An on-line adaptive hybrid PID autopilot of ship heading control using auto-tuning BP & RBF neurons

Ship navigation has the characteristics of nonlinearity, uncertainties and disturbances, which cause difficulties in ship heading control. It is presented a new hybrid PID (Proportional-Integral-Derivative) control technique for the ship navigation systems in this paper, the hybrid PID algorithm is composed of on-line adaptive adjustable parameters of the auto-tuning neurons based on BP (Back Propagation) and RBF (Radial Basic Function). The hybrid activation function is a modified hyperbolic tangent function, which is used as the auto-tuning NN (Neural Network), it is adjusted the magnitude and the shape of the Sigmoid and Gaussian function. The on-line adaptive hybrid PID autopilot adopts steepest descent adaptation laws and is tuned on-line; the operation results show the effectiveness of the proposed ship heading controller.