A number of research workers have applied intelligent approaches for robotic applications. In the recent literature there is an increasing role of fuzzy and Neuro fuzzy approaches for unmanned vehicles. Both these approaches are based on intelligent rules. However for these applications the rules become very large and so computational time is very high. It is important to explore the approaches so as to reduce the computation time. In this paper a combination of factor analysis and clustering approaches is suggested so as to reduce the number of rules. The factor analysis can be used to reduce the number of parameters while clustering approach can be used to reduce the number of observations. Based on this methodology a new algorithm is developed which reduces the original parameters and observations into a set of new data. An algorithm is proposed and applied on a real robotic data available in a previous paper. Some of the applications for future work are proposed.