Speaker
Description
This research presents a novel approach utilizing deep learning techniques to enhance the accuracy of indoor localization technology using UWB sensors, with the specific goal of developing an indoor autonomous driving system that can follow or move near humans, such as a human-following robot. Notably, unlike previous UWB devices that rely on CIR data, the latest UWB devices utilize the signal delay between the target and anchor UWB to measure distance with high accuracy, and our proposed localization system is based on this distance data. Additionally, to address the issue of sensor noise and variable position error in non-line-of-sight environments, the proposed system calibrates the bilateration localization method that underlies multilateration techniques. Our experimental results demonstrate that our learning-based localization system outperforms traditional localization systems in reducing errors. It is well-known that deep learning algorithms can face challenges such as overfitting and generalization issues. However, the proposed algorithm in this research addresses these problems through several means, resulting in a more robust and effective deep learning approach for improving indoor localization technology.
References
[1] Zafari F, Gkelias A, Leung KK. “A Survey of Indoor Localization System and Technologies.” IEEE
Comm. Surveys & Tutorials. 2019; 21(3): 2568-2509.
[2] Heungju Ahn, V. C. Dang, H. C. Seo, Sang C. Lee, “Mapping property of bilateration and its application to human-following robot, in”:Fuzzy Systems and Data Mining VI, Vol. 331 of Frontiers 75 in Articial Intelligence and Applications, 2020, pp. 546-552.
[3] Heungju Ahn, V. C. Dang, Sang C Lee, “Bilateration localization system: Complexification of data for the cyber-physical understanding”,preprint
[4] Y. -M. Lu, J. -P. Sheu and Y. -C. Kuo, "Deep Learning for Ultra-Wideband Indoor Positioning," 2021 IEEE 32nd Annual International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC), 2021, pp. 1260-1266.
[5] V. C. Dang, Heungju Ahn, H. C. Seo, Sang C. Lee, “A cognitive robotic system for a human-following robot, in:” Fuzzy Systems and Data Mining VI, Vol. 331 of Frontiers in Articial Intelligence 80 and Applications, 2020, pp. 601-607.
[6] Feng T, Yu Y,Wu L, Bai Y, Xiao Z, Lu Z. A “Human-Tracking Robot Using Ultra Wideband Technology.” IEEE Access. 2018; 6: 42541-42550
| Keywords | Learning-based, IPS, localization, human-following robot, UWB,deep learning |
|---|
