A POSITION ESTIMATION SYSTEM FOR MOBILE ROBOTS USING A MONOCULAR IMAGE OF A 3-D LANDMARK

This paper presents an absolute position estimation system for a mobil e robot moving on a flat surface. In this system, a 3-D landmark with four coplanar points and a non-coplanar point is utilized to improve t he accuracy of position estimation and to guide the robot during navig ation. Applying theoretical analysis, we investigate the image sensiti vity of the proposed 3-D landmark compared with the conventional 2-D l andmark. In the camera calibration stage of the experiments, we employ a neural network as a computational tool. The neural network is train ed from a set of learning data collected at various points around the mark so that the extrinsic and intrinsic parameters of the camera syst em can be resolved. The overall estimation algorithm from the mark ide ntification to the position determination is implemented in a 32-bit p ersonal computer with an image digitizer and an arithmetic accelerator . To demonstrate the effectiveness of the proposed 3-D landmark and th e neural network-based calibration scheme, a series of navigation expe riments were performed on a wheeled mobile robot (LCAR) in an indoor e nvironment. The results show the feasibility of the position estimatio n system applicable to mobile robot's real-time navigation.