Simultaneous localization and mapping (SLAM) is becoming one of the most attractive research focuses of robot control and visual processing. In this paper, robot performs a SLAM mission in an unknown andstructured in...
详细信息
ISBN:
(纸本)9783319168418;9783319168401
Simultaneous localization and mapping (SLAM) is becoming one of the most attractive research focuses of robot control and visual processing. In this paper, robot performs a SLAM mission in an unknown andstructured indoor environment with the Microsoft XBOX Kinect obtaining visual anddepth information. The line features of the object as mapping elements are extracted from visual images, and according to the extraction result, the distance between line features of the object and the robot can be obtained to portray the object edge onto the map in 2d, and the distance is provided by depth data from the Kinect. Meanwhile the robot motion model is created for the trajectory plan. A method using the extended Kalman filter (EKF) is applied to provide a pose estimate for the robot motion trajectory. The SLAM strategy is demonstrated in simulation and experimental environment.
暂无评论