Sliding mode adaptive control for ship path following with sideslip angle observer

An adaptive sliding mode control algorithm based on radial basis function neural networks(RBF-NNs) is proposed to solve the problems of external disturbances, internal model uncertainty and sideslip angle unknown in the process of underactuated ship path following. Firstly, the three dimensional path following control is transformed into one dimensional heading control by backstepping algorithm. Because the sideslip angle is unknown, the RBF-NNs are used to estimate the sideslip angle. Taking the ship's heading error and the yaw rate error as the input of neural network, 10 hidden layers are used to classify the input linearly, and an adaptive law is designed to update the weight online to solve the problem of unknown sideslip angle. The linear extended state observer (LESO) is introduced to estimate the total disturbances in the path following process, so as to solve the problem of external disturbances and internal model uncertainty. Finally, the mathematical model group(MMG) model is used for simulation. The simulation results show that the designed controller can follow the preset waypoints well, and the RBF-NNs can estimate the sideslip angle well, LESO can well observe the total disturbances. It is demonstrate that this scheme can provide a potential reference for the path following of underactuated ships with unknown sideslip and disturbances.

Automated summary

An adaptive sliding mode control algorithm based on radial basis function neural networks(RBF-NNs) is proposed to solve the problems of external disturbances, internal model uncertainty and sideslip angle unknown in the process of underactuated ship path following. The RBF-NNs are used to estimate the sideslip angle, while the linear extended state observer (LESO) is introduced to estimate the total disturbances. This approach provides a potential reference for underactuated ship path following with unknown sideslip and disturbances.