Neural network based reinforcement learning control of autonomous underwater vehicles with control input saturation

2014 ◽  
Author(s):  
Rongxin Cui ◽  
Chenguang Yang ◽  
Yang Li ◽  
Sanjay Sharma
Keyword(s):  
Neural Network ◽  
Reinforcement Learning ◽  
Autonomous Underwater Vehicles ◽  
Input Saturation ◽  
Learning Control ◽  
Underwater Vehicles ◽  
Control Input

Neural Network-Based Tracking Control of Underactuated Autonomous Underwater Vehicles With Model Uncertainties

2014 ◽  
Vol 137 (2) ◽  
Author(s):  
Bong Seok Park
Keyword(s):  
Neural Network ◽  
Tracking Control ◽  
Control Method ◽  
Autonomous Underwater Vehicles ◽  
Underwater Vehicles ◽  
Reference Trajectory ◽  
Control Input ◽  
Model Uncertainties ◽  
Dynamic Surface ◽  
Hydrodynamic Damping

In this paper, we propose a neural network (NN)-based tracking control method for underactuated autonomous underwater vehicles (AUVs) with model uncertainties. In order to solve the difficulties in designing the controller for underactuated AUVs, the additional virtual control input is developed, and the approach angle, which generates the desired yaw angle to track any reference trajectory, is introduced. Moreover, the NNs are used to deal with model uncertainties in the hydrodynamic damping terms of AUVs. Finally, the proposed controller is designed based on the dynamic surface control (DSC) method, and the boundedness of all tracking errors is proved by using the Lyapunov stability theory. Some simulation results demonstrate the performance of the proposed control method.

Sign in / Sign up

Export Citation Format

Share Document