A Hybrid Algorithm For Feature-Based Approach In Visual Simultaneous Localization And Mapping

Abstract

Robot navigation means the robot's ability to determine its own position in its frame of reference (map) and then to plan a path towards some goal location by avoiding dangerous situations such as collisions and unsafe conditions. The use of obstacle avoidance for the navigation of the robot has been increasing due to its importance in many sectors such as for autonomous cars, for military surveillance robots, for indoor surveillance robots, and for assisting visual impaired people with robot. Therefore, the ability to observe and understand the meaning of the robot is an essential part of navigating the robot safe and reaching the desired end without any difficulty. Therefore, integrating fast vision algorithm that can enable the robot to understand its environment and avoid obstacle make the robot to navigate safely within an unknown indoor environment. This study enables the robot to detect and recognize a map with high accuracy and enable the robot to avoid any damage. But there are a small robot who only got one camera, those robots need an accurate map and avoid collision this thesis shows solving this problem only using a hybrid algorithm for feature-based approach. For micro-robot which only got one camera to discover the environment so such a future need to be applied. Since navigating the robot through an unfamiliar environment requires an understanding of details and other useful visual information, the robot has to have a method for detecting recognizing the environment as accurately as possible .so our work focus on this behave. 1) It needs to model and understand its environment, with much keyframe points along with its attendant uncertainties, as well as find its own location within the environment, using only the noisy observations from its monocular camera so we propose hybrid algorism for VSLAM based on monocular camera and the algorithms are Minutiae extraction, commonly used in fingerprint recognition and ORB-VSLAM one of the common used VSLAM algorism 2) The proposed algorithm got tested bygone through some identified simulation tools with well-known data and local data then finally got the appreciable results this thesis paper explains this in detail.