Multi-objective optimization for AUV conceptual design based on NSGA-II

AUV (autonomous underwater vehicle) design is requiring the coordination of many different disciplines, which involved in assemblity, mission and performance. Conceptual design is the least defined stage of the AUV design process. An approach for multidisciplinary optimization, multi-objective optimization study of AUV's principal parameters in conceptual design is proposed in present analysis. First, a MDO (Multi-Disciplinary Optimization) flow is constructed for AUV design circle. The total system can be divided into one system control level and 6 sub-systems. Then, a Non-dominated Sorting Genetic Algorithm II (NSGA-II) is employed to search the design space for optimal, feasible designs by considering three measures of performance (MOPs): cost, effectiveness, and risk. The result shows that, NSGA-II is effective in multi-objective optimization of an AUV conceptual design.