Affiliations: [a] School of Faculty of Electrical Engineering and Computer Science, Ningbo University, Ningbo, China
| [b] School of Faculty of Mechanical Engineering & Mechanics, Ningbo University, Ningbo, China
| [c] School of Information Technology, Deakin University, Melbourne, Australia
| [d] School of Computer Science and Engineer, The University of New South Wales, Sydney, Australia
Correspondence:
[*]
Corresponding author. Zhijun Xie, School of Faculty of Electrical Engineering and Computer Science, Ningbo University, Ningbo, China. E-mail: [email protected].
Note: [1] This work was supported by National Natural Science Foundation of China (Grant No. U20A20121); Ningbo public welfare project (Grant No. 202002N3109, 2022S094); Natural Science Foundation of Zhejiang Province (Grant No. LY21F020006); The international cooperation project of Ningbo (Grant No. 2023H012, 2023H007); Science and Technology Innovation 2025 Major Project of Ningbo (Grant No. 2019B10125, 2019B10028, 2020Z016, 2021Z031, 2022Z074, 2022Z241, 2023Z132, 2023Z133, 2023Z216); Ningbo Fenghua District industrial chain key core technology “unveiled the commander” project (Grant No. 202106206).
Abstract: Traditional Simultaneous Localization and Mapping application in dynamic situations is constrained by static assumptions. However, the majority of well-known dynamic SLAM systems use deep learning to identify dynamic objects, which creates the issue of trade-offs between accuracy and real-time. To tackle this issue, this work suggests a unique dynamic semantics method(DYS-SLAM) for semantic simultaneous localization and mapping that strikes a trade-off between high accuracy and high real-time performance. We propose M-LK, an enhanced Lucas-Kanade(LK) optical flow method. This technique keeps the continuous motion and greyscale consistency assumptions from the original method while switching out the spatial consistency assumption for a motion consistency assumption, reducing sensitivity to image gradients to identify dynamic feature points and regions efficiently. In order to increase segmentation accuracy while maintaining real-time performance, we develop a segmentation refinement scheme that projects 3D point cloud segmentation results into 2D object detection zones. A dense semantic octree graph is built in the interim to expedite the high-level process. Compared to the four equivalent dynamic SLAM approaches, experiments on the publicly available TUM RGB-D dataset demonstrate that the DYS-SLAM method offers competitive localization accuracy and satisfactory real-time performance in both high and low-dynamic scenarios.
Keywords: Visual SLAM, object detection, dynamic environment, deep learning for visual perception
