Evolutionary Recognition of Corridor and Turning using Adaptive Model with 3D Structure

The paper describes a new method for recognizing corridors for autonomous mobile robot. The method detects the edge markers of a corridor based on the genetic-algorithm (GA) and 3D model-based matching. The edge markers with distinct color are generally located at the bottom of the walls for human visual recognition. To recognize the edge markers and the branch of corridor stably and precisely, a set of 3D models of the corridor which change its model length in 3D space is utilized to adapt the variation of the appearance of the markers. The experimental results show the method is effective for real-time recognition of corridor and its branch