Collaborative Learning using Fuzzy Logic (CLIFF): Part 1

There are a growing number of aerospace applications demonstrating the effectiveness of emulating human decision making with fuzzy logic. This overall project has three main parts: an investigation of the benefits of type-1 and type-2 fuzzy logic, creation of a fuzzy system that can adapt to a situation using one of the two aforementioned fuzzy logic types, and creation of a fuzzy robotic coach who has the ability to better its team in a previously created fuzzy MATLAB game of PONG based upon its ability to adapt and learn from its opponent. In this effort, part one was attempted, and, through extensive studying and understanding of type-2 fuzzy theory, the beginning of a type-2 fuzzy logic system was developed in the MATLAB environment. This system was compared to a benchmark problem, which raised the question of type-2 logic effectiveness, however, as a created type-1 system showed better results than any results in the paper. Therefore due to type-2 systems being more computationally heavy, and difficult to program, it was concluded that cascading type-1 logic may be more suitable than type-2 logic in control applications. Overall, the developments of intelligence and learning within this project should prove to be an interesting stepping stone towards highly capable robots with the ability to learn from their environment; something that would be extremely helpful in disaster recovery, space exploration and military applications. Nomenclature