抄録
Evolutionary Multi-objective Optimization (EMO) is expected to be a powerful optimization framework for real world problems such as engineering design. Recent progress in automatic control and instrumentation provides us with a smart environment called Hardware In the Loop Simulation (HILS) for experiment-based optimization. However, since conventional technique of Multi-Objective Evolutionary Algorithms (MOEAs) requires a large number of evaluations, it is difficult to apply it to real world problems of costly evaluation. To make Experiment-Based EMO (EBEMO) using the HILS environment feasible, the most important pre-requisite is reduction of the number of necessary fitness evaluations. In the EBEMO, the performance of the evaluation reduction under uncertainty such as observation noise is highly important, although the previous works often assume noise-free environments. In this paper, we propose an evaluation reduction to overcome the above-mentioned problem by selecting the solution candidates by means of the estimated fitness before applying them to the real experiment in MOEAs. This technique is called ‘Pre-selection’. For the estimation of fitness, we adopt locally weighted regression. The effectiveness of the proposed method was examined by some numerical experiments and also two-objective four-variable optimization problem of a real internal-combustion engine using HILS.
