Abstract
This research work proposes a computer-aided algorithm for the conversion
of building plans into a graph for the navigation of robots. Rescuing people from
burning buildings manually is a tedious process. Robots play a pivotal role in industrial
automation and the deployment of robots in rescue operations can save many people.
Indoor navigation for robots is a challenging task since every building has a unique
structure. A routing graph is inevitable to find the path in a building quickly for the
navigation of robots to perform the rescue operation. The automatic extraction of the
routing graph from the image of the floor plan is offered in this research work. The
floor plan images are acquired and converted into a raster image. Then, by using the
predefined kernels, the white pixels are eroded for obtaining the routing path of
common walkways through corridors and rooms. The Extended Conditional Erosion
Algorithm is used for the extraction of the routing graph from the floor plan images.
The resultant graph as output aids the navigation of the robot.