Rotation tracking control strategy of underwater flexible telescopic manipulator based on neural network compensation for water environment disturbance

The underwater flexible telescopic manipulator (UFTM) can be installed on a submersible to achieve underwater operation, which is conducive to the exploration and development of deep-sea resources. However, the motion of the UFTM is affected by underwater disturbance torque, which may lead to a decrease in operational accuracy. In addition, the flexible components in the UFTM, including flexible deformation and torsional deformation, will exacerbate fluctuations in the rotation angle underwater environment disturbance torque. In this paper, a control method using neural networks to identify underwater disturbance torques is proposed to improve the rotational tracking accuracy of the UFTM. Firstly, based on the assumed modal method (AMM) and Lagrange equation, the UFTM dynamic model considering two-dimensional deformation and underwater disturbance torque is derived. Then, based on the UFTM dynamic model and Lyapunov theorem, the adaptive law and control law of neural networks are designed. Finally, simulation and control experiments prove the effectiveness of the neural network compensation for underwater disturbance torque control strategy. Compared with the classical PID strategy and the sliding mode control, the control strategy of neural network compensation for underwater disturbance can effectively improve the tracking control accuracy of the UFTM.