ABSTRACT
Human intelligence with uncertainty reflects human adaptability to the uncertain environment. People have been utilizing three classical experiments to simulate this kind of uncertain intelligence: chess playing between human and computer, the soccer match between robots, and the inverted pendulum. During the past 50 years, artificial intelligence (AI) has made great progress in the symbolic theorem proving and logic reasoning, resulting in defeating Kasparov by the computer “Deep Blue.” The robot soccer match also reflects the progress in uncertain processing related to computer vision, planning, cooperation, and control. In the control of an inverted pendulum, which is a classical research topic in automatic control, the balance and robustness are regarded as the effective way to evaluate intelligent control. This is another important content in the research of AI with uncertainty. In this chapter, we plan to utilize the reasoning and control by qualitative knowledge rather than through the precise mathematical model of the plant, and implement stable control by qualitative rule controller reflecting the control experience by human natural language.
