Unmanned ship autonomous navigation method based on deep reinforcement learning and genetic algorithm

Abstract

The invention discloses an unmanned ship autonomous navigation method based on deep reinforcement learning and a genetic algorithm and belongs to the field of an unmanned ship. The method comprises two parts: path planning and path tracking, wherein in the path planning, obtaining an overview of the environment through satellite images in advance, and predetermining a start point and an end pointof a planned path; through an elitist-genetic algorithm, obtaining discrete, ordered and optimal Nw barrier-free way points in any environment, and carrying out fitting on the start point, the Nw barrier-free way points and the end point through a K order B spline algorithm formula to obtain a continuous unmanned ship path planning curve; and according to all parameters of the unmanned ship path planning curve and a state immediate reward function at any time as well as a kinematics model and a dynamics model of the unmanned ship, obtaining control torque of the unmanned ship based on the self-learning ability of the deep deterministic strategy gradient in deep reinforcement learning in finite time, and carrying out the unmanned ship path tracking. Through the elitist-genetic algorithm, moderate number of safe path points can be obtained in a narrow environment.

Description

Technical field [0001] The present invention relates to the field of unmanned ships, and in particular, to a method for autonomous navigation of unmanned ships based on deep reinforcement learning and genetic algorithms. Background technique [0002] With the widespread application of automation theory and practice in ocean engineering, underactuated surface vessels, as a highly autonomous unmanned vehicle, can flexibly and conveniently complete a series of high-risk ocean tasks. Among them, high-precision autonomous navigation technology plays a crucial role in developing the autonomy of under-actuated surface unmanned ships. In fact, autonomous navigation problems can be well solved by combining path planning with path tracking synthesis. In the design of path planning subsystem, many traditional path search algorithms have been proposed. Sampling-based algorithms, such as random expansion trees, probabilistic roadmap path planners, etc. Through random sampling points i...

AI Technical Summary

Problems solved by technology

In the system design of path tracking, backstepping method, singular perturbation method, fuzzy partition method and various adaptive control methods have been effectively used in ship motion control, but these control methods need to design different controllers according to the task, This reduces the flexibility and mobility of the control system. In the existing autonomous navigation methods, real-time control is usually performed through cameras or radars, which makes the overall operating state of the unmanned ship largely dependent on the sensor. performance and computing power of the on-board chip

Images