Abstract:
Nature's problem solving strategies are very different than our engineering problem solving methodologies. Nature's problem solving strategies produce highly robust solutions to very difficult problems. The main thrust of this seminar is to discuss how we can apply the nature's problem solving strategies to inverse problems in engineering. The basic tools of biologically inspired soft computing methods are: neural network (inspired by the structure of human brain), genetic algorithm (inspired by natural evolution) and fuzzy logic (inspired by the linguistic methods). These tools are being widely used in engineering applications. However, the soft computing tools have potentials far beyond those exploited in the vast majority of the current applications. In order to exploit the full potential of these methods, it is important to understand the fundamental differences between the nature's problem solving strategies and our mathematically based methods. These issues will be discussed through a series of examples from the research conducted by my colleagues and I, and our graduate students.