Learning and optimization using the clonal selection principle
IEEE Transactions on Evolutionary Computation
| Hi-index | 0.00 |
Autonomous underwater vehicles(AUVs) have been drawing increasing interests in various marine applications such as coastal structure inspection, sea floor exploration, and oceanographic monitoring. Due to the complexity of underwater stream dynamics and the prevalence of unexpected underwater obstacles, it is imperative to develop self-adjustable, intelligent navigation control functions for AUVs. Among various control techniques, we focus on the proportional-integral-derivative(PID) controller since it is still one of the dominant techniques in actual underwater vehicle control systems. We propose to apply the Clonal Selection Algorithm to determine optimal combination of three gain coefficients, KP, KD, KI of the PID controller. Our simulation shows that the proposed technique provides better responses than the existing Ziegler-Nichols technique with respect to the settling time, overshoot and an affinity in submerging under water and turning the yaw angle through simulation. We expect that AUVs could autonomously regulate three coefficients of six degree-of-freedom(DOF) PID controllers through real-time onboard processing.