Keyframe extraction and loop closure detection considering robot motion

Appearance based loop closure detection is a crucial technique for robot simultaneous localization and mapping. Extracting appropriate keyframes can reduce computational cost. In this paper, we propose a keyframe extraction and loop closure detection algorithm considering robot motion information. Both linear and angular velocities of the robot are utilized to decide when to extract the query keyframe. Comparative experiments are carried out on a public dataset to verify the proposed algorithm. The results demonstrate that our algorithm can extract appropriate keyframes at key points on robot trajectory and can be used in loop closure detection.