关键词: 图像分割, 三维目标检测, 拓扑地图, 图匹配, 语义SLAM

Abstract:

Semantic information perception of the external environment and accurate positioning are the keys to autonomous navigation and operation of mobile robots. This paper proposes a method of semantic simultaneous localization and mapping (SLAM) based on a monocular camera. The system completes three-dimensional (3D) object detection while estimating the trajectory. We model the 3D objects with cuboids. Then, the semantic meanings, color distribution, size and neighborhood topology of the objects are extracted as descriptors for the accurate matching of objects between different frames. The camera pose, map points and object landmarks are optimized jointly in the backend of the system. The weight coefficient of each error term in the cost function is autonomously adjusted to improve the estimation accuracy and robustness of each state variable of the system. The experimental results show that the proposed method has high accuracy in map construction.

Key words: image segmentation, 3D object detection, topological map, graph matching, semantic SLAM