Ground Plane Detection using Kinect sensor

Ground plane detection is essential for successful navigation of vision based mobile robots. We introduce a novel and robust ground plane detection algorithm using depth information acquired by a Kinect sensor. Unlike similar methods from the literature, we do not assume that the ground plane covers the largest area in the scene. Furthermore our algorithm handles two different conditions: fixed and changing view angle of the sensor. We show that the algorithm is robust if the view angle is fixed whereas an additional procedure handles different view angles satisfactorily.