| Summary: | This study mainly deals with the various aspects of modeling the learning processes within the brain limbic system and studying the various aspects of using it for different applications in control engineering. The current study is a multi-aspect research effort which not only requires a background of control engineering, but also a basic knowledge of some biomorphic systems. The main focus of this study is on biological systems which are involved in emotional processes. In mammalians, a part of the brain called the limbic system is mainly responsible for emotional processes. Therefore, general brain emotional processes and specific aspects of the limbic system are reviewed in the early parts of this study. Next, we describe developing a computational model of the limbic system based on these concepts. Since the focus of this study is on the application of the model in engineering systems and not on the biological concepts, the model established is not a very complicated model and does not include all the components of the limbic system. In fact, we are trying to develop a model which captures the minimal and basic properties of the limbic system which are mainly known as the Amygdala-Orbitofrontal Cortex system. The main chapter of this thesis, Chapter IV, shows the utilization of the Brain Emotional Learning (BEL) model in different applications of control and signal fusion systems. The main effort is focused on applying the model to control systems where the model acts as the controller block. Furthermore, the application of the model in signal fusion is also considered where simulation results support the applicability of the model. Finally, we studied different analytical aspects of the model including the behavior of the system during the adaptation phase and the stability of the system. For the first issue, we simplify the model, e.g. remove the nonlinearities, to develop mathematical formulations for behavior of the system. To study the stability of the system, we use the cell-to-cell mapping algorithm which reveals the stability conditions of the system in different representations. This thesis finishes with some concluding remarks and some topics for future research on this field.
|
|---|
