Although many continuous situations seem appropriate for case-based reasoning, there has been little work done in this area. Case-based reasoning in such problem domains requires significant enhancements to the basic case-based reasoning methods used in traditional case-based reasoning systems. In this paper, we describe the stepwise case based reasoning (SCBR) approach to autonomous robot navigation. In the SCBR approach, two different types of cases are distinguished: situation cases and route cases. This approach provides an appropriate architectural framework for autonomous navigation systems in which situation cases are used to support the situation module (reactive subsystem) and route cases are used to support the high-level route planning module. The route planning module generates high-level sketchy plans, based on route cases. The situation module controls the sequence of the primitive actions based on an actual situation case.

are many approaches to the face recognition. This paper presents an approach that combines advantage of generalization ability of Principal Component Analysis (PCA) and specialization ability of Case-based reasoning (CBR). CBR is expected to improve the generalization ability of PCA in the recognition process. By using PCA the new image is projected into its eigenface components. The projected vector represents a description component of a new face case. The CBR module compares the similarity of the new face case and previously stored face cases in the face casebase and retrieves the most similar case. The solution component of the retrieved case represents the results of the recognition process.