Unmanned boat heading active disturbance rejection control system based on RBF neural network

Abstract

The invention discloses an unmanned boat heading active disturbance rejection control system based on an RBF neural network. The system comprises a sensor magnetic compass for collecting practical heading angle information of an unmanned boat; an active disturbance rejection unit for receiving the practical heading angle information transmitted by the sensor magnetic compass, set heading angle information transmitted by a user and control quantity information of the unmanned boat; and an RBF neural network identifier for receiving the practical heading angle information transmitted by the sensor magnetic compass and the control quantity information of the unmanned boat. According to the system, through utilization of the active disturbance rejection unit, heading control over the unmannedboat is realized, through utilization of the RBF neural network identifier, an error feedback coefficient beta1 and a differential error feedback coefficient beta2 in the active disturbance rejectionunit are optimized, parameter adaptation is realized, unmanned boat heading control precision is improved, and system robustness is improved.

Description

technical field [0001] The invention relates to the field of navigation control of unmanned ships, in particular to an RBF neural network-based active disturbance rejection control system for unmanned ships heading. Background technique [0002] The unmanned boat is an intelligent motion platform that can realize autonomous navigation at sea and complete designated tasks. It can perform extremely dangerous tasks such as mine clearance, investigation, surveillance, and anti-terrorism, reducing casualties. In the future, it can also cooperate with other The unmanned platforms perform tasks together. In order to complete specific tasks, unmanned ships must have good motion characteristics, and the heading control of unmanned ships is the basis of its motion control research. The development of boats is of great significance. [0003] At present, there are many methods for the course control of unmanned ships, such as PID control, adaptive control and other conventional contro...

AI Technical Summary

Problems solved by technology

[0003] At present, there are many methods for the course control of unmanned ships, such as PID control, adaptive control and other conventional control methods, but due to the complexity of the unmanned ship model, the uncertainty of model parameters, and the external environment Interference, the design is only effective within a certain range, the anti-interference ability is poor, and the robustness is not strong

Active Disturbance Rejection Control (ADRC) is widely used because it does not rely on specific models and has the ability to estimate unknown disturbances. It can even be applied to the course control of unmanned ships. However, due to its complex structure and many internal parameters, It is difficult to adjust the parameters to the optimum. In order to achieve accurate heading control of unmanned ships, a large number of experiments are required, and repeated adjustment of parameters can be successful, which undoubtedly increases the difficulty of experiments.

Images