Autonomous underwater vehicle path planning method, system and electronic equipment

Abstract

The invention relates to the technical field of vehicle path planning, and discloses an autonomous underwater vehicle path planning method and system and electronic equipment, and the method comprises the steps: obtaining and processing initial feature field data, and obtaining high-resolution feature field data; distributing target observation areas for the plurality of aircrafts by using the high-resolution characteristic field data; extracting local feature field data corresponding to each aircraft, and initializing a local Gaussian process model corresponding to the aircraft; dynamically updating the local Gaussian process model by using real-time feature field data acquired by the aircraft, and predicting local feature field prediction data around each aircraft by using the local Gaussian process model; based on the local feature field prediction data, planning a navigation path of each aircraft by using an improved ant colony algorithm; and the task execution state of each aircraft is monitored, and task redistribution or task ending is carried out. According to the invention, task areas can be efficiently allocated for a plurality of aircrafts, and paths of the aircrafts can be planned, so that self-adaptive efficient observation of ocean dynamic characteristics can be realized.

Description

Technical Field

[0001] The present application relates to the technical field of underwater vehicle path planning, and particularly to a path planning method, system and electronic device for an autonomous underwater vehicle. Background Art

[0002] Autonomous Underwater Vehicles (AUVs) play an important role in fields such as ocean observation, and their adaptive sampling path planning method is one of the key technologies. Existing AUV underwater adaptive sampling path methods are mainly divided into two categories. One is the planning method based on preset rules. The AUV dynamically adjusts the sampling strategy during the task execution according to the real-time observed environmental information or prior knowledge. Such methods include the threshold method, the trigger mechanism method, and the region division method, etc. Although these methods are simple to implement and have a low computational cost, they rely on manually designed rules, are difficult to adapt to complex and dynamically changing environments, and since the sampling strategy is adjusted based on local information, it is impossible to optimize the sampling path from a global perspective, which easily leads to resource waste or important areas not being covered. The other is the planning method based on the dynamically predicted information of the environmental characteristics of the model. This method relies on mathematical models and optimization algorithms to accurately describe the environmental characteristics and predict the change trend. The purpose is to optimize the sampling behavior and plan the path systematically, so that the AUV can obtain as much useful data as possible within limited resources, improving the task execution efficiency and data quality. However, in this traditional model-based method, the optimization algorithm usually has a large amount of computation and is difficult to meet the performance requirements in tasks with high real-time requirements, and sufficient historical data and real-time observation data are required to support the model construction and update. When the data is insufficient, the path optimization effect will be greatly reduced. Therefore, in view of the dilemma of complex and dynamically changing environments, how to efficiently allocate task areas and improve the adaptability and accuracy of path planning to meet the needs of feature field observation, so as to achieve efficient observation of the ocean environment, is an urgent problem to be solved. Summary of the Invention

[0003] The main purpose of the present application is to overcome the shortcomings and deficiencies of the prior art, and provide a path planning method, system and electronic device for an autonomous underwater vehicle, which realizes local feature field prediction based on Gaussian process regression, breaks through the limitation that traditional adaptive path planning cannot cope with complex and dynamically changing environments, and solves the problems of complex global model prediction calculation and large computational amount; and can efficiently allocate task areas and plan the paths of each AUV to meet the needs of feature field observation, so as to achieve efficient adaptive observation of the ocean environment.

[0004] To achieve the above purpose, the present application adopts the following technical solutions:

[0005] In the first aspect, the present application provides a path planning method for an autonomous underwater vehicle, including:

[0006] Obtain the initial feature field data, preprocess the initial feature field data to obtain high-resolution feature field data;

[0007] Use the high-resolution feature field data to allocate target observation areas for multiple vehicles to perform observation tasks;

[0008] Based on the high-resolution feature field data, extract the local feature field data around the current position of each vehicle, and initialize the local Gaussian process model corresponding to the observation task performed by each vehicle based on the local feature field data;

[0009] Use the real-time feature field data collected by each vehicle to dynamically update the local Gaussian process model, and predict the local feature field prediction data around each vehicle based on the dynamically updated local Gaussian process model;

[0010] Based on the local feature field prediction data of each vehicle, use an improved ant colony algorithm to plan the navigation path of each vehicle;

[0011] Based on the navigation path, monitor the task execution status of each vehicle, and perform task reallocation or end the task based on the task execution status of all vehicles.

[0012] As a preferred technical solution, obtaining the initial feature field data and preprocessing the initial feature field data to obtain high-resolution feature field data includes:

[0013] Obtain the initial feature field data of the area to be observed, and use a preset interpolation method to perform interpolation processing on the initial feature field data to obtain high-resolution feature field data, and the resolution of the high-resolution feature field data is higher than that of the initial feature field data.

[0014] As a preferred technical solution, using the high-resolution feature field data to allocate target observation areas for multiple vehicles to perform observation tasks includes:

[0015] Based on the high-resolution feature field data, divide multiple areas to be observed with different feature information through a clustering algorithm, and calculate the feature identification criterion values of the areas to be observed;

[0016] Based on a preset feature identification criterion threshold, determine the interesting feature values from the feature identification criterion values, and mark the area corresponding to the clustering of the interesting feature values as the interesting area;

[0017] Perform secondary classification on the interesting areas to determine multiple sub-interesting areas, and divide the multiple sub-interesting areas into several target observation areas to be observed according to a preset evaluation index and the number of vehicles;

[0018] According to the evaluation results of the target observation area, different numbers of vehicles are dispatched to the divided target observation areas for observation.

[0019] As a preferred technical solution, the local Gaussian process model is dynamically updated using the real-time feature field data collected by each vehicle, and the local feature field prediction data around each vehicle is predicted based on the dynamically updated local Gaussian process model, including:

[0020] Define the input feature matrix of the local Gaussian process model;

[0021] Based on the real-time feature field data collected by each vehicle on the navigation path in the target observation area, obtain the target output vector;

[0022] Use the input feature matrix and the target output vector to update the hyperparameters through an optimization function, and update the local Gaussian process model;

[0023] Input the input feature matrix into the updated local Gaussian process model to generate the local feature field prediction data of each vehicle.

[0024] As a preferred technical solution, based on the local feature field prediction data of each vehicle, use an improved ant colony algorithm to plan the navigation path of each vehicle, including:

[0025] Define the pheromone matrix of the ant colony algorithm and obtain the pheromone concentration of the target observation area;

[0026] Based on the local feature field prediction data of the current path point of the vehicle, calculate the feature identification criterion value of the change of the feature value between the current path point and the adjacent path point with respect to the spatial position;

[0027] Based on a preset state transition function, combine the traveling direction of the vehicle from the current path point to the adjacent path point, the feature identification criterion value, and the pheromone concentration of the target observation area, calculate the transition probability from the current path point to the adjacent path point, and determine the next path point based on the transition probability;

[0028] Obtain the real-time feature field data of the transition from the current path point to the next path point, determine the change value of the feature field data, update the pheromone matrix based on the change value of the feature field data, and a single ant completes a path planning;

[0029] When all ants complete the path planning, determine the best navigation path through a preset path evaluation function as the navigation path of each vehicle.

[0030] As a preferred technical solution, determining the next path point based on the transition probability further includes:

[0031] Check whether the vehicle has left the target observation area. If it has left the target observation area, start the re-discovery behavior, detect adjacent waypoints in a preset search manner, and detect the adjacent waypoints that meet the preset conditions as the next waypoint.

[0032] As a preferred technical solution, based on the navigation path, monitor the task execution status of each vehicle, and perform task re-allocation or end the task based on the task execution status of all vehicles, including:

[0033] Based on the distribution of the navigation paths, calculate the coverage rate of the navigation paths of the vehicles on the target observation area in real time, and determine the task execution status of each vehicle based on the coverage rate;

[0034] When it is monitored that the task execution status of the vehicle meets the preset task completion conditions, construct a global Gaussian process model based on the real-time feature field data collected by all vehicles;

[0035] Generate the global uncertainty information of the feature field using the global Gaussian process model, and allocate a new target observation area for the vehicles that have completed the task based on the global uncertainty information.

[0036] As a preferred technical solution, it further includes:

[0037] When the global uncertainty information drops to the preset uncertainty threshold, end the observation tasks of all vehicles.

[0038] In a second aspect, the present application provides an autonomous underwater vehicle path planning system for executing the autonomous underwater vehicle path planning method as described in the first aspect, including:

[0039] A data acquisition module for acquiring initial feature field data and preprocessing the initial feature field data to obtain high-resolution feature field data;

[0040] A task allocation module for allocating target observation areas for multiple vehicles to perform observation tasks using the high-resolution feature field data;

[0041] A local model initialization module for extracting local feature field data around the current position of each vehicle based on the high-resolution feature field data, and initializing the local Gaussian process model corresponding to the observation task of each vehicle based on the local feature field data;

[0042] A local feature field prediction module for dynamically updating the local Gaussian process model using the real-time feature field data collected by each vehicle, and predicting the local feature field prediction data around each vehicle based on the dynamically updated local Gaussian process model;

[0043] A path planning module, configured to plan the navigation path of each vehicle based on the local feature field prediction data of each vehicle by using an improved ant colony algorithm;

[0044] A task monitoring module, configured to monitor the task execution status of each vehicle based on the navigation path, and perform task reallocation or end the task based on the task execution status of all vehicles.

[0045] In a third aspect, the present application provides an electronic device, where the electronic device includes:

[0046] At least one processor; and a memory communicatively connected to the at least one processor;

[0047] Wherein, the memory stores computer program instructions executable by the at least one processor, and the computer program instructions are executed by the at least one processor so that the at least one processor can execute the autonomous underwater vehicle path planning method described above.

[0048] In summary, compared with the prior art, the effective effects brought by the technical solution provided by the present application at least include:

[0049] The present application proposes an autonomous underwater vehicle path planning method. By obtaining initial feature field data, preprocessing the initial feature field data to obtain high-resolution feature field data; using the high-resolution feature field data to allocate target observation areas for multiple vehicles to perform observation tasks; based on the high-resolution feature field data, extracting local feature field data around the current position of each vehicle, and initializing a local Gaussian process model corresponding to the observation task performed by each vehicle based on the local feature field data; dynamically updating the local Gaussian process model by using the real-time feature field data collected by each vehicle, and predicting local feature field prediction data around each vehicle based on the dynamically updated local Gaussian process model; planning the navigation path of each vehicle based on the local feature field prediction data of each vehicle by using an improved ant colony algorithm; based on the navigation path, monitoring the task execution status of each vehicle, and performing task reallocation or ending the task based on the task execution status of all vehicles. It can obtain local environment information more accurately, perform local prediction through the local Gaussian process model, and break through the limitation that traditional adaptive path planning cannot cope with complex and dynamically changing environments; secondly, by combining the improved ant colony adaptive path planning algorithm with the local feature field prediction information obtained through the local Gaussian process model, the navigation paths of multiple autonomous underwater vehicles are optimized, enabling the autonomous underwater vehicles to better adapt to the complex and dynamically changing marine environment, planning more reasonable navigation paths, improving the adaptability and accuracy of path planning, and thus efficiently observing the marine environment.

[0050] In addition, based on the distribution of the navigation path, the present application calculates in real time the coverage rate of the navigation path of the vehicle on the target observation area, determines the task execution status of each vehicle based on the coverage rate; when it is monitored that the task execution status of the vehicle meets the preset task completion condition, a global Gaussian process model is constructed based on the real-time characteristic field data collected by all vehicles, the global uncertainty information of the characteristic field is generated by using the global Gaussian process model, and a new target observation area is allocated to the vehicle that has completed the task based on the global uncertainty information. When the global uncertainty information is reduced to the preset uncertainty threshold, the overall observation task is completed, and the uncertainty of the global information of the characteristic field can be reduced; based on the global uncertainty information of the characteristic field, multiple autonomous underwater vehicles are dynamically re-allocated to different areas for observation, which can make the observation area range of the multi-autonomous underwater vehicle collaborative observation wider, improve the observation efficiency, obtain more useful sampling data, and effectively solve the problems of resource waste or important areas not being covered in the traditional method. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application, and those of ordinary skill in the art can obtain other drawings without creative efforts based on these drawings.

[0052] Figure 1 It is a schematic flowchart of the path planning method for an autonomous underwater vehicle provided by an embodiment of the present application;

[0053] Figure 2 It is a schematic flowchart of the path planning method for an autonomous underwater vehicle provided by another embodiment of the present application;

[0054] Figure 3 It is a schematic flowchart of the task area allocation of multiple autonomous underwater vehicles provided by an embodiment of the present application;

[0055] Figure 4 It is a schematic flowchart of the Gaussian process model generating local characteristic field prediction information or uncertainty information provided by an embodiment of the present application;

[0056] Figure 5 It is a schematic flowchart of using an improved ant colony algorithm for adaptive path planning provided by an embodiment of the present application;

[0057] Figure 6 It is a schematic flowchart of using an improved ant colony algorithm for adaptive path planning provided by another embodiment of the present application;

[0058] Figure 7Schematic diagram of the structure of an autonomous underwater vehicle path planning system provided by an embodiment of the present application;

[0059] Explanation of reference numerals:

[0060] Data acquisition module 101, task allocation module 102, local model initialization module 103, local feature field prediction module 104, path planning module 105, task monitoring module 106. Detailed implementation manners

[0061] In order to enable those skilled in the art to better understand the solutions of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of the present application.

[0062] Referring to "embodiment" in the present application means that the specific features, structures or characteristics described in connection with the embodiment may be included in at least one embodiment of the present application. The phrase appears in various positions in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art explicitly and implicitly understand that the embodiments described in the present application may be combined with other embodiments.

[0063] Please refer to Figure 1 and Figure 2 In an embodiment of the present application, an autonomous underwater vehicle path planning method is provided, including the following steps:

[0064] S1: Obtain initial feature field data, preprocess the initial feature field data to obtain high-resolution feature field data.

[0065] First, obtain the initial feature field data of the area to be monitored in the ocean through satellite remote sensing technology. The initial feature field data is low-resolution feature field data, that is, feature field data with a low spatial resolution. A low spatial resolution is manifested as a sampling interval in the spatial dimension being greater than a preset threshold (the sampling interval in the spatial dimension is large), resulting in a relatively sparse distribution of data points and a small amount of data. Among them, the feature field data can be temperature data, salinity data, flow field data, etc. Temperature data can identify temperature fronts, salinity data can identify salinity gradients, and flow field data can analyze vortices or coastal currents, etc.

[0066] S2: Use the high-resolution feature field data to allocate target observation areas for multiple vehicles to perform observation tasks.

[0067] Specifically, the area to be observed is divided into multiple target observation areas by using the high-resolution feature field data through a clustering algorithm and assigned to each vehicle.

[0068] S3: Based on the high-resolution feature field data, local feature field data around the current position of each vehicle is extracted, and a local Gaussian process model corresponding to the observation task of each vehicle is initialized based on the local feature field data.

[0069] Specifically, the local Gaussian process model is initialized through local feature field information. The expression of the initially constructed local Gaussian process model is the expression of a normal Gaussian process, which will transmit an initial hyperparameter. This process provides a local feature information framework for the local area and reveals the main trends and change patterns of the local area.

[0070] S4: The local Gaussian process model is dynamically updated using the real-time feature field data collected by each vehicle, and local feature field prediction data around each vehicle is predicted based on the dynamically updated local Gaussian process model.

[0071] S5: Based on the local feature field prediction data of each vehicle, an improved ant colony algorithm is used to plan the navigation path of each vehicle.

[0072] S6: Based on the navigation path, the task execution status of each vehicle is monitored, and task reallocation or task termination is performed based on the task execution status of all vehicles.

[0073] Among them, in step S1, initial feature field data is obtained, and the initial feature field data is preprocessed to obtain high-resolution feature field data, including:

[0074] The initial feature field data of the area to be observed is obtained, and the initial feature field data is interpolated using a preset interpolation method to obtain high-resolution feature field data. The resolution of the high-resolution feature field data is higher than that of the initial feature field data. The high-resolution feature field data with high resolution refers to the sampling interval in the spatial dimension being less than the preset threshold.

[0075] In this embodiment, the Kriging interpolation method is used to interpolate the initial feature field data to obtain high-resolution feature field data.

[0076] Please refer to Figure 3 , step S2 specifically includes:

[0077] S21: Based on the high-resolution feature field data, multiple areas to be observed with different feature information are divided through a clustering algorithm, and the feature identification criterion value of the area to be observed is calculated; the feature identification criterion value refers to the key quantitative index of the area to be observed.

[0078] In this embodiment, the K_MEANS clustering method is used to cluster the high-resolution feature field data to obtain the regions to be observed with different feature information.

[0079] S22: Based on the preset threshold of the feature identification criterion, determine the feature values of interest from the feature identification criterion values, and mark the regions corresponding to the clustering of the feature values of interest as the regions of interest;

[0080] Specifically, compare the feature identification criterion values of multiple regions to be observed with different feature information with the preset feature identification criterion threshold, and determine the feature values corresponding to the feature identification criterion values greater than the preset feature identification criterion threshold as the feature values of interest.

[0081] S23: Perform secondary classification on the regions of interest to determine multiple sub-regions of interest, and divide the multiple sub-regions of interest into several target observation regions to be observed according to the preset evaluation index and the number of vehicles.

[0082] In actual implementation, perform secondary classification on the regions of interest according to the density-based spatial clustering of applications with noise (DBSCAN) method. Use the area and feature identification criterion value of the region after secondary classification as the evaluation function, and divide the classified regions into high-quality regions to be observed and low-quality regions to be observed from high to low according to the evaluation index.

[0083] S24: According to the evaluation results of the target observation regions, dispatch different numbers of vehicles to the divided target observation regions for observation.

[0084] Specifically, according to the quality of the target observation regions, dispatch different numbers of vehicles to the high-quality regions to be observed and the low-quality regions to be observed for observation. More autonomous underwater vehicles are dispatched to the high-quality regions.

[0085] Further, please refer to Figure 4 , the specific steps of step S4 include:

[0086] S41: Define the input feature matrix of the local Gaussian process model;

[0087] S42: Based on the real-time feature field data collected by each vehicle on the navigation path in the target observation region, obtain the target output vector;

[0088] S43: Use the input feature matrix and the target output vector to update the hyperparameters through the optimization function and update the local Gaussian process model;

[0089] S44: Input the input feature matrix into the updated local Gaussian process model to generate the local feature field prediction data of each vehicle.

[0090] Among them, by using the input feature matrix and the target output vector, the hyperparameters are updated through an optimization function to update the local Gaussian process model, which specifically further includes:

[0091] Calculate the covariance matrix K: According to the input two-dimensional feature matrix X and the covariance function k(x, x′), calculate the covariance matrix K; for example, the covariance function adopts the isotropic Rational Quadratic (RQ) kernel function:

[0092]

[0093] In the formula, σ 2 is the signal variance, α is the shape parameter, and l is the length scale. It should be noted that other kernel functions can also be selected according to requirements.

[0094] In the embodiments of the present application, the logarithmic marginal likelihood estimation method is used for hyperparameter optimization, and the Laplace approximation method, the Leave-One-Out approximation method, the sparse likelihood approximation method, etc. can also be used for optimization.

[0095] Taking the logarithmic marginal likelihood estimation method as an example for illustration:

[0096] Calculate the logarithmic marginal likelihood function: According to the likelihood function Calculate the logarithmic marginal likelihood function:

[0097]

[0098] where y is the output target vector, X is the input two-dimensional feature matrix, K is the covariance matrix, is the noise variance, I is the identity matrix, and n is the number of data points.

[0099] Optimize the hyperparameters: By maximizing the logarithmic marginal likelihood function, optimize the hyperparameters; during the optimization process, continuously adjust the hyperparameters until the logarithmic marginal likelihood function reaches the maximum value or meets the convergence condition.

[0100] Finally, through the above optimization process, the updated hyperparameter values are obtained. The updated hyperparameter values will replace the initial hyperparameters and be used to update the local Gaussian process regression model. The updated hyperparameters can better describe the distribution characteristics of the data and improve the prediction accuracy of the model.

[0101] Use the updated local Gaussian process regression model for prediction, and the prediction distribution is:

[0102] f * |X, y, X * ~N(m * , ∑ * );

[0103] where m * is the predicted mean, and Σ * is the predicted covariance matrix; through the prediction distribution, the local feature prediction field data can be obtained for the selection of the next path point of the autonomous underwater vehicle.

[0104] In this application, by dynamically updating the parameters in the local Gaussian process model, the accuracy of the local environmental information is gradually improved. As time goes by and the amount of data increases, the model is gradually corrected and optimized, making the accuracy of local observations continuously improved. Through the updated Gaussian process regression model, the local feature field prediction information is continuously obtained for the selection of the next step of the autonomous underwater vehicle.

[0105] Referring to Figure 5 , in step S5, based on the local feature field prediction data of each vehicle, the improved ant colony algorithm is used to plan the navigation path of each vehicle, which specifically includes:

[0106] S51: Define the pheromone matrix of the ant colony algorithm and obtain the pheromone concentration of the target observation area;

[0107] S52: Based on the local feature field prediction data of the current path point of the vehicle, calculate the feature identification criterion value of the change of the feature values of the current path point and the adjacent path points of the vehicle with the spatial position;

[0108] S53: Based on the preset state transition function, combine the traveling direction of the vehicle from the current path point to the adjacent path point, the feature identification criterion value, and the pheromone concentration of the target observation area, calculate the transfer probability from the current path point to the adjacent path point, and determine the next path point based on the transfer probability;

[0109] S54: Obtain the real-time feature field data of the transfer from the current path point to the next path point, determine the change value of the feature field data, update the pheromone matrix based on the change value of the feature field data, and a single ant completes a path planning;

[0110] S55: When all ants complete the path planning, determine the best navigation path through the preset path evaluation function as the navigation path of each vehicle.

[0111] Furthermore, after determining the next path point based on the transfer probability, it further includes:

[0112] Check whether the vehicle has left the target observation area. If it has left the target observation area, start the rediscovery behavior, detect the adjacent path points in a preset search method, and detect the adjacent path points that meet the preset conditions as the next path point.

[0113] Illustrate with an example, please refer to Figure 6, Initialize the state of each vehicle and the parameters of the improved ant colony algorithm. Among them, the parameters of the improved ant colony algorithm include the basic parameters of the ant colony algorithm, such as the number of ants, the initial pheromone concentration, the pheromone evaporation factor, the heuristic factor, the maximum number of iterations, etc.; in each iteration, each ant uses the local Gaussian process model according to the current position to obtain the local feature field prediction data of each vehicle centered on the current position; the local feature field prediction data includes temperature eigenvalue, salinity eigenvalue or flow field eigenvalue; in the local feature field prediction data, combine the traveling direction of the vehicle from the current point to the adjacent path point, the feature identification criterion value of the feature difference between the current point and the next point changing with the spatial position, and the pheromone concentration of the target observation area to calculate the transfer probability of the next point.

[0114] The feature difference ΔT is expressed as:

[0115] ΔT = |T(x curren x)-T(x next )|,

[0116] where, T(x current ) is the eigenvalue of the current position, and T(x next ) is the eigenvalue of the candidate position.

[0117] Check whether the current point is in the taboo list. If it is in the taboo list, skip this point and select the next point; this taboo list is used to record the points that the vehicle has visited and the points within the preset range of the visited points to ensure that the recently visited nodes are not selected repeatedly to avoid circular paths.

[0118] Use the roulette wheel algorithm to select the next point according to the calculated transfer probability;

[0119] By selecting the next point according to the transfer probability and combining the roulette wheel algorithm, the diversity of the path can be increased and getting stuck in the local optimum can be avoided.

[0120] Determine the position of the next point, and add the current point and the adjacent points within the preset range of the current point to the taboo list;

[0121] Based on the position of the next point, the vehicle moves to the next point and obtains the eigenvalue data on the path as real-time update data;

[0122] After each step of movement of the vehicle, update the pheromone matrix based on the eigenvalue data on the path;

[0123] After each step of movement, update the pheromone concentration on the path according to the collected feature differences. For example, in the path segment where the feature change is significant, increase the pheromone concentration to guide the subsequent autonomous underwater vehicle to preferentially select these paths.

[0124] Based on the updated pheromone matrix, the best path is updated through a path evaluation function, and a path with a feature difference greater than a preset value between different features is selected as the new best path; including:

[0125] Update of the cumulative eigenvalue: The feature differences after each selection of the next point are accumulated and used as the basis for path evaluation. The cumulative eigenvalue change ΔT all The formula for is:

[0126] ΔT all = ΔT all + ΔT.

[0127] Path evaluation update: Compare the cumulative feature differences of each path, and select the path with a higher cumulative feature difference as the new best path. The formula for path evaluation update is:

[0128] ΔT all_best = max(ΔT all_best , ΔT all,k ),

[0129] where k represents the k-th path in the current iteration.

[0130] Detect whether all ants have completed a path selection. If not, repeat the above steps to complete one iteration;

[0131] Detect whether the current iteration count has reached the maximum iteration count. If it has reached the maximum iteration count, output the optimal path.

[0132] This application improves the state transition function and pheromone update method of the traditional ant colony algorithm according to eigenvalues, and realizes the optimized observation for target features.

[0133] Specifically, during the execution of step S6, based on the distribution of the navigation path, the coverage rate of the navigation path of the vehicle on the target observation area is calculated in real time, and the task execution status of each vehicle is determined based on the coverage rate;

[0134] When it is detected that the task execution status of the vehicle meets the preset task completion condition, a global Gaussian process model is constructed based on the real-time feature field data collected by all vehicles;

[0135] Use the global Gaussian process model to generate the global uncertainty information of the feature field, and allocate a new target observation area for the vehicle that has completed the task based on the global uncertainty information. Allocate a target observation area with high uncertainty for the vehicle that has completely observed the task for further secondary observation, thereby reducing the global uncertainty.

[0136] Furthermore, when the global uncertainty information is reduced to the preset uncertainty threshold, the observation tasks of all vehicles are ended.

[0137] In the embodiment of the present application, after the initial allocation is completed, during the process of the autonomous underwater vehicle moving and observing, when it is determined that a certain area has been observed, in order to optimize the observation, a second task allocation is performed. The autonomous underwater vehicles that have completed the observation task go to the area with an uncertainty greater than the preset uncertainty (i.e., high uncertainty) according to the uncertainty information, so that the observation area range of the multi-autonomous underwater vehicle collaborative observation is more extensive.

[0138] The task allocation of multi-autonomous underwater vehicles maximizes the observation of the unknown feature field based on the prediction information, dynamically allocates the observation tasks according to the online feature field prediction information, and at the same time uses the uncertainty information of the Gaussian process for task reallocation, making the observation more efficient and obtaining more useful sampling data.

[0139] The embodiment of the present application uses simple K_MEANS clustering and DBSCAN clustering to cluster the eigenvalues to achieve area allocation, and at the same time combines the uncertainty after the global Gaussian process to achieve secondary allocation, improving the observation efficiency of multi-autonomous underwater vehicles. According to the real-time prediction information of the feature field, the task allocation of multi-autonomous underwater vehicles is carried out. When the multi-autonomous underwater vehicles detect that a certain area has been observed during the observation, they will use the global Gaussian process regression to make a global prediction to determine the uncertainty of the global prediction information, and use this uncertainty to re-allocate the tasks again. After the re-allocation, the observation is carried out again to maximize the observation of the feature area, which is closer to the observation of the complex dynamic feature field by multi-autonomous underwater vehicles.

[0140] In the present application, local prediction is carried out through local Gaussian process regression, breaking through the limitation that the traditional adaptive path planning cannot cope with complex and dynamically changing environments. At the same time, it improves the problem that the global model prediction calculation in the current planning method based on the dynamic prediction information of the environmental characteristics of the model is complex and the calculation amount is large, realizing the purpose of quickly predicting with simple information to optimize the observation; at the same time, updating the hyperparameters of the model according to the real-time data to increase the reliability of the model prediction.

[0141] Based on the same idea as an autonomous underwater vehicle path planning method in the above embodiment, the present application also provides an autonomous underwater vehicle path planning system, which can be used to execute the above autonomous underwater vehicle path planning method. For the sake of illustration, in the structural schematic diagram of an embodiment of the autonomous underwater vehicle path planning system, only the parts related to the embodiment of the present application are shown. Those skilled in the art can understand that the illustrated structure does not constitute a limitation on the system, and it may include more or fewer components than those illustrated, or combine some components, or arrange different components.

[0142] Please refer to Figure 7, in another embodiment of the present application, an autonomous underwater vehicle path planning system is provided. The system includes: a data acquisition module 101, a task assignment module 102, a local model initialization module 103, a local feature field prediction module 104, a path planning module 105, and a task monitoring module 106;

[0143] Among them, the data acquisition module 101 is used to acquire initial feature field data, preprocess the initial feature field data, and obtain high-resolution feature field data;

[0144] The task assignment module 102 is used to use the high-resolution feature field data to assign target observation areas for multiple vehicles to perform observation tasks;

[0145] The local model initialization 103 is used to extract local feature field data around the current position of each vehicle based on the high-resolution feature field data, and initialize the local Gaussian process model corresponding to each vehicle's observation task based on the local feature field data;

[0146] The local feature field prediction module 104 is used to dynamically update the local Gaussian process model using the real-time feature field data collected by each vehicle, and predict the local feature field prediction data around each vehicle based on the dynamically updated local Gaussian process model;

[0147] The path planning module 105 is used to plan the navigation path of each vehicle based on the local feature field prediction data of each vehicle using an improved ant colony algorithm;

[0148] The task monitoring module 106 is used to monitor the task execution status of each vehicle based on the navigation path, and reassign tasks or end tasks based on the task execution status of all vehicles.

[0149] In some preferred embodiments, the data acquisition module 101 is specifically used to perform interpolation processing on the initial feature field data using the interpolation method to obtain high-resolution feature field data, and the resolution of the high-resolution feature field data is higher than that of the initial feature field data.

[0150] In some preferred embodiments, the task assignment module 102 is specifically used for:

[0151] Using a clustering algorithm to divide the high-resolution feature field data into multiple to-be-observed areas with different feature information, and calculating the feature identification criterion values of the to-be-observed areas;

[0152] Based on a preset feature identification criterion threshold, determining the interesting feature values from the feature identification criterion values, and marking the area corresponding to the clustering of the interesting feature values as the interesting area;

[0153] Perform secondary classification on the region of interest to determine multiple sub-regions of interest, and divide the multiple sub-regions of interest into several target observation regions to be observed according to the preset evaluation index and the number of vehicles;

[0154] According to the preset evaluation index of the target observation region, dispatch different numbers of vehicles to the divided target observation regions for observation.

[0155] It should be noted that an autonomous underwater vehicle path planning system of the present application corresponds one-to-one with the autonomous underwater vehicle path planning method of the present application. The technical features and beneficial effects described in the embodiments of the above autonomous underwater vehicle path planning method are applicable to the embodiments of the autonomous underwater vehicle path planning system. For specific content, reference can be made to the description in the method embodiments of the present application, which will not be repeated here. This is hereby declared.

[0156] In addition, in the implementation manner of the autonomous underwater vehicle path planning system of the above embodiment, the logical division of each program module is only an example. In practical applications, according to needs, for example, considering the configuration requirements of the corresponding hardware or the convenience of software implementation, the above functions can be allocated to different program modules to complete, that is, the internal structure of the autonomous underwater vehicle path planning system is divided into different program modules to complete all or part of the functions described above.

[0157] In another embodiment, an electronic device for implementing an autonomous underwater vehicle path planning method is provided, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor; when the processor executes the computer program, an autonomous underwater vehicle path planning method according to any embodiment of the present application is implemented.

[0158] Exemplarily, in this embodiment, the computer program can be divided into one or more modules, and the one or more modules are stored in the memory and executed by the processor to complete the present application. The one or more module elements can be a series of computer program instruction segments capable of completing specific functions, and the instruction segments are used to describe the execution process of the computer program in the device.

[0159] The device can be a computing device such as a desktop computer, a notebook, a palm computer, and a cloud server. The device may include, but is not limited to, a processor and a memory.

[0160] The processor may be a Central Processing Unit (CPU), or may also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The processor is the control center of the device, connecting various parts of the entire device using various interfaces and lines. The memory can be used to store the computer programs and / or modules. By running or executing the computer programs and / or modules stored in the memory, and by calling the data stored in the memory, the processor realizes various functions of the device. The memory mainly includes a program storage area and a data storage area. Among them, the program storage area can store an operating system, application programs required for at least one function, etc. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as a hard disk, internal memory, plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, Flash Card, at least one magnetic disk storage device, flash memory device, or other volatile solid-state storage devices.

[0161] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium used in the embodiments provided in the present application can include non-volatile and / or volatile memories. Non-volatile memories can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memories can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0162] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, 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, it should be considered as the scope recorded in this specification.

[0163] The above embodiments are preferred embodiments of the present application, but the embodiments of the present application are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications made without departing from the spirit and principle of the present application should be equivalent replacement methods and are all included in the protection scope of the present application.

Claims

1. An autonomous underwater vehicle path planning method, characterized in that, Including: Obtain initial feature field data, preprocess the initial feature field data to obtain high-resolution feature field data; Use the high-resolution feature field data to allocate target observation areas for multiple vehicles to perform observation tasks; Based on the high-resolution feature field data, extract local feature field data around the current position of each vehicle, and initialize a local Gaussian process model corresponding to each vehicle's observation task based on the local feature field data; Dynamically update the local Gaussian process model using the real-time feature field data collected by each vehicle, and predict the local feature field prediction data around each vehicle based on the dynamically updated local Gaussian process model; Based on the local feature field prediction data of each vehicle, use an improved ant colony algorithm to plan the navigation path of each vehicle; Based on the navigation path, monitor the task execution status of each vehicle, and re-allocate tasks or end tasks based on the task execution status of all vehicles.

2. The autonomous underwater vehicle path planning method according to claim 1, characterized in that Obtain initial feature field data, preprocess the initial feature field data to obtain high-resolution feature field data, including: Obtain the initial feature field data of the area to be observed, and perform interpolation processing on the initial feature field data using a preset interpolation method to obtain high-resolution feature field data, and the resolution of the high-resolution feature field data is higher than that of the initial feature field data.

3. The autonomous underwater vehicle path planning method according to claim 1, characterized in that Use the high-resolution feature field data to allocate target observation areas for multiple vehicles to perform observation tasks, including: Based on the high-resolution feature field data, divide multiple areas to be observed with different feature information through a clustering algorithm, and calculate the feature identification criterion values of the areas to be observed; Based on a preset feature identification criterion threshold, determine the interesting feature values from the feature identification criterion values, and mark the area corresponding to the clustering of the interesting feature values as the interesting area; Perform secondary classification on the interesting area to determine multiple sub-interesting areas, and divide the multiple sub-interesting areas into several target observation areas to be observed according to a preset evaluation index and the number of vehicles; According to the evaluation results of the target observation areas, dispatch different numbers of vehicles to the divided target observation areas for observation.

4. The autonomous underwater vehicle path planning method according to claim 1, characterized in that Dynamically update the local Gaussian process model using the real-time feature field data collected by each vehicle, and predict the local feature field prediction data around each vehicle based on the dynamically updated local Gaussian process model, including: Define the input feature matrix of the local Gaussian process model; Based on the real-time feature field data collected by each vehicle on the navigation path in the target observation area, obtain the target output vector; Use the input feature matrix and the target output vector to update the hyperparameters through an optimization function to update the local Gaussian process model; Input the input feature matrix into the updated local Gaussian process model to generate the local feature field prediction data of each vehicle.

5. The autonomous underwater vehicle path planning method according to claim 1, characterized in that, Based on the local feature field prediction data of each vehicle, use an improved ant colony algorithm to plan the navigation path of each vehicle, including: Define the pheromone matrix of the ant colony algorithm, and obtain the pheromone concentration of the target observation area; Based on the local feature field prediction data of the current path point of the vehicle, calculate the feature identification criterion value of the eigenvalues of the current path point and adjacent path points of the vehicle changing with spatial position; Based on a preset state transition function, combine the traveling direction of the vehicle from the current path point to the adjacent path point, the feature identification criterion value, and the pheromone concentration of the target observation area to calculate the transition probability from the current path point to the adjacent path point, and determine the next path point based on the transition probability; Obtain the real-time feature field data of the transition from the current path point to the next path point, determine the change value of the feature field data, update the pheromone matrix based on the change value of the feature field data, and a single ant completes one path planning; When all ants complete path planning, determine the optimal navigation path through a preset path evaluation function as the navigation path of each vehicle; 6. The autonomous underwater vehicle path planning method according to claim 5, wherein, After determining the next path point based on the transition probability, it includes: Check whether the vehicle has left the target observation area. If it has left the target observation area, start a rediscovery behavior, detect adjacent path points in a preset search manner, and detect adjacent path points that meet the preset conditions as the next path point; 7. The autonomous underwater vehicle path planning method according to claim 1, characterized in that Based on the navigation path, monitor the task execution status of each vehicle, and perform task reallocation or end the task based on the task execution status of all vehicles, including: Based on the distribution of the navigation path, calculate in real time the coverage rate of the navigation path of the vehicle for the target observation area, and determine the task execution status of each vehicle based on the coverage rate; When it is monitored that the task execution status of the vehicle meets the preset task completion condition, construct a global Gaussian process model based on the real-time feature field data collected by all vehicles; Generate the global uncertainty information of the feature field using the global Gaussian process model, and allocate a new target observation area for the vehicles that have completed the task based on the global uncertainty information; 8. The autonomous underwater vehicle path planning method according to claim 7, characterized in that, It also includes: When the global uncertainty information drops to a preset uncertainty threshold, end the observation tasks of all vehicles; 9. An autonomous underwater vehicle path planning system, characterized in that, For executing the method according to any one of claims 1 to 8, it includes: A data acquisition module for acquiring initial feature field data, preprocessing the initial feature field data to obtain high-resolution feature field data; A task allocation module for using the high-resolution feature field data to allocate target observation areas for multiple vehicles to perform observation tasks; A local model initialization module for extracting local feature field data around the current position of each vehicle based on the high-resolution feature field data, and initializing the local Gaussian process model corresponding to each vehicle to perform observation tasks based on the local feature field data; A local feature field prediction module for dynamically updating the local Gaussian process model using the real-time feature field data collected by each vehicle, and predicting the local feature field prediction data around each vehicle based on the dynamically updated local Gaussian process model; A path planning module for planning the navigation path of each vehicle using an improved ant colony algorithm based on the local feature field prediction data of each vehicle; A task monitoring module, configured to monitor the task execution status of each vehicle based on the navigation path, and perform task reallocation or end tasks based on the task execution status of all vehicles.

10. An electronic device, characterized in that, The electronic device includes: At least one processor; and a memory communicatively connected to the at least one processor; Wherein, the memory stores computer program instructions executable by the at least one processor, and the computer program instructions are executed by the at least one processor so that the at least one processor can execute the autonomous underwater vehicle path planning method according to any one of claims 1 to 8.

Images