Sample capacity for autonomous underwater vehicle range accreditation

摘要

In this paper we are interested in the hypothesis test (HT) solution in the presence of prior information and conjugated prior distribution. Truncated Bayesian sequential posterior odd test (SPOT) is introduced to predict the required test sample size in autonomous underwater vehicle (AUV) range accreditation. We focus on the connection between risks (type I and type II) and truncated threshold, required test sample size. And the computational approach based on numerical approaching is given in steps. Supported by prior information after fusion on credibility, we deeply analyzed the effect of risk base value on SPOT performance, and an array of random bootstrapping data is generated to validate the conclusion. A mature solution to small-size-sample test accreditation is proposed in this paper, which provides vital suggestions for test sample size prediction in small-size-sample parameter hypothesis test.