FILTERING DATA BY USING THREE ERROR THEORIES TOGETHER: THE GUESS FILTER
Olivier Strauss, Eric Aime · 1996
Most papers dealing with noisy measurements combination try and use a single tool providing the best estimation. But, as pointed out by Dubois and Prade (DUB 90), most tools dealing with imperfect information have different purposes. The purpose of this paper is to try and find a way of using three different frameworks together in order to obtain a filter that combines robustness, accuracy, reliability and easy implementation. To perform this estimation, possibility theory is considered to deal with precision while a statistical tool reduces uncertainty. Then, we propose to use the rough sets theory to perform robust and easy computation of the resulting filter. Some results on real data provided by the compass of a submarine robot are presented.