There have been tremendous efforts to understand the biological nature of human grasping, in such a way that it can be learned and copied to prosthesis–robotics and dextrous grasping applications. Several biomimetic methods and techniques have been adopted, hence applied to analytically comprehend ways human performs grasping to duplicate human knowledge. A major topic for further study, is related to decoding the resulting EEG brainwaves during motorizing of fingers and moving parts. To accomplish this, there are a number of phases that are performed, including recording, pre-processing, filtration, and understanding of the waves. However, there are two important phases that have received substantial research attentions. The classification and decoding, of such massive and complex brain waves, as they are two important steps towards understanding patterns during grasping. In this respect, the fundamental objective of this research is to demonstrate how to employ advanced pattern recognition methods, like fuzzy c-mean clustering for understanding resulting EEG brain waves, in such a way to control a prosthesis or robotic hand, while relying sets of detected EEG brainwaves. There are a number of decoding and classification methods and techniques, however we shall look into fuzzy based clustering blended with principle component analysis (PAC) technique to help for the decoding mechanism. EEG brainwaves during a grasping and manipulation have been used for this analysis. This involves, movement of almost five fingers during a grasping defined task. The study has found that, it is not a straight forward task to decode all human fingers motions, as due to the complexity of grasping tasks. However, the adopted analysis was able to classify and identify the different narrowly performed and related fundamental events during a simple grasping task.
Part of the book: Biomimetic Prosthetics
Learning mobile robot space and navigation behavior, are essential requirements for improved navigation, in addition to gain much understanding about the navigation maps. This chapter presents mobile robots feature-based SLAM behavior learning, and navigation in complex spaces. Mobile intelligence has been based on blending a number of functionaries related to navigation, including learning SLAM map main features. To achieve this, the mobile system was built on diverse levels of intelligence, this includes principle component analysis (PCA), neuro-fuzzy (NF) learning system as a classifier, and fuzzy rule based decision system (FRD).
Part of the book: Applications of Mobile Robots