A motion control method for nautical ships based on reinforcement learning intelligent algorithm

Abstract

A reinforcement learning intelligent algorithm-based navigation ship motion control method disclosed by the present invention comprises the steps of utilizing a multi-mode sensor to acquire various types of important data of the ships; filtering and fusing the obtained data information to obtain various states of data satisfying the precision requirements, wherein the data comprises the attitude information of the ships, the actual positions of the ships, the ship speed, the sea surface wind speed, the ocean current speed and the like; evaluating the actual state data via an evaluation function to obtain an award value at the current moment; learning continuously via a reinforcement learning intelligent algorithm, and finally obtaining the motion control rules of the ships on the differentenvironmental state conditions. The ship motion control method of the present invention is an optimization algorithm that does not depend on an environment model, and compared with a conventional motion control algorithm, is more flexible to control, and can optimize and learn continuously during the usage.

Description

technical field [0001] The invention relates to the field of nautical ships, in particular to a motion control method for a nautical ship based on an enhanced learning intelligent algorithm. Background technique [0002] With the prosperity of the world economy and trade, ships play an increasingly important role in the transportation process, but their high energy consumption also limits the development of long-distance transportation of ships. According to estimates, the fuel costs of shipping companies account for about 40% of the total operating costs. The annual fuel cost of a large shipping company can reach tens or even tens of billions. Ships need to formulate a sailing plan before sailing and adjust it as real-time conditions change; however, there are more than 40 factors that affect the energy saving and emission reduction of ships, and the relationship is intricate, and the real-time and effective dynamic control of ships is one of the most important key point....

AI Technical Summary

Problems solved by technology

[0002] With the prosperity of the world economy and trade, ships play an increasingly important role in the transportation process, but their high energy consumption also limits the development of long-distance transportation of ships

Images