Hey folks! I’m thrilled to share with you a research paper that I’ve been working on, diving deep into the fascinating realm of vision-based navigation for indoor robots.
In my research paper titled “Real-time Vision-based Navigation for a Robot in an Indoor Environment”, I explore an alternative approach to traditional robot navigation. Instead of relying on more complex sensors, I propose utilizing a simple monocular camera as the robot’s primary visual input.
My method uses computer vision to develop a basic understanding of the robot’s surroundings. By analyzing visual data from a monocular camera, I segment the image to identify walkable surfaces and create a cost map. Transforming the cost map into a bird’s-eye view perspective builds a search-space for the environment, which enables the robot to plan an optimal path while avoiding obstacles.
To provide you with a visual demonstration of my approach, I’ve prepared a 5-minute video showcasing the capabilities of my vision-based navigation system. It’s exciting to see the robustness of the planned path by the robot in the presence of dynamic obstacles like the roomba (shown in the video below).
