A Review of V-SLAM

Visual Simultaneous Localization and Mapping (V-SLAM) uses the camera as a main external sensor and creates environmental map while locating itself in the absence of a prior knowledge of the environment. The quality of map created by SLAM will affect the performance of the autonomous location, path planning and obstacle avoidance. Though the SLAM problem is considered to be a theoretically solved problem, there are still many issues remain in in practical implement. This paper will introduce research history of SLAM briefly and the cameras used in V-SLAM, then classify the SLAM method based on the feature-based and direct method. The existing V-SLAM methods will be summarized. Finally, the key issues of V-SLAM and development trends are discussed.