Mobile Robot Neuro-Fuzzy Navigation Based VSLAM Features Learning

Abstract

The presented approach was focused around building intelligence for mobile robot navigation. That was achieved by creating navigation intelligence capabilities while the robot is in motion. The adopted learning paradigm was a five layers Neuro-Fuzzy (NF) learning architecture, due to ability to create and inference for enhanced navigation. To meet such visual data gathering, the mobile robot platform have fully computer-interfaced stereo vision, and reliable 3D perception. Mobile robot intelligence (NF), hence learns navigation (SLAM) maps visual features, as it travels within spaces. Blinding intelligence with visual maps has resulted in better navigation capabilities.