Towards a hybrid brain based robot

Autonomous robots performing various tasks without human guidance have always been our dream for a better future. A problem that faces autonomous robots is dealing with novel situations. One of the recent technologies that has proven its ability to deal with such situation is Brain-Based Device(BBD). However, the large computational power needed to simulate its nervous system is a major limitation. To overcome this problem, this paper presents a hybrid system that consists of neuronal and non-neuronal areas. In non-neuronal areas, computer vision algorithms are used to extract some features from images. While neuronal-areas are connected together based on a detailed neuroanatomical structure to mimic the human learning process. OpenCV is used for extracting features and obtaining invariant object-recognition. Nengo python package is used for simulating neuronal areas in the system and monitoring activities of neuronal units. Moreover, the successful integration of the subsystems leads to the perceptual categorization based on invariant object-recognition of various visual cues. The time needed for the simulation is significantly decreased compared to the BBDs in the literature.