First of all, I apologize if this question has been answered before. We have in mind to work with a laser pointing downwards, so the obstacles lying in the floor in front of the robot could be detected. If I am not wrong, ROS works with a 2D map, more precisely with a 2D projection of a 3D cloud of points. When performing local planning, the new discovered obstacles are added to the map. My question is: how does ROS consider the floor when it is not detected exactly (maybe with some tolerance?) where it is supossed to be? My guess is that it is considered an obstacle and then a laser readings filtering would be necessary and the detection of obstacles of small height not straightforward. Am I right? I would be very grateful for any suggestion from people working in this area. Cheers! Maria -- View this message in context: http://ros-users.122217.n3.nabble.com/Floor-detected-as-an-obstacle-tp2397002p2397002.html Sent from the ROS-Users mailing list archive at Nabble.com.