Appearance-invariant Visual Localization for Long-term Navigation

Zerong Su1; Xubin Lin1; Li He2; Zhihao Xu1; Xuefeng Zhou1

10.1109/EECR60807.2024.10607284

2024-03-31

979-8-3503-7001-0

2024 10th International Conference on Electrical Engineering, Control and Robotics (EECR)

29-31 March 2024

Guangzhou, China

Long-term navigation refers to the ability of a mobile robot to navigate effectively over extended periods in environments that may undergo changes. Unlike short-term navigation, which focuses on immediate and local motion estimation, long-term navigation aims to identify environmental changes and maintain a sustainable map that is beneficial to the association between current observations and the visited location. In this paper, a Convolutional Neural Network (CNN) based discriminating model is proposed to quantify the repeatability and the discriminativity of visual features, aiming to filter out the unstable keypoints for the mapping process. Moreover, a novel self-adaptive mechanism for map maintenance is proposed to incrementally update the keypoint dataset in a memory-saving manner. Extensive experimental evaluations are conducted on the CMU Seasons dataset, which suggests that our method increases the success rate of localization from 50% to 86% on average.

其他

EI

1.Guangdong Key Laboratory of Modern Control Technology, Institute of Intelligent Manufacturing, Guangdong Academy of Sciences, Guangzhou, China
2.Department of Electronic and Electrical Engineering, Southern University of Science and Technology, Shenzhen, China

Zerong Su,Xubin Lin,Li He,et al. Appearance-invariant Visual Localization for Long-term Navigation[C],2024.