CN-121994234-A - Unmanned underwater vehicle autonomous decision-making route planning method and system

CN121994234ACN 121994234 ACN121994234 ACN 121994234ACN-121994234-A

Abstract

The invention provides an autonomous decision-making route planning method and system for an unmanned underwater vehicle, which comprise the steps of acquiring environmental data in real time by utilizing various sensors carried by the unmanned underwater vehicle, realizing autonomous decision-making route planning of the unmanned underwater vehicle under a dynamic change environment by combining decision-making capability of a dynamic Bayesian network and planning capability of a path nearest junction algorithm according to the real-time sensor data, and being strong in dynamic adaptability and capable of coping with the environmental change in real time. The dynamic Bayesian network can continuously update the environment state according to the collected information of internal waves, mesoscale vortex, dense-jump layers, offshore obstacles and the like, and adjust the path planning according to the real-time information, so that the change of the dynamic environment is effectively treated.

Inventors

CHENG XIMING
SHEN ZHIYU
JIANG HUAI
JIN MANMAN
CAO JINGPU

Assignees

北京中安智能信息科技有限公司

Dates

Publication Date: 20260508
Application Date: 20260121

Claims (10)

1. An autonomous decision-making route planning method for an unmanned underwater vehicle, comprising the steps of: acquiring environmental data in real time by utilizing various sensors carried by the unmanned underwater vehicle; Constructing a dynamic Bayesian network according to the real-time sensor data; Inputting the data acquired by the sensor into a Bayesian network for reasoning, and calculating posterior probability distribution of each node to obtain the current state of the environment; predicting the current environment state of the unmanned underwater vehicle according to the Bayesian network reasoning result, and obtaining the optimal sailing path and adjustment scheme through reasoning; Generating a plurality of candidate paths according to the positions of a plurality of target points and the dynamic environment change; Calculating intersection points among a plurality of target paths, and selecting the intersection point closest to the current position of the unmanned underwater vehicle as a selection point of an optimal path; Selecting an optimal path from a plurality of target path junction points, and optimizing a navigation path through a CPA algorithm; According to the environment information updated in real time by the Bayesian network, dynamically adjusting the current path of the unmanned underwater vehicle to ensure that the unmanned underwater vehicle always runs along the optimal path; And generating a course and speed control instruction according to the path planning result, and implementing path adjustment through an execution system of the unmanned underwater vehicle.
2. The unmanned underwater vehicle autonomous decision-making route planning method according to claim 1, further comprising feeding back the route planning effect in real time according to the actual navigation situation and environmental change of the unmanned underwater vehicle, and enabling the route planning to be more accurate by continuously iterating and optimizing the bayesian network and the CPA algorithm.
3. The unmanned underwater vehicle autonomous decision-making route planning method according to claim 2, wherein the decision-making process of the dynamic bayesian network and the CPA algorithm is continuously optimized through feedback learning in the long-time sailing process, so that the self-adaption capability and the path planning efficiency of the system are improved.
4. An unmanned underwater vehicle autonomous decision routing method as claimed in claim 1, wherein the environmental data includes water flow rate, temperature, obstacle location multidimensional information.
5. An unmanned underwater vehicle autonomous decision routing method as claimed in claim 1, wherein the nodes of the bayesian network include water flow rate, obstacle location, target location, and the nodes represent conditional dependencies of environmental variables by directed edges.
6. The unmanned underwater vehicle autonomous decision-making routing method of claim 1, wherein the candidate paths combine with water flow, obstacle dynamics, ensuring that each path can accommodate real-time environmental changes.
7. The unmanned underwater vehicle autonomous decision-making routing method of claim 1, wherein the optimal path combines the avoidance of obstacles and the shortest path factor of the target point.
8. An unmanned underwater vehicle autonomous decision-making routing system, the unmanned underwater vehicle autonomous decision-making routing method according to any of claims 1-7, comprising: the data acquisition module is used for acquiring marine environment data in real time; the environment modeling module is used for constructing a dynamic Bayesian network; the reasoning module is used for carrying out environment reasoning according to the acquired data; the CPA path planning module is used for generating an optimal intersection point and optimizing a route; the control execution module is used for generating navigation control instructions and implementing path adjustment; and the feedback optimization module is used for iterating the optimization algorithm according to the actual sailing situation.
9. An electronic device, comprising: A memory and a processor; The memory is configured to store computer executable instructions that, when executed by the processor, implement the steps of an autonomous decision-making routing method for an unmanned underwater vehicle according to any of claims 1 to 7.
10. A computer readable storage medium storing computer executable instructions which when executed by a processor implement the steps of an autonomous decision routing method for an unmanned underwater vehicle as claimed in any of claims 1 to 7.

Description

Unmanned underwater vehicle autonomous decision-making route planning method and system Technical Field The invention relates to the field of underwater vehicles, in particular to an autonomous decision-making route planning method and system for an unmanned underwater vehicle. Background With the continuous development of unmanned underwater vehicle technology, the unmanned underwater vehicle autonomous navigation system has increasingly wide application in the fields of ocean exploration, environmental monitoring and the like, the autonomous navigation capability of the unmanned underwater vehicle is one of key factors for improving the operation efficiency and the task completion degree of the unmanned underwater vehicle, the planning and decision-making of the unmanned underwater vehicle navigation path depend on an effective path planning algorithm and an adaptive decision-making mechanism in a complex and dynamic ocean environment, however, the conventional unmanned underwater vehicle autonomous navigation planning technology still faces a plurality of challenges in practical application, and particularly, the dynamic adaptability, the accuracy and the reliability of the path planning and decision-making mechanism. At present, research on autonomous path planning of unmanned underwater vehicles mainly focuses on path planning methods based on static environment models and static decision mechanisms, for example, classical A-algorithm, dijkstra algorithm and other methods are widely applied to path planning of unmanned underwater vehicles, the methods generally assume that environments are static and perform path calculation based on preset maps or models, however, in practical application, the marine environment is full of complex dynamic factors such as ocean currents, internal waves, mesoscale vortexes, obstacle movement and the like, so that the static path planning methods cannot effectively cope with environmental changes, and the planned path may not be optimal or insufficient in adaptability. To cope with this problem, methods based on a Dynamic Bayesian Network (DBN) and a path closest junction (CPA) algorithm have been gradually introduced, but these methods have some limitations, respectively. The dynamic Bayesian network is a method for processing time series data and dynamic change uncertainty through a probability map model, is widely applied to state prediction and decision in a complex system, is used for modeling environment uncertainty in path planning of an unmanned underwater vehicle, particularly for processing the change which is difficult to accurately predict in marine environment, and can update environment states according to current sensor data through an inference mechanism to provide real-time feedback for path planning. The method shows better advantage in coping with the multi-objective problem because the method can find a natural intersection point among a plurality of objectives, reduce the complexity of path selection and improve the efficiency. The existing dynamic Bayesian network and CPA algorithm have good results in respective fields, but are still challenged when being used independently, particularly the problem that multi-objective and real-time path planning cannot be effectively conducted in a dynamic environment is solved, the dynamic Bayesian network can handle uncertainty and the dynamic environment, but has the defects of low calculation efficiency and insufficient adaptability in the multi-objective path planning, and the CPA algorithm has higher efficiency in the multi-objective path planning, but has no flexibility in the dynamic and variable environments and is easily affected by environmental changes. Disclosure of Invention The invention aims to overcome the defects in the prior art, and provides an autonomous decision-making route planning method and system for an unmanned underwater vehicle. In order to solve the technical problems, the invention provides the following technical scheme: in a first aspect, the present invention provides a method for autonomous decision-making route planning for an unmanned underwater vehicle, comprising: acquiring environmental data in real time by utilizing various sensors carried by the unmanned underwater vehicle; Constructing a dynamic Bayesian network according to the real-time sensor data; Inputting the data acquired by the sensor into a Bayesian network for reasoning, and calculating posterior probability distribution of each node to obtain the current state of the environment; predicting the current environment state of the unmanned underwater vehicle according to the Bayesian network reasoning result, and obtaining the optimal sailing path and adjustment scheme through reasoning; Generating a plurality of candidate paths according to the positions of a plurality of target points and the dynamic environment change; Calculating intersection points among a plurality of target paths, and selecting the interse