This article proposes a new kind of static state estimator based on Genetic
Algorithms (GAs) in order to accomplish an estimation-based search techniq
ue for multimodal adjustment processes. To adapt GA to the estimation probl
em, the general procedure of the conventional GAs are modified in two aspec
ts; first, the reproduction strategy uses two different criteria (survival
cost and parent cost) for the elimination of weaker members and parent sele
ction, and secondly, a new genetic operator based on a priori knowledge of
the system is developed to generate more effective offspring solutions. The
performance and applicability of the proposed estimator was investigated v
ia three numerical examples of the multimodal adjustment system, and the re
sults obtained show that this estimator performs very well even with the pr
esence of a large amount of measurement noise. (C) 1998 Elsevier Science Lt
d. All rights reserved.