The idea for this research work is to use the fuzzy logic as a novel technique for modelling infantry tactics and doctrine that are currently being documented in flow charts. In flow charts we can only have yes-no decisions where one query branches either to another query (if the answer is no) or branches to an action (if the answer is yes). In such methods there are no various degrees of reasoning, just crisp yes or crisp no. This is not the way that humans usually reason. Hence, the introduction of fuzzy logic gives more flexibility in modelling human decision making process. / On the other hand, the knowledge-based systems are designed to mimic the performance of a human expert by transferring his/her expertise in a specific field to a computer-based model most oftenly in the form of a software package. This knowledge is often imprecise, or not all facts are available. Still humans are capable of making good decisions within such uncertain environment. / It was decided to use fuzzy logic for modelling infantry tactics and doctrine because fuzzy logic provides a basis for representing uncertain and imprecise knowledge and forms a basis for human reasoning and decision making in such incompletely defined systems. Therefore the fuzzy logic seems a suitable choice in modelling infantry tactics and doctrine. / In this research work we first show how selected infantry tactics could be modelled utilising MATLAB® Fuzzy Toolbox. Then we develop a software package in Java programming language, which is then used to model the same infantry tactics. Using Java is necessary as the final aim of this project is to implement the developed models in the intelligent agent software (Jack). / The first part of this thesis is an overview of the fuzzy logic as one of the artificial intelligence paradigms. This part also briefly introduces intelligent agents, what they are and what they are not. The second part of this thesis shows the implementation of fuzzy reasoning in modelling selected infantry tactics and doctrine. The simulation results from all applications are also presented. / Thesis MEng(ElectronicEngineering byResearch)--University of South Australia, 2055.
| Identifer | oai:union.ndltd.org:ADTP/267508 |
| Creators | Nedic, Vladimir. |
| Source Sets | Australiasian Digital Theses Program |
| Language | English |
| Detected Language | English |
| Rights | copyright under review |
Page generated in 0.0023 seconds