Simulation Analysis of K-modification Adaptive Control for Remotely Operated Underwater Vehicles

Due to the non-linear properties of the underwater dynamics, a controller with the ability to adapt to unknown uncertainties is considered as the most important characteristic of an adaptive controller. With that consideration, there is one other factor which affects the adaptation of a system, i.e., the adaptation learning rates. A decade old question that is hovering in adaptive control literature is, how to achieve fast adaptation with higher adaptation gains? This paper presents a novel adaptive control architecture for underwater vehicles, which is called k-modification in adaptive control. The proposed control scheme shows that it could achieve fast adaptation with zero oscillations in the control signal. The adaptive controller was tested for disturbances, for thruster failure and for buoyancy changes, and the results were compared with the results of the standard adaptive controller Model Reference Adaptive Controller (MRAC) for the same scenarios. The results showed the differences between the two controllers. MRAC was performing better at low adaptation gains, while the k-mod controller was performing better at high adaptation gains.