Underwater autonomous motion control of a small-scaled spherical robot with neural networks

Considering the complex and variability of the operating environment of underwater spherical robot, usually it is difficult to solve the control problem when the robot changes its motion state or it is subject to waves and ocean currents, in those cases wherein robots are subject to continuous parametric changes or external disturbances, online gains tuning is a desirable choice. In this paper, with the goal of supporting some autonomous tasks of our small-scaled spherical robot, such as ecological observations and intelligent surveillance, a neural network-based auto-tuning control system was designed and implemented, which has a great advantage of processing online for the robot due to their nonlinear dynamics. The neural network plays the role of automatically estimating the suitable set of control gains that achieves the stability of the system. Simulation results are presented for the underwater swimming, in terms of the motion performance, stability, and velocity of the robot. Finally, the effectiveness of the proposed method was demonstrated by showing that the underwater horizontal and desired triangular trajectory motion were stable, and the design presented in this paper is able to meet future demands of underwater robots in biological monitoring and multi-robot cooperation.