Predicting accelerometer baseline correction and non-divergent deformation velocity based on convolutional neural network (CNN) during GNSS Downgrade

Jing, Ce, Bertolesi, Elisa ORCID: https://orcid.org/0000-0003-3258-0743, Huang, Guanwen, Li, Xin, Zhang, Qin, Zhai, Weiwei, Liu, Guolin and Li, Hang 2025. Predicting accelerometer baseline correction and non-divergent deformation velocity based on convolutional neural network (CNN) during GNSS Downgrade. IEEE Sensors Journal 10.1109/JSEN.2025.3543726

Preview

PDF - Accepted Post-Print Version Download (12MB) | Preview

Abstract

Accelerometer and Global Navigation Satellite System (GNSS) can be effectively combined to establish a robust multi sensor deformation monitoring system. However, GNSS signals may get downgraded in challenging environments, and then destroy the Kalman filter data fusion model. As a result, the accelerometer becomes the only reliable sensor for deformation monitoring, but relying on only accelerometer data may lead to rapid error accumulation due to its potential baseline shift error. To mitigate this challenge, especially in the slow-moving deformation scenarios, we propose a baseline correction prediction algorithm named CNN-BC, based on convolutional neural networks. This algorithm utilizes high-frequency acceleration and baseline correction as input and output features, respectively. The baseline correction of training dataset is derived from the accelerometer and GNSS coupled algorithm. By incorporating the reliable prediction from the network, we can correct the original accelerometer data and reduce error accumulation. To further address the divergence in deformation velocity, we develop CNN-dVel, which uses high-frequency acceleration and velocity difference as input and output features, respectively. We validated the proposed algorithms through two slow deformation experiments utilizing both high-precision and low-cost accelerometers. The results demonstrate that the CNN-BC can predict reliable baseline correction, with an average root mean square (RMS) of 0.37cm/s2, and the CNN-dVel achieves non-divergent deformation velocity prediction, with an average RMS of 0.42 cm/s. Furthermore, optimizing the training dataset with acceleration standard deviation (STD) basis enhances prediction accuracy.

Item Type:	Article
Date Type:	Published Online
Status:	In Press
Schools:	Schools > Engineering
Subjects:	T Technology > TA Engineering (General). Civil engineering (General)
Publisher:	Institute of Electrical and Electronics Engineers
ISSN:	1530-437X
Date of First Compliant Deposit:	4 March 2025
Date of Acceptance:	1 March 2025
Last Modified:	06 Mar 2025 12:15
URI:	https://orca.cardiff.ac.uk/id/eprint/176551

Actions (repository staff only)

Edit Item