1
|
Liu M, Wang Z, Jiang P, Yan G. Temperature Compensation Method for Piezoresistive Pressure Sensors Based on Gated Recurrent Unit. SENSORS (BASEL, SWITZERLAND) 2024; 24:5394. [PMID: 39205088 PMCID: PMC11359372 DOI: 10.3390/s24165394] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Figures] [Subscribe] [Scholar Register] [Received: 08/02/2024] [Revised: 08/19/2024] [Accepted: 08/19/2024] [Indexed: 09/04/2024]
Abstract
Piezoresistive pressure sensors have broad applications but often face accuracy challenges due to temperature-induced drift. Traditional compensation methods based on discrete data, such as polynomial interpolation, support vector machine (SVM), and artificial neural network (ANN), overlook the thermal hysteresis, resulting in lower accuracy. Considering the sequence-dependent nature of temperature drift, we propose the RF-IWOA-GRU temperature compensation model. Random forest (RF) is used to interpolate missing values in continuous data. A combination of gated recurrent unit (GRU) networks and an improved whale optimization algorithm (IWOA) is employed for temperature compensation. This model leverages the memory capability of GRU and the optimization efficiency of the IWOA to enhance the accuracy and stability of the pressure sensors. To validate the compensation method, experiments were designed under continuous variations in temperature and actual pressure. The experimental results show that the compensation capability of the proposed RF-IWOA-GRU model significantly outperforms that of traditional methods. After compensation, the standard deviation of pressure decreased from 10.18 kPa to 1.14 kPa, and the mean absolute error and root mean squared error were reduced by 75.10% and 76.15%, respectively.
Collapse
Affiliation(s)
- Mian Liu
- School of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University, Shanghai 200240, China; (M.L.); (P.J.)
| | - Zhiwu Wang
- School of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University, Shanghai 200240, China; (M.L.); (P.J.)
| | - Pingping Jiang
- School of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University, Shanghai 200240, China; (M.L.); (P.J.)
| | - Guozheng Yan
- School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai 200240, China;
| |
Collapse
|
2
|
Weber C, Eggert M, Udelhoven T. Flight Attitude Estimation with Radar for Remote Sensing Applications. SENSORS (BASEL, SWITZERLAND) 2024; 24:4905. [PMID: 39123952 PMCID: PMC11314695 DOI: 10.3390/s24154905] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Subscribe] [Scholar Register] [Received: 07/11/2024] [Revised: 07/21/2024] [Accepted: 07/22/2024] [Indexed: 08/12/2024]
Abstract
Unmanned aerial vehicles (UAVs) and radar technology have benefitted from breakthroughs in recent decades. Both technologies have found applications independently of each other, but together, they also unlock new possibilities, especially for remote sensing applications. One of the key factors for a remote sensing system is the estimation of the flight attitude. Despite the advancements, accurate attitude estimation remains a significant challenge, particularly due to the limitations of a conventional Inertial Measurement Unit (IMU). Because these sensors may suffer from issues such as drifting, additional effort is required to obtain a stable attitude. Against that background, this study introduces a novel methodology for making an attitude estimation using radar data. Herein, we present a drone measurement system and detail its calculation process. We also demonstrate our results using three flight scenarios and outline the limitations of the approach. The results show that the roll and pitch angles can be calculated using the radar data, and we conclude that the findings of this research will help to improve the flight attitude estimation of remote sensing flights with a radar sensor.
Collapse
Affiliation(s)
- Christoph Weber
- Engineering, Computer Science and Economics, TH Bingen University of Applied Sciences, 55411 Bingen am Rhein, Germany
| | - Marius Eggert
- Faculty of Design Computer Science Media, RheinMain University of Applied Sciences, 65197 Wiesbaden, Germany;
| | - Thomas Udelhoven
- Environmental Remote Sensing & Geoinformatics Department, University of Trier, 54286 Trier, Germany;
| |
Collapse
|
3
|
Özateş ME, Yaman A, Salami F, Campos S, Wolf SI, Schneider U. Identification and interpretation of gait analysis features and foot conditions by explainable AI. Sci Rep 2024; 14:5998. [PMID: 38472287 PMCID: PMC10933258 DOI: 10.1038/s41598-024-56656-4] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/19/2023] [Accepted: 03/08/2024] [Indexed: 03/14/2024] Open
Abstract
Clinical gait analysis is a crucial step for identifying foot disorders and planning surgery. Automating this process is essential for efficiently assessing the substantial amount of gait data. In this study, we explored the potential of state-of-the-art machine learning (ML) and explainable artificial intelligence (XAI) algorithms to automate all various steps involved in gait analysis for six specific foot conditions. To address the complexity of gait data, we manually created new features, followed by recursive feature elimination using Support Vector Machines (SVM) and Random Forests (RF) to eliminate low-variance features. SVM, RF, K-nearest Neighbor (KNN), and Logistic Regression (LREGR) were compared for classification, with a Majority Voting (MV) model combining trained models. KNN and MV achieved mean balanced accuracy, recall, precision, and F1 score of 0.87. All models were interpreted using Local Interpretable Model-agnostic Explanation (LIME) method and the five most relevant features were identified for each foot condition. High success scores indicate a strong relationship between selected features and foot conditions, potentially indicating clinical relevance. The proposed ML pipeline, adaptable for other foot conditions, showcases its potential in aiding experts in foot condition identification and planning surgeries.
Collapse
Affiliation(s)
| | - Alper Yaman
- Fraunhofer IPA, Nobelstrasse 12, Stuttgart, Germany.
| | - Firooz Salami
- Clinic for Orthopedics, Heidelberg University Hospital, Schlierbacher Landstrasse 200a, 69118, Heidelberg, Germany
| | - Sarah Campos
- Clinic for Orthopedics, Heidelberg University Hospital, Schlierbacher Landstrasse 200a, 69118, Heidelberg, Germany
| | - Sebastian I Wolf
- Clinic for Orthopedics, Heidelberg University Hospital, Schlierbacher Landstrasse 200a, 69118, Heidelberg, Germany
| | | |
Collapse
|
4
|
Brotchie J, Li W, Greentree AD, Kealy A. RIOT: Recursive Inertial Odometry Transformer for Localisation from Low-Cost IMU Measurements. SENSORS (BASEL, SWITZERLAND) 2023; 23:3217. [PMID: 36991926 PMCID: PMC10057007 DOI: 10.3390/s23063217] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Subscribe] [Scholar Register] [Received: 02/24/2023] [Revised: 03/10/2023] [Accepted: 03/14/2023] [Indexed: 06/19/2023]
Abstract
Inertial localisation is an important technique as it enables ego-motion estimation in conditions where external observers are unavailable. However, low-cost inertial sensors are inherently corrupted by bias and noise, which lead to unbound errors, making straight integration for position intractable. Traditional mathematical approaches are reliant on prior system knowledge, geometric theories and are constrained by predefined dynamics. Recent advances in deep learning, which benefit from ever-increasing volumes of data and computational power, allow for data-driven solutions that offer more comprehensive understanding. Existing deep inertial odometry solutions rely on estimating the latent states, such as velocity, or are dependent on fixed-sensor positions and periodic motion patterns. In this work, we propose taking the traditional state estimation recursive methodology and applying it in the deep learning domain. Our approach, which incorporates the true position priors in the training process, is trained on inertial measurements and ground truth displacement data, allowing recursion and learning both motion characteristics and systemic error bias and drift. We present two end-to-end frameworks for pose invariant deep inertial odometry that utilises self-attention to capture both spatial features and long-range dependencies in inertial data. We evaluate our approaches against a custom 2-layer Gated Recurrent Unit, trained in the same manner on the same data, and tested each approach on a number of different users, devices and activities. Each network had a sequence length weighted relative trajectory error mean ≤0.4594 m, highlighting the effectiveness of our learning process used in the development of the models.
Collapse
Affiliation(s)
- James Brotchie
- School of Science, RMIT University, Melbourne, VIC 3001, Australia
| | - Wenchao Li
- School of Science, RMIT University, Melbourne, VIC 3001, Australia
| | - Andrew D. Greentree
- ARC Centre of Excellence for Nanoscale BioPhotonics, School of Science, RMIT University, Melbourne, VIC 3001, Australia
| | - Allison Kealy
- School of Science, RMIT University, Melbourne, VIC 3001, Australia
- Victorian Department of Environment, Land, Water and Planning, Melbourne, VIC 3000, Australia
| |
Collapse
|
5
|
Shan G, Li G, Wang Y, Xing C, Zheng Y, Yang Y. Application and Prospect of Artificial Intelligence Methods in Signal Integrity Prediction and Optimization of Microsystems. MICROMACHINES 2023; 14:344. [PMID: 36838043 PMCID: PMC9958958 DOI: 10.3390/mi14020344] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Subscribe] [Scholar Register] [Received: 12/30/2022] [Revised: 01/26/2023] [Accepted: 01/26/2023] [Indexed: 06/18/2023]
Abstract
Microsystems are widely used in 5G, the Internet of Things, smart electronic devices and other fields, and signal integrity (SI) determines their performance. Establishing accurate and fast predictive models and intelligent optimization models for SI in microsystems is extremely essential. Recently, neural networks (NNs) and heuristic optimization algorithms have been widely used to predict the SI performance of microsystems. This paper systematically summarizes the neural network methods applied in the prediction of microsystem SI performance, including artificial neural network (ANN), deep neural network (DNN), recurrent neural network (RNN), convolutional neural network (CNN), etc., as well as intelligent algorithms applied in the optimization of microsystem SI, including genetic algorithm (GA), differential evolution (DE), deep partition tree Bayesian optimization (DPTBO), two stage Bayesian optimization (TSBO), etc., and compares and discusses the characteristics and application fields of the current applied methods. The future development prospects are also predicted. Finally, the article is summarized.
Collapse
Affiliation(s)
- Guangbao Shan
- School of Microelectronics, Xidian University, Xi’an 710071, China
| | - Guoliang Li
- School of Microelectronics, Xidian University, Xi’an 710071, China
| | - Yuxuan Wang
- School of Microelectronics, Xidian University, Xi’an 710071, China
| | - Chaoyang Xing
- Beijing Institute of Aerospace Control Devices, Beijing 100039, China
| | - Yanwen Zheng
- School of Microelectronics, Xidian University, Xi’an 710071, China
| | - Yintang Yang
- School of Microelectronics, Xidian University, Xi’an 710071, China
| |
Collapse
|
6
|
Liu Y, Cui J, Liang W. A hybrid learning-based stochastic noise eliminating method with attention-Conv-LSTM network for low-cost MEMS gyroscope. Front Neurorobot 2022; 16:993936. [PMID: 36590082 PMCID: PMC9797827 DOI: 10.3389/fnbot.2022.993936] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/14/2022] [Accepted: 11/22/2022] [Indexed: 12/23/2022] Open
Abstract
Low-cost inertial measurement units (IMUs) based on microelectromechanical system (MEMS) have been widely used in self-localization for autonomous robots due to their small size and low power consumption. However, the low-cost MEMS IMUs often suffer from complex, non-linear, time-varying noise and errors. In order to improve the low-cost MEMS IMU gyroscope performance, a data-driven denoising method is proposed in this paper to reduce stochastic errors. Specifically, an attention-based learning architecture of convolutional neural network (CNN) and long short-term memory (LSTM) is employed to extract the local features and learn the temporal correlation from the MEMS IMU gyroscope raw signals. The attention mechanism is appropriately designed to distinguish the importance of the features at different times by automatically assigning different weights. Numerical real field, datasets and ablation experiments are performed to evaluate the effectiveness of the proposed algorithm. Compared to the raw gyroscope data, the experimental results demonstrate that the average errors of bias instability and angle random walk are reduced by 57.1 and 66.7%.
Collapse
Affiliation(s)
- Yaohua Liu
- School of Nano-Tech and Nano-Bionics, University of Science and Technology of China, Hefei, China,Institute of Nano-Tech and Nano-Bionics, Chinese Academy of Sciences, Suzhou, China,Department of Mathematics and Theories, Peng Cheng Laboratory, Shenzhen, China
| | - Jinqiang Cui
- Department of Mathematics and Theories, Peng Cheng Laboratory, Shenzhen, China,*Correspondence: Jinqiang Cui
| | - Wei Liang
- School of Nano-Tech and Nano-Bionics, University of Science and Technology of China, Hefei, China,Institute of Nano-Tech and Nano-Bionics, Chinese Academy of Sciences, Suzhou, China
| |
Collapse
|
7
|
Drift compensation of a holonomic mobile robot using recurrent neural networks. INTEL SERV ROBOT 2022. [DOI: 10.1007/s11370-022-00430-w] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/17/2022]
|
8
|
Hybrid neural networks for noise reductions of integrated navigation complexes. ARTIF INTELL 2022. [DOI: 10.15407/jai2022.01.288] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/17/2022]
Abstract
The necessity of integrated navigation complexes (INC) construction is substantiated. It is proposed to include in the complex the following inertial systems: inertial, satellite and visual. It helps to increase the accuracy of determining the coordinates of unmanned aerial vehicles. It is shown that in unfavorable cases, namely the suppression of external noise of the satellite navigation system, an increase in the errors of the inertial navigation system (INS), including through the use of accelerometers and gyroscopes manufactured using MEMS technology, the presence of bad weather conditions, which complicates the work of the visual navigation system. In order to ensure the operation of the navigation complex, it is necessary to ensure the suppression of interference (noise). To improve the accuracy of the INS, which is part of the INC, it is proposed to use the procedure for extracting noise from the raw signal of the INS, its prediction using neural networks and its suppression. To solve this problem, two approaches are proposed, the first of which is based on the use of a multi-row GMDH algorithm and single-layer networks with sigm_piecewise neurons, and the second is on the use of hybrid recurrent neural networks, when neural networks were used, which included long-term and short-term memory (LSTM) and Gated Recurrent Units (GRU). Various types of noise, that are inherent in video images in visual navigation systems are considered: Gaussian noise, salt and pepper noise, Poisson noise, fractional noise, blind noise. Particular attention is paid to blind noise. To improve the accuracy of the visual navigation system, it is proposed to use hybrid convolutional neural networks.
Collapse
|
9
|
Ru X, Gu N, Shang H, Zhang H. MEMS Inertial Sensor Calibration Technology: Current Status and Future Trends. MICROMACHINES 2022; 13:879. [PMID: 35744491 PMCID: PMC9228165 DOI: 10.3390/mi13060879] [Citation(s) in RCA: 11] [Impact Index Per Article: 5.5] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 03/09/2022] [Revised: 05/18/2022] [Accepted: 05/29/2022] [Indexed: 12/10/2022]
Abstract
A review of various calibration techniques of MEMS inertial sensors is presented in this paper. MEMS inertial sensors are subject to various sources of error, so it is essential to correct these errors through calibration techniques to improve the accuracy and reliability of these sensors. In this paper, we first briefly describe the main characteristics of MEMS inertial sensors and then discuss some common error sources and the establishment of error models. A systematic review of calibration methods for inertial sensors, including gyroscopes and accelerometers, is conducted. We summarize the calibration schemes into two general categories: autonomous and nonautonomous calibration. A comprehensive overview of the latest progress made in MEMS inertial sensor calibration technology is presented, and the current state of the art and development prospects of MEMS inertial sensor calibration are analyzed with the aim of providing a reference for the future development of calibration technology.
Collapse
Affiliation(s)
| | | | | | - Heng Zhang
- School of Computer and Information Science, Southwest University, Chongqing 400700, China; (X.R.); (N.G.); (H.S.)
| |
Collapse
|
10
|
Short-Term Streamflow Forecasting Using Hybrid Deep Learning Model Based on Grey Wolf Algorithm for Hydrological Time Series. SUSTAINABILITY 2022. [DOI: 10.3390/su14063352] [Citation(s) in RCA: 5] [Impact Index Per Article: 2.5] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 12/10/2022]
Abstract
The effects of developing technology and rapid population growth on the environment have been expanding gradually. Particularly, the growth in water consumption has revealed the necessity of water management. In this sense, accurate flow estimation is important to water management. Therefore, in this study, a grey wolf algorithm (GWO)-based gated recurrent unit (GRU) hybrid model is proposed for streamflow forecasting. In the study, daily flow data of Üçtepe and Tuzla flow observation stations located in various water collection areas of the Seyhan basin were utilized. In the test and training analysis of the models, the first 75% of the data were used for training, and the remaining 25% for testing. The accuracy and success of the hybrid model were compared via the comparison model and linear regression, one of the most basic models of artificial neural networks. The estimation results of the models were analyzed using different statistical indexes. Better results were obtained for the GWO-GRU hybrid model compared to the benchmark models in all statistical metrics except SD at the Üçtepe station and the whole Tuzla station. At Üçtepe, the FMS, despite the RMSE and MAE of the hybrid model being 82.93 and 85.93 m3/s, was 124.57 m3/s, and it was 184.06 m3/s in the single GRU model. We achieved around 34% and 53% improvements, respectively. Additionally, the R2 values for Tuzla FMS were 0.9827 and 0.9558 from GWO-GRU and linear regression, respectively. It was observed that the hybrid GWO-GRU model could be used successfully in forecasting studies.
Collapse
|
11
|
Adaptive Navigation Algorithm with Deep Learning for Autonomous Underwater Vehicle. SENSORS 2021; 21:s21196406. [PMID: 34640726 PMCID: PMC8512330 DOI: 10.3390/s21196406] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 07/21/2021] [Revised: 09/22/2021] [Accepted: 09/23/2021] [Indexed: 11/16/2022]
Abstract
Precise navigation is essential for autonomous underwater vehicles (AUVs). The measurement deviation of the navigation sensors, especially the microelectromechanical systems (MEMS) sensors, is a crucial factor that affects the localization accuracy. Deep learning is a novel method to solve this problem. However, the calculation cycle and robustness of the deep learning method may be insufficient in practical application. This paper proposes an adaptive navigation algorithm with deep learning to address these questions and realize accurate navigation. Firstly, this algorithm uses deep learning to generate low-frequency position information to correct the error accumulation of the navigation system. Secondly, the χ2 rule is selected to judge if the Doppler velocity log (DVL) measurement fails, which could avoid interference from DVL outliers. Thirdly, the adaptive filter, based on the variational Bayesian (VB) method, is employed to estimate the navigation information simultaneous with the measurement covariance, improving navigation accuracy even more. The experimental results, based on AUV field data, show that the proposed algorithm could realize robust navigation performance and significantly improve position accuracy.
Collapse
|