Research Article
Features Extraction of Reconstruction Model Using in Augmented Reality System of Teleoperation Mobile Robots
@INPROCEEDINGS{10.1007/978-3-031-34497-8_7, author={Zhang Dongpu and Xu Fang and Chen Huimin and Wu Wen and Cui Zhaojing and Guo Weiguo}, title={Features Extraction of Reconstruction Model Using in Augmented Reality System of Teleoperation Mobile Robots}, proceedings={Mobile Wireless Middleware, Operating Systems and Applications. 11th EAI International Conference, MOBILWARE 2022, Virtual Event, December 28-29, 2022, Proceedings}, proceedings_a={MOBILWARE}, year={2023}, month={5}, keywords={augmented reality teleoperation mobile robots reconstruction model SIFT features extraction}, doi={10.1007/978-3-031-34497-8_7} }- Zhang Dongpu
Xu Fang
Chen Huimin
Wu Wen
Cui Zhaojing
Guo Weiguo
Year: 2023
Features Extraction of Reconstruction Model Using in Augmented Reality System of Teleoperation Mobile Robots
MOBILWARE
Springer
DOI: 10.1007/978-3-031-34497-8_7
Abstract
In manned deep space exploration, teleoperation robot can cooperate with or even replace astronauts to perform various dangerous tasks in unfamiliar environment. By combining augmented reality system with teleoperation robot, the higher reliability and accuracy required by teleoperation robot can be obtained. The real-time tracking algorithm is one of the key and difficult points in the realization of on orbit teleoperation augmented reality system. Feature extraction based on reconstruction model is the basis of real-time tracking of the system. This paper studies the feature extraction algorithm based on reconstruction model, analyzes the feature points extraction of Harris and sift from the principle level combined with the task characteristics, studies the relationship among the time, image size and number of feature points of the two kinds of feature extraction through experiments, and tests and analyzes the practical application effect of the selected feature extraction methods in the system, and obtains the SIFT The conclusion is that feature point extraction is more suitable for on orbit augmented reality teleoperation system, and the future work is prospected.
