Wyniki wyszukiwania - BazTech

1

Position fixing and uncertainty

Filipowicz W.

TransNav : International Journal on Marine Navigation and Safety of Sea Transportation

|

2023

|

Vol. 17 No. 4

887--893

EN

Taken random observations are usually accompanied by rectified knowledge regarding their behaviour. In modern computer applications, raw data sets are usually exploited at learning phase. At this stage, available data are explored in order to extract necessary parameters required within the inference scheme computations. Crude data processing enables conditional dependencies extraction. It starts with upgrading histograms and their uncertainty estimation. Exploiting principles of fuzzy systems one can obtain modified step-wise structure in the form of locally injective density functions. They can be perceived as conditional dependency diagrams with identified uncertainty that enables constructing basic probability assignments. Belief, uncertainty and plausibility measures are extracted from initial raw data sets. The paper undertakes problem of belief structures upgraded from uncertainty model in order to solve the position fixing problem. The author intention is presenting position fixing as an inference scheme. The scheme engages evidence, hypothesis and revokes concept of conditional relationships.

2

A logical device for processing nautical data

Filipowicz W.

Zeszyty Naukowe Akademii Morskiej w Szczecinie

|

2017

|

nr 52 (124)

65--73

EN

Nautical measurements are randomly and systematically corrupted. There is a rich scope of knowledge regarding the randomness shown by results of observations. The distribution of stochastic distortions remains an estimate and is imprecise with respect to their parameters. Uncertainties can also occur through the subjective assessment of each piece of available data. The ability to model and process all of the aforementioned items through traditional approaches is rather limited. Moreover, the results of observations, the final outcome of a quality evaluation, can be estimated prior to measurements being taken. This a posteriori analysis is impaired and it is outside the scope of traditional, inaccurate data handling methods. To propose new solutions, one should start with an alternative approach towards modelling doubtfulness. The following article focusses on belief assignments that may benefit from the inclusion of uncertainty. It starts with a basic interval uncertainty model. Then, assignments engaging fuzzy locations around nautical indications are discussed. This fragment includes transformation from density functions to probability distributions of random errors. Diagrams of the obtained conversions are included. The presentation concludes with a short description of a computer application that implements the presented ideas.

3

On the mathematical theory of evidence in navigation

Filipowicz W.

Zeszyty Naukowe Akademii Morskiej w Szczecinie

|

2016

|

nr 45 (117)

159--167

EN

In most problems encountered in navigation, imprecision and uncertainty dominate. Methods of their processing rely on rather obsolete formalisms based on probability and statistics. Available solutions exploit a limited amount of available data, and knowledge is necessary to interpret the achieved results. Profound a posteriori analysis is rather limited; thus, the informative context of solutions is rather poor. Including knowledge in a nautical data processing scheme is impossible. Remaining stuck with the traditional formal apparatus, based on probability theory, one cannot improve the informative context of obtained results. Traditional approaches toward solving problems require assumptions imposed by the probabilistic model that exclude possibility of modelling uncertainty. It should be noticed that the flexibility of exploited formalism decide the quality of upgrading models and, subsequently, on the universality of the final results. Therefore, extension of the available formalisms is a challenge to be met. Many publications devoted to the mathematical theory of evidence (MTE) and its adaptation for nautical science in order to support decision making in navigational processes have enabled one to submit and defend the following proposition. Many practical problems related to navigational ship conducting and to feature uncertainty can be solved with MTE; the informative context of the obtained results is richer when compared to those acquired by traditional methods. Additionally, a posteriori analysis is an inherent feature of the new foundations. The brief characteristics of a series of publications devoted to the new methodology are the main topics of this paper.