RSV-SLAM: Toward Real-Time Semantic Visual SLAM in Indoor Dynamic Environments

Simultaneous Localization and Mapping (SLAM) plays an important role in many robotics fields, including social robots. Majority of the existing visual SLAM methods rely on a static world assumption and fail in dynamic environments. In this paper, we proposed a real-time semantic RGB-D SLAM system for dynamic environments that is capable of detecting moving objects and maintaining a static map for robust camera tracking. The proposed model is to eliminate the influence of dynamic objects by introducing deep learning-based semantic information to SLAM systems. Furthermore, we augment the semantic segmentation process using an Extended Kalman filter module to detect dynamic objects that are temporarily idle. We have also implemented a generative network to fill in the missing regions of input images belonging to dynamic objects. This highly modular framework has been implemented on the ROS platform and can achieve around 22 fps on a GTX1080. Benchmarking the developed pipeline on dynamic sequences from the TUM dataset suggests that the proposed approach achieves competitive localization error compared to the state-of-the-art methods while operating in near real-time. The source code is publicly available at https://github.com/mobiiin/rsv_slam .