Path Planning Method for Autonomous Underwater Vehicle Based on Double Neural Network Reinforcement Learning

Abstract

The invention discloses a path planning method for an autonomous underwater vehicle based on dual neural network reinforcement learning, and solves the path planning problem of the autonomous underwater vehicle based on the dual neural network reinforcement learning algorithm. Aiming at the problems of large storage space and long search time in the Q-learning learning algorithm, optimize the processing; on the basis of the Q-learning learning algorithm, integrate the target network and memory pool experience playback technology to obtain a double neural network reinforcement learning algorithm, and build a network based on AUV intelligent path planning framework of dual neural network reinforcement learning algorithm; quantitatively process the environmental state information of autonomous underwater vehicles, combine it with dual neural network reinforcement learning algorithm, and judge the relationship between the AUV movement direction and the position of the target point, and obtain the AUV Decision-making for intelligent planning paths. The invention significantly reduces the complexity of calculation, meets the requirements of real-time decision-making, has outstanding migration ability and environmental adaptability, and provides a safe, fast and reliable path planning scheme for AUV.

Description

technical field [0001] The invention belongs to the field of artificial intelligence and path planning, and in particular relates to a path planning method for an autonomous underwater vehicle based on double neural network reinforcement learning. Background technique [0002] With the gradual development of the field of artificial intelligence, more and more experts pay attention to and study how to use artificial intelligence technology to solve the path planning problem of autonomous underwater vehicles, and a large number of intelligent algorithms have emerged in the scientific community, including deep learning, reinforcement Learning, etc., make the path planning of Autonomous Underwater Vehicle (AUV) more and more accurate. However, what AUV ultimately faces is how to navigate accurately in an unknown environment, so when studying how AUV can improve its own performance, it is also necessary to fully consider the problem of accurately navigating in an unknown environm...

AI Technical Summary

Problems solved by technology

[0003] Purpose of the invention: In order to solve the problems of self-learning ability and environmental self-adaptation in the face of unknown environment existing in AUV path planning in the prior art, the present invention provides an autonomous underwater vehicle path planning method based on double neural network reinforcement learning

Images