अमूर्त

Computer Vision-Based Sensor of Autonomous Mobile Robot for Detection of Stairs

A.Papsys,G.Daunys.

Autonomous navigation of a robot in an unknown environment is still a challenge for engineers. This paper deals with a simple method for detecting obstacles in a robot’s way and stairs in multi-storey buildings. As there may be low light conditions, a structured light-based approach was chosen. A mathematical model was developed to describe the formation of traces of structured light in images. Various scenarios of a robot’s position in relation to stairs were simulated. It was established that light source traces depend on the position and distance of the camera from the light source. Computer simulation revealed characteristic changes of light source patterns in images. It was proved that a cross-form pattern of is attractive for obstacle detection and object recognition.