| The research of mobile robot involves a lot of knowledge such as automatic control, computer science, pattern recognition and artificial intelligence. With the fast development of new technology and the wide use of advanced sensors, mobile robot has become the focus in the field of robotics and automation.Intelligent robots are a kind of robots that are able to work in complex environments with the capacity of self-reaction. The aim is to move purposely and do the job without aids. Simultaneous localization and mapping (SLAM) is a hot issue concerned in researching on such kind of robots. SLAM is an essential capability for mobile robots in unknown environments.According to working environments, it can be classified into indoor robot and outdoor robot. The thesis aims to study simultaneous localization and mapping in indoor environment. SLAM techniques and algorithms are firstly reviewed. The existing problems of SLAM are analyzed and the different implementations are described. In the past 20 years, SLAM based on sonar and other sensors has been mature. But it is far from our demand for characteristics of unknown enviroments. So based on vision SLAM has been proposed. Then the focus of this thesis is introduced: SLAM based on Extended Kalman Filter. This approach has been programmed and successfully tested in both simulations and experiments.In addition to motion-model, sensor-model, the camera calibration and the binocular vision based on image matching is described in details. Last, the method based on vision SLAM has been achieved through experiments. |
PDF Full Text Request