Optimization of an Inertial Sensor De-Noising Method using a Hybrid Deep Learning Algorithm

Boronakhin, A. M.; Shalymov, Roman V.; Larionov, Daniil Yu.; Yen, Nguyen Trong

doi:10.1109/elconrus54750.2022.9755529

Cited by 1 publication

(1 citation statement)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This method achieved good signal enhancement results on the training set, but the signal enhancement results on the test data were unsatisfactory. Further, Boronakhin et al (Boronakhin et al 2022) used Bayesian optimization to optimize the hyperparameters of the GRU-LSTM model, which improved the model performance on the test set, but due to the lack of interpretability of deep learning models, the generated signal has poor trustworthiness and even loses the semantic information of the input signals (Shamwell et al 2019). Even though the generated signals were better than the input signals in some quantization indicators (such as quantization noise, angle random walk, velocity random walk, and bias instability), they performed worse than the original signals in downstream tasks such as attitude estimation and trajectory reconstruction (Huang et al 2019).…”

Section: Data-driven Methods For Imu Signal Denoisingmentioning

confidence: 98%

Wavelet Dynamic Selection Network for Inertial Sensor Signal Enhancement

Wang,

Zhao

2024

AAAI

View full text Add to dashboard Cite

As attitude and motion sensing components, inertial sensors are widely used in various portable devices, covering consumer electronics, sports health, aerospace, etc. But the severe intrinsic errors of inertial sensors heavily restrain their function implementation, especially the advanced functionality, including motion trajectory recovery and motion semantic recognition, which attracts considerable attention. As a mainstream signal processing method, wavelet is hailed as the mathematical microscope of signal due to the plentiful and diverse wavelet basis functions. However, complicated noise types and application scenarios of inertial sensors make selecting wavelet basis perplexing. To this end, we propose a wavelet dynamic selection network (WDSNet), which intelligently selects the appropriate wavelet basis for variable inertial signals. In addition, existing deep learning architectures excel at extracting features from input data but neglect to learn the characteristics of target categories, which is essential to enhance the category awareness capability, thereby improving the selection of wavelet basis. Therefore, we propose a category representation mechanism (CRM), which enables the network to extract and represent category features without increasing trainable parameters. Furthermore, CRM transforms the common fully connected network into category representations, which provide closer supervision to the feature extractor than the far and trivial one-hot classification labels. We call this process of imposing interpretability on a network and using it to supervise the feature extractor the feature supervision mechanism, and its effectiveness is demonstrated experimentally and theoretically in this paper. The enhanced inertial signal can perform impracticable tasks with regard to the original signal, such as trajectory reconstruction. Both quantitative and visual results show that WDSNet outperforms the existing methods. Remarkably, WDSNet, as a weakly-supervised method, achieves the state-of-the-art performance of all the compared fully-supervised methods.

show abstract

Section: Data-driven Methods For Imu Signal Denoisingmentioning

confidence: 98%