Multi-objective optimization of remotely operated vehicle control system using surrogate modeling

Abstract

This paper discussed the idea of using surrogate modeling to optimize a multi-objective problem. The proposed method is adopted on PD controllers of a remotely-operated vehicle RRC ROV II designed by the Robotic Research Centre in the Nanyang Technological University (NTU). The main emphasis of this study is to approximate a set of controller parameters from a few sample and search for Pareto front. Through the simulation, Radial Basis Function Neural Network (RBFNN) was able to give a good approximation to the Pareto-front of controller parameters and perform about three times faster compare by using brute-force search approach.