphibious vehicle and environmental factors.

[0056] The amphibious vehicle surface energy consumption model includes a propulsion energy consumption model and an environmental resistance energy consumption model; the expression for the amphibious vehicle surface energy consumption model is:

[0057] E = E p +E r ;

[0058] Where E represents the amphibious vehicle's surface energy consumption; E p To promote energy consumption; E r Energy consumption due to environmental resistance.

[0059] The expression for the propulsion energy consumption model is:

[0060]

[0061] Among them, F p η is the propulsion force; d is the travel distance; and η is the efficiency of the propulsion system.

[0062] The expression for the environmental resistance energy consumption model is:

[0063] E r =E r_wind +E r_current +E r_wave ;

[0064] E r_wind =w c ·k c ·(v w ·cos(θ-d w ));

[0065] E r_current =w f ·k f ·(v c ·cos(θ-d c ));

[0066] E r_wave =w w ·k w ·(h w ·f w );

[0067] Among them, E r_wind Energy consumption due to wind resistance; E r_current Energy consumption due to ocean current resistance; E r_wave Energy consumption due to wave resistance; w c k is the weighting coefficient for wind resistance energy consumption. c The drag coefficient is v. w θ is the wind speed; θ is the heading angle; d w For wind direction; cos(θ-d) w ) is the angle between the amphibious vehicle's heading angle and the wind direction; w f k is the weighting coefficient for ocean current resistance energy consumption. f v is the ocean current drag coefficient; c d represents the ocean current velocity. c The direction of the ocean current; cos(θ-d) c ) is the angle between the amphibious vehicle's heading angle and the direction of the ocean current; h w f represents the wave height. w The wave frequency; k w The wave resistance coefficient is w. w This is the weighting coefficient for wave resistance energy consumption.

[0068] In one embodiment, the expression for the heuristic function is:

[0069]

[0070] Where h(n) is a heuristic function; w d The weights are the path lengths; (x) n ,y n (x) represents the coordinates of the current node; g ,y g E represents the coordinates of the target location node. r Energy consumption due to environmental resistance; E p To promote energy consumption.

[0071] As an optional implementation method, the method for determining the path and energy consumption optimization model specifically includes:

[0072] Acquire training information data; construct a deep Q-network; the deep Q-network includes an experience replay buffer.

[0073] Initialize the experience replay buffer and weight parameters of the deep Q-network.

[0074] The state space and action space are determined based on the training data; the expression for the state space s is: s = [v w ,d w ,v c ,d c ,h w ,f w The expression for the action space a is: a = [Δw], x, y, v, θ]. d ,Δw c ,Δw f ,Δw w ,Δθ,Δv];v w For wind speed; d w For wind direction; v c d represents the ocean current velocity. c In the direction of ocean currents; h w f represents the wave height. w θ is the wave frequency; θ is the heading angle; x is the x-coordinate of the position; y is the y-coordinate of the position; v is the velocity; Δw c Adjustment of the wind resistance energy consumption weighting coefficient; Δw d Adjust the path length weight; Δw f Adjustment of the ocean current resistance energy consumption weighting coefficient; Δw w Δθ is the adjustment of the wave resistance energy consumption weighting coefficient; Δθ is the heading angle adjustment; Δv is the speed adjustment.

[0075] For any time step, the corresponding action space is selected based on the current state space, and the next state space is determined according to the corresponding reward function; the time step is determined after time point alignment processing based on the timestamp; the reward function is determined based on the state space and the action space.

[0076] Randomly sample data from the stored experience replay buffer, and then base the sample data on A. * The algorithm performs path search and, based on a heuristic function, trains and updates the weight parameters of the deep Q-network with the goal of network convergence, resulting in the trained deep Q-network. The stored experience replay buffer contains the state space, action space, reward function, and next state space corresponding to each time step.

[0077] The trained deep Q-network was determined as the path and energy consumption optimization model.

[0078] Specifically, the expression for the reward function is:

[0079]

[0080] Where R(s,a) is the reward function; E is the amphibious vehicle's surface energy consumption; α, β, and γ are all weighting parameters; a iFor the action at the i-th time step; δ j This is the result of the j-th training update adjustment; n is the total number of time steps; m is the total number of training updates; a i+1 This refers to the action at the (i+1)th time step.

[0081] In one embodiment, acquiring information data specifically includes: acquiring initial information data; aligning the initial information data according to timestamps; using a Kalman filter to remove noise from the aligned initial information data to obtain processed data; and using a weighted average method to fuse the processed data to obtain information data.

[0082] This application addresses the path planning and energy consumption optimization problems of amphibious vehicles in dynamic marine environments. It modifies the heuristic function and applies an improved A... * The algorithm solves the path decision problem. Simultaneously, by collecting real-time environmental data such as wind speed, wind direction, ocean current speed, ocean current direction, wave height, and frequency, it dynamically adjusts the heuristic function parameters for path planning, optimizing path planning and energy management. By combining reinforcement learning and real-time data processing techniques, it dynamically adapts to environmental changes, comprehensively considering path planning and energy optimization to improve the amphibious vehicle's endurance and mission completion efficiency.

[0083] In practical applications, the overall flowchart corresponding to the method mentioned in this application is shown below. Figure 2 Specifically, it includes the following steps:

[0084] Step 1: Real-time Environmental Data Collection and Processing. Sensors are used to collect real-time environmental data and information related to the amphibious vehicle itself, including wind speed v. w Wind direction d w Ocean current speed v c Ocean current direction d c Wave height h w Ocean wave frequency f w The sensors collect data on the amphibious vehicle's current position, speed, acceleration, and other information. Data is collected every second to ensure real-time accuracy. Each data collection is timestamped for time synchronization during subsequent processing.

[0085] After collecting the required data, all data needs to be timestamped to ensure that each type of data is processed and fused at the same point in time. A Kalman filter is used to process the raw data, removing noise and improving accuracy. After data preprocessing, data fusion is performed to improve overall accuracy. Weighted averaging can be used to fuse similar data from multiple sensors to further enhance accuracy.

[0086] Step 2: Establish an amphibious vehicle surface energy consumption model. To optimize path planning and energy management, it is necessary to rationally plan the path with minimum energy consumption based on the amphibious vehicle's dynamic characteristics and environmental factors. The energy consumption of the amphibious vehicle during mission execution includes propulsion energy consumption and environmental drag energy consumption.

[0087] Propulsion energy consumption mainly depends on the amphibious vehicle's propulsion force, travel distance, and propulsion system efficiency. Specifically, it can be expressed as:

[0088]

[0089] Among them, F p η is the propulsion force, d is the travel distance, and η is the efficiency of the propulsion system.

[0090] In dynamic marine environments, amphibious vehicles need to overcome environmental resistance to maintain their mobility. Environmental resistance mainly includes wind resistance, ocean current resistance, and wave resistance, which are the main sources of external resistance for amphibious vehicles when operating in dynamic marine environments.

[0091] The wind resistance energy consumption model represents the energy consumed by the amphibious vehicle due to air resistance during its movement, specifically expressed as:

[0092] E r_wind =w c ·k c ·(v w ·cos(θ-d w )).

[0093] In the formula, v w d represents wind speed. w Indicates wind direction, cos(θ-d) w θ is the angle between the amphibious vehicle's heading angle and the wind direction, used to calculate the actual wind speed felt in the amphibious vehicle's forward direction. c w is the drag coefficient. c This is the weighting coefficient for wind resistance energy consumption, used to adjust the proportion of wind resistance energy consumption in total energy consumption.

[0094] Ocean current resistance energy consumption represents the energy consumed by an amphibious vehicle due to ocean current resistance during its movement. Specifically, it can be expressed as:

[0095] E r_current =w f ·k f ·(v c ·cos(θ-d c )).

[0096] In the formula, v c d represents the ocean current velocity. c Indicates the direction of ocean current, cos(θ-d) cθ is the angle between the amphibious vehicle's heading angle and the ocean current direction, used to calculate the actual ocean current velocity felt by the amphibious vehicle in its forward direction. f w is the ocean current drag coefficient. f This is a weighting coefficient for ocean current resistance energy consumption, used to adjust the proportion of ocean current resistance energy consumption in the total energy consumption.

[0097] Wave drag energy consumption represents the energy consumed by an amphibious vehicle due to wave drag during its movement. Specifically, it can be expressed as:

[0098] E r_wave =w w ·k w ·(h w ·f w ).

[0099] In the formula, h w f represents wave height. w h represents the product of wave frequency, wave height, and frequency. w ·f w This indicates the combined drag effect of waves on the amphibious vehicle. w w is the wave resistance coefficient. w The weighting coefficient for wave resistance energy consumption is used to adjust the proportion of wave resistance energy consumption in the total energy consumption.

[0100] The combined environmental resistance energy consumption is expressed as the sum of wind resistance, ocean current resistance, and wave resistance. The specific formula is:

[0101] E r =E r_wind +E r_current +E r_wave .

[0102] In summary, by combining propulsion energy consumption and environmental resistance energy consumption, a total energy consumption model is obtained:

[0103] E = E p +E r .

[0104] like Figure 3 As shown, in practical applications, the process of sensor data collection and environmental modeling is as follows:

[0105] First, environmental data was collected, based on wind speed and direction sensors, ocean current speed and direction sensors, wave and frequency sensors, and GPS and IMU data were also acquired.

[0106] The acquired data undergoes Coleman filtering and data fusion processing before the environmental model is updated. The environmental model includes: ocean current model (i.e., ocean current drag energy consumption), wind field model (i.e., wind drag energy consumption model), and wave model (i.e., wave drag energy consumption).

[0107] Step 3: Reinforcement Learning Parameter Training. A Deep Q-Network (DQN) is used to learn and adjust the parameters of the path planning heuristic function, dynamically adjusting the path based on real-time environmental data and an energy consumption model to optimize path planning. The input to the reinforcement learning model includes the current environmental state and the amphibious vehicle's state; the output is the updated heuristic function parameters and the dynamic adjustment strategy. The specific process is as follows:

[0108] The state space includes wind speed v w Wind direction d w Ocean current speed v c Ocean current direction d c Wave height h w Ocean wave frequency f w The amphibious vehicle's current position (x, y), velocity v, and heading angle θ. Therefore, the state can be represented as:

[0109] s = [v w ,d w ,v c ,d c ,h w ,f w ,x,y,v,θ].

[0110] Accordingly, the action space can be represented as:

[0111] a=[Δw d ,Δw c ,Δw f ,Δw w ,Δθ,Δv].

[0112] This vector represents all the actions that a reinforcement learning model can take, including adjusting the parameters of the heuristic function and the specific actions of dynamically adjusting the path.

[0113] The reward function is defined as the negative of the total energy consumption of the path, taking into account both path smoothness and adjustment effects:

[0114]

[0115] The specific steps for training and updating are as follows:

[0116] Initialize the weights and experience replay buffer of the deep Q network.

[0117] At each time step, collect the current environmental state s and the amphibious vehicle state, and select action a (adjust the heuristic function parameters or dynamically adjust the path).

[0118] Perform action a to obtain a new state s' and a reward R(s,a).

[0119] Store the current state s, action a, reward R(s,a), and next state s' into the experience replay buffer.

[0120] Randomly sample mini-batches of samples from the experience replay buffer and use these samples to update the weights of the Q-network. Repeat training until the Q-network converges to obtain the optimal heuristic function parameter tuning and path dynamic adjustment strategy.

[0121] Step 4: Path Planning and Adjustment. Based on the current environment model and reinforcement learning strategy, the path planning is dynamically calculated and adjusted. The optimized heuristic function is used to calculate the path cost, and the A* algorithm is combined for path search. Dynamic adjustments are made during execution. Specifically:

[0122] The heuristic function, which comprehensively considers factors such as path length, environmental resistance energy consumption, and propulsion energy consumption, is expressed as:

[0123]

[0124] Among them, w d The weights for path length, (x) n ,y n (x) represents the coordinates of the current node. g ,y g E represents the coordinates of the target location node. r For environmental resistance energy consumption, E p To promote energy consumption.

[0125] The A* algorithm is used to search for and generate paths, which are dynamically adjusted based on real-time environmental data and the amphibious vehicle's status during execution. The reinforcement learning model continuously updates the heuristic function parameters and calculates the optimal path in real time. If significant environmental changes occur (such as wind speed or ocean current direction), the system recalculates the path to ensure optimal path planning is always maintained. For details on reinforcement learning in deep Q-networks, as well as the process of experience replay and model updates, please refer to [link to documentation / documentation]. Figure 4 .

[0126] Step 5: Experience Replay and Model Update. During path planning, historical data is stored and utilized through an experience replay mechanism to train the neural network. The specific process is as follows:

[0127] Experience storage: During each path planning process, the current state, action, reward, and next state are stored in the experience replay buffer.

[0128] Mini-batch update: Randomly draw a mini-batch of samples from the experience replay buffer and use these samples to update the weights of the Q network.

[0129] Model update: Through repeated training and updates, the reinforcement learning model can better adjust the heuristic function parameters and dynamically adjust the path, thus optimizing the path planning strategy.

[0130] Advantages of this invention:

[0131] First, this application establishes a dynamic environment model by collecting and processing environmental data in real time (such as wind speed, wind direction, ocean current speed, ocean current direction, wave height, and frequency), ensuring that path planning can adapt to environmental changes in real time, thus improving accuracy and robustness. Second, this application combines reinforcement learning algorithms and the A algorithm, optimizing the heuristic function parameters of the A algorithm to improve path search efficiency and accuracy. The heuristic function comprehensively considers factors such as path length, propulsion energy consumption, and environmental drag energy consumption, ensuring the path planning with the lowest energy consumption. Furthermore, the reinforcement learning model not only optimizes the heuristic function parameters but also dynamically adjusts them based on real-time environmental data and the amphibious vehicle's status during path execution, ensuring that path planning is always optimal. This application significantly improves the amphibious vehicle's autonomous operation capability and mission completion efficiency, enabling the amphibious vehicle to complete more tasks under limited energy conditions through real-time data processing and dynamic path adjustment.

[0132] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0133] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A dynamic adaptive amphibious vehicle path and energy consumption optimization method, characterized in that, The dynamic adaptive amphibious vehicle path and energy consumption optimization method includes: Acquire information data; the information data includes: environmental data and amphibious vehicle-related data; the environmental data includes: wind speed, wind direction, ocean current speed, ocean current direction, wave height, and wave frequency; the amphibious vehicle-related data includes: position, speed, and acceleration; the information data contains timestamps; The information data is input into the path and energy consumption optimization model to obtain optimized planned path information; the optimized planned path information is the lowest energy-consuming planned path determined based on a heuristic function; the path and energy consumption optimization model employs reinforcement learning algorithms and A... * The algorithm is obtained by updating and optimizing the amphibious vehicle surface energy consumption model and heuristic function; the amphibious vehicle surface energy consumption model is a mathematical model determined based on the dynamic characteristics of the amphibious vehicle and environmental factors. The amphibious vehicle surface energy consumption model includes a propulsion energy consumption model and an environmental resistance energy consumption model; the expression for the amphibious vehicle surface energy consumption model is: E=E p +E r ; Where E represents the amphibious vehicle's surface energy consumption; E p To promote energy consumption; E r Energy consumption due to environmental resistance; The expression for the propulsion energy consumption model is: Among them, E p η is the propulsion force; d is the travel distance; η is the efficiency of the propulsion system. The expression for the environmental resistance energy consumption model is as follows: AND r =And r_wind +E r_current +E r_wave ; E r_wind =w c ·k c ·(v w ·cos(θ-d w )); E r_current =w f ·k f ·(v c ·cos(θ-d c )); E r_wave =w w ·k w ·(h w ·f w ); Among them, E r_wind Energy consumption due to wind resistance; E r_current Energy consumption due to ocean current resistance; E r_wave Energy consumption due to wave resistance; w c k is the weighting coefficient for wind resistance energy consumption. c The drag coefficient is v. w θ is the wind speed; θ is the heading angle; d w For wind direction; cos(θ-d) w ) is the angle between the amphibious vehicle's heading angle and the wind direction; w f k is the weighting coefficient for ocean current resistance energy consumption. f v is the ocean current drag coefficient; c d represents the ocean current velocity. c The direction of the ocean current; cos(θ-d) c ) is the angle between the amphibious vehicle's heading angle and the direction of the ocean current; h w f represents the wave height. w The wave frequency; k w The wave resistance coefficient is w. w This is the weighting coefficient for wave resistance energy consumption.

2. The dynamic adaptive amphibious vehicle path and energy consumption optimization method according to claim 1, characterized in that, The expression for the heuristic function is: Where h(n) is a heuristic function; w d The weights are the path lengths; (x) n ,y n (x) represents the coordinates of the current node; g ,y g E represents the coordinates of the target location node. r Energy consumption due to environmental resistance; E p To promote energy consumption.

3. The dynamic adaptive amphibious vehicle path and energy consumption optimization method according to claim 1, characterized in that, The method for determining the path and energy consumption optimization model specifically includes: Obtain training information data; Construct a deep Q-network; the deep Q-network includes an experience replay buffer; Initialize the experience replay buffer and weight parameters of the deep Q-network; The state space and action space are determined based on the training data; the expression for the state space s is: s = [v w ,d w ,v c ,d c ,h w ,f w The expression for the action space a is: a = [Δw], x, y, v, θ]. d ,Δw c ,Δw f ,Δw w ,Δθ,Δ v ];v w For wind speed; d w For wind direction; v c d represents the ocean current velocity. c In the direction of ocean currents; h w f represents the wave height. w θ is the wave frequency; θ is the heading angle; x is the x-coordinate of the position; y is the y-coordinate of the position; v is the velocity; Δw c Adjustment of the wind resistance energy consumption weighting coefficient; Δw d Adjust the path length weight; Δw f Adjustment of the ocean current resistance energy consumption weighting coefficient; Δw w For adjusting the wave resistance energy consumption weighting coefficient; Δθ is for adjusting the heading angle; Δv is for adjusting the speed; For any time step, the corresponding action space is selected based on the current state space, and the next state space is determined according to the corresponding reward function; the time step is determined after time point alignment processing based on the timestamp; the reward function is determined based on the state space and the action space. Randomly sample data from the stored experience replay buffer, and base the sample data on A. * The algorithm performs path search and, based on the heuristic function, trains and updates the weight parameters of the deep Q network with network convergence as the goal, to obtain the trained deep Q network; the stored experience replay buffer stores the state space, action space, reward function, and next state space corresponding to each time step; The trained deep Q-network was determined as the path and energy consumption optimization model.

4. The dynamic adaptive amphibious vehicle path and energy consumption optimization method according to claim 3, characterized in that, The expression for the reward function is: Where R(s,a) is the reward function; E is the amphibious vehicle's surface energy consumption; α, β, and γ are all weight parameters; a i For the action at the i-th time step; δ j This is the result of the j-th training update adjustment; n is the total number of time steps; m is the total number of training updates; a i+1 This refers to the action at the (i+1)th time step.

5. The dynamic adaptive amphibious vehicle path and energy consumption optimization method according to claim 1, characterized in that, The information data obtained specifically includes: Obtain initial information data; The initial information data is aligned according to timestamps; A Kalman filter is used to remove noise from the aligned initial information data to obtain the processed data. The processed data is fused using a weighted average method to obtain the information data.

Images