Inertial Measurement Unit Sensor Research Articles

On-Device Semi-Supervised Activity Detection: A New Privacy-Aware Personalized Health Monitoring Approach

This paper presents an on-device semi-supervised human activity detection system that can learn and predict human activity patterns in real time. The clinical objective is to monitor and detect the unhealthy sedentary lifestyle of a user. The proposed semi-supervised learning (SSL) framework uses sparsely labelled user activity events acquired from Inertial Measurement Unit sensors installed as wearable devices. The proposed cluster-based learning model in this approach is trained with data from the same target user, thus preserving data privacy while providing personalized activity detection services. Two different cluster labelling strategies, namely, population-based and distance-based strategies, are employed to achieve the desired classification performance. The proposed system is shown to be highly accurate and computationally efficient for different algorithmic parameters, which is relevant in the context of limited computing resources on typical wearable devices. Extensive experimentation and simulation study have been conducted on multi-user human activity data from the public domain in order to analyze the trade-off between classification accuracy and computation complexity of the proposed learning paradigm with different algorithmic hyper-parameters. With 4.17 h of training time for 8000 activity episodes, the proposed SSL approach consumes at most 20 KB of CPU memory space, while providing a maximum accuracy of 90% and 100% classification rates.

Using artificial intelligence algorithms to approximate data from inertial measurement unit sensors and strain gauges in basketball players

Using artificial intelligence algorithms to approximate data from inertial measurement unit sensors and strain gauges in basketball players

An Inertial-Based Wearable System for Monitoring Vital Signs during Sleep.

This study explores the feasibility of a wearable system to monitor vital signs during sleep. The system incorporates five inertial measurement units (IMUs) located on the waist, the arms, and the legs. To evaluate the performance of a novel framework, twenty-three participants underwent a sleep study, and vital signs, including respiratory rate (RR) and heart rate (HR), were monitored via polysomnography (PSG). The dataset comprises individuals with varying severity of sleep-disordered breathing (SDB). Using a single IMU sensor positioned at the waist, strong correlations of more than 0.95 with the PSG-derived vital signs were obtained. Low inter-participant mean absolute errors of about 0.66 breaths/min and 1.32 beats/min were achieved, for RR and HR, respectively. The percentage of data available for analysis, representing the time coverage, was 98.3% for RR estimation and 78.3% for HR estimation. Nevertheless, the fusion of data from IMUs positioned at the arms and legs enhanced the inter-participant time coverage of HR estimation by over 15%. These findings imply that the proposed methodology can be used for vital sign monitoring during sleep, paving the way for a comprehensive understanding of sleep quality in individuals with SDB.

Cognitive driven gait freezing phase detection and classification for neuro-rehabilitated patients using machine learning algorithms

BackgroundThe significance of diagnosing illnesses associated with brain cognitive and gait freezing phase patterns has led to a recent surge in interest in the study of gait for mental disorders. A more precise and effective way to characterize and classify many common gait problems, such as foot and brain pulse disorders, can improve prognosis evaluation and treatment options for Parkinson patients. Nonetheless, the primary clinical technique for assessing gait abnormalities at the moment is visual inspection, which depends on the subjectivity of the observer and can be inaccurate. Research questionThis study investigates whether it is possible to differentiate between gait brain disorder and the typical walking pattern using machine learning driven supervised learning techniques and data obtained from inertial measurement unit sensors for brain, hip and leg rehabilitation. MethodThe proposed method makes use of the Daphnet freezing of Gait Data Set, consisted of 237 instances with 9 attributes. The method utilizes machine learning and feature reduction approaches in leg and hip gait recognition. ResultsFrom the obtained results, it is concluded that among all classifiers RF achieved highest accuracy as 98.9 % and Perceptron achieved lowest i.e. 70.4 % accuracy. While utilizing LDA as feature reduction approach, KNN, RF and NB also achieved promising accuracy and F1-score in comparison with SVM and LR classifiers. SignificanceIn order to distinguish between the different gait disorders associated with brain tissues freezing/non-freezing and normal walking gait patterns, it is shown that the integration of different machine learning algorithms offers a viable and prospective solution. This research implies the need for an impartial approach to support clinical judgment.

Robot navigation on inclined terrain using social force model

<p><span lang="EN-US">This research introduces an innovative approach to address the limitations of the commonly used social force model-based robot navigation method on flat terrain when applied to sloped terrain. The incline of the terrain becomes a crucial factor in calculating the robot’s steering output when navigating from the initial position to the target position while avoiding obstacles. Therefore, we propose a social forced model-based robot navigation system that can adapt to inclined terrain using inertial measurement unit sensor assistance. The system can detect the surface incline in real time and dynamically adjust friction and gravitational forces, ensuring the robot’s speed and heading direction are maintained. Simulation results conducted using CoppeliaSim show a significant improvement in speed adjustment efficiency. With this new navigation system, the robot can reach its destination in 59.935089 seconds, compared to the conventional social forced model which takes 63.506442 seconds, the robot is also able to reduce slip to reduce wasted movement. This method shows the potential of implementing a faster and more efficient navigation system in the context of inclined terrain.</span></p>

Real-Time Estimation of Tree Position, Tree Height, and Tree Diameter at Breast Height Point, Using Smartphones Based on Monocular SLAM

Precisely estimating the position, diameter at breast height (DBH), and height of trees is essential in forest resource inventory. Augmented reality (AR)-based devices help overcome the issue of inconsistent global point cloud data under thick forest canopies with insufficient Global Navigation Satellite System (GNSS) coverage. Although monocular simultaneous localization and mapping (SLAM) is one of the current mainstream systems, there is still no monocular SLAM solution for forest resource inventories, particularly for the precise measurement of inclined trees. We developed a forest plot survey system based on monocular SLAM that utilizes array cameras and Inertial Measurement Unit (IMU) sensors provided by smartphones, combined with augmented reality technology, to achieve a real-time estimation of the position, DBH, and height of trees within forest plots. Our results from the tested plots showed that the tree position estimation is unbiased, with an RMSE of 0.12 m and 0.11 m in the x-axis and y-axis directions, respectively; the DBH estimation bias is −0.17 cm (−0.65%), with an RMSE of 0.83 cm (3.59%), while the height estimation bias is −0.1 m (−0.95%), with an RMSE of 0.99 m (5.38%). This study will be useful in designing an algorithm to estimate the DBH and position of inclined trees using point clouds constrained by sectional planes at the breast height of the trunk, developing an algorithm to estimate the height of inclined trees utilizing the relationship between rays and plane positions, and providing observers with visual measurement results using augmented reality technology, allowing them to judge the accuracy of the estimates intuitively. Clearly, this system has significant potential applications in forest resource management and ecological research.

Development of an ICT Laparoscopy System with Motion-Tracking Technology for Solo Laparoscopic Surgery: A Feasibility Study

The increasing demand for laparoscopic surgery due to its cosmetic benefits and rapid post-surgery recovery is juxtaposed with a shortage of surgical support staff. This juxtaposition highlights the necessity for improved camera management in laparoscopic procedures, encompassing positioning, zooming, and focusing. Our feasibility study introduces the information and communications technology (ICT) laparoscopy system designed to aid solo laparoscopic surgery. This system tracks a surgeon’s body motion using a controller, manipulating an embedded camera to focus on specific surgical areas. It comprises a camera module, a camera movement controller, and a motor within the main body, operating connected wires according to controller commands for camera movement. Surgeon movements are detected by an inertial measurement unit (IMU) sensor, facilitating precise camera control. Additional features include a foot pedal switch for motion tracking, a dedicated trocar for main body stability, and a display module. The system’s effectiveness was evaluated using an abdomen phantom model and animal experimentation with a porcine model. The camera responded to human movement within 100 ms, a delay that does not significantly affect procedural performance. The ICT laparoscopy system with advanced motion-tracking technology is a promising tool for solo laparoscopic surgery, potentially improving surgical outcomes and overcoming staff shortages.

Autonomous Full 3D Coverage Using an Aerial Vehicle, Performing Localization, Path Planning, and Navigation Towards Indoors Inventorying for the Logistics Domain

Over the last years, a rapid evolution of unmanned aerial vehicle (UAV) usage in various applications has been observed. Their use in indoor environments requires a precise perception of the surrounding area, immediate response to its changes, and, consequently, a robust position estimation. This paper provides an implementation of navigation algorithms for solving the problem of fast, reliable, and low-cost inventorying in the logistics industry. The drone localization is achieved with a particle filter algorithm that uses an array of distance sensors and an inertial measurement unit (IMU) sensor. Navigation is based on a proportional–integral–derivative (PID) position controller that ensures an obstacle-free path within the known 3D map. As for the full 3D coverage, an extraction of the targets and then their final succession towards optimal coverage is performed. Finally, a series of experiments are carried out to examine the robustness of the positioning system using different motion patterns and velocities. At the same time, various ways of traversing the environment are examined by using different configurations of the sensor that is used to perform the area coverage.

Head Gesture Recognition Combining Activity Detection and Dynamic Time Warping.

The recognition of head movements plays an important role in human-computer interface domains. The data collected with image sensors or inertial measurement unit (IMU) sensors are often used for identifying these types of actions. Compared with image processing methods, a recognition system using an IMU sensor has obvious advantages in terms of complexity, processing speed, and cost. In this paper, an IMU sensor is used to collect head movement data on the legs of glasses, and a new approach for recognizing head movements is proposed by combining activity detection and dynamic time warping (DTW). The activity detection of the time series of head movements is essentially based on the different characteristics exhibited by actions and noises. The DTW method estimates the warp path distances between the time series of the actions and the templates by warping under the time axis. Then, the types of head movements are determined by the minimum of these distances. The results show that a 100% accuracy was achieved in the task of classifying six types of head movements. This method provides a new option for head gesture recognition in current human-computer interfaces.

Dual-Modality Cross-Interaction-Based Hybrid Full-Frame Video Stabilization

This study aims to generate visually useful imagery by preventing cropping while maintaining resolution and minimizing the degradation of stability and distortion to enhance the stability of a video for Augmented Reality applications. The focus is placed on conducting research that balances maintaining execution speed with performance improvements. By processing Inertial Measurement Unit (IMU) sensor data using the Versatile Quaternion-based Filter algorithm and optical flow, our research first applies motion compensation to frames of input video. To address cropping, PCA-flow-based video stabilization is then performed. Furthermore, to mitigate distortion occurring during the full-frame video creation process, neural rendering is applied, resulting in the output of stabilized frames. The anticipated effect of using an IMU sensor is the production of full-frame videos that maintain visual quality while increasing the stability of a video. Our technique contributes to correcting video shakes and has the advantage of generating visually useful imagery at low cost. Thus, we propose a novel hybrid full-frame video stabilization algorithm that produces full-frame videos after motion compensation with an IMU sensor. Evaluating our method against three metrics, the Stability score, Distortion value, and Cropping ratio, results indicated that stabilization was more effectively achieved with robustness to flow inaccuracy when effectively using an IMU sensor. In particular, among the evaluation outcomes, within the “Turn” category, our method exhibited an 18% enhancement in the Stability score and a 3% improvement in the Distortion value compared to the average results of previously proposed full-frame video stabilization-based methods, including PCA flow, neural rendering, and DIFRINT.

Effects of IMU sensor-to-segment calibration on clinical 3D elbow joint angles estimation.

Introduction: Inertial Measurement Units (IMU) require a sensor-to-segment calibration procedure in order to compute anatomically accurate joint angles and, thereby, be employed in healthcare and rehabilitation. Research literature proposes several algorithms to address this issue. However, determining an optimal calibration procedure is challenging due to the large number of variables that affect elbow joint angle accuracy, including 3D joint axis, movement performed, complex anatomy, and notable skin artefacts. Therefore, this paper aims to compare three types of calibration techniques against an optical motion capture reference system during several movement tasks to provide recommendations on the most suitable calibration for the elbow joint. Methods: Thirteen healthy subjects were instrumented with IMU sensors and optical marker clusters. Each participant performed a series of static poses and movements to calibrate the instruments and, subsequently, performed single-plane and multi-joint tasks. The metrics used to evaluate joint angle accuracy are Range of Motion (ROM) error, Root Mean Squared Error (RMSE), and offset. We performed a three-way RM ANOVA to evaluate the effect of joint axis and movement task on three calibration techniques: N-Pose (NP), Functional Calibration (FC) and Manual Alignment (MA). Results: Despite small effect sizes in ROM Error, NP displayed the least precision among calibrations due to interquartile ranges as large as 24.6°. RMSE showed significant differences among calibrations and a large effect size where MA performed best (RMSE = 6.3°) and was comparable with FC (RMSE = 7.2°). Offset showed a large effect size in the calibration*axes interaction where FC and MA performed similarly. Conclusion: Therefore, we recommend MA as the preferred calibration method for the elbow joint due to its simplicity and ease of use. Alternatively, FC can be a valid option when the wearer is unable to hold a predetermined posture.

Research on autonomous navigation system of greenhouse electric crawler tractor based on LiDAR.

The application of autonomous navigation technology of electric crawler tractors is an important link in the development of intelligent greenhouses. Aiming at the characteristics of enclosed and narrow space and uneven ground potholes in greenhouse planting, to improve the intelligence level of greenhouse electric crawler tractors, this paper develops a navigation system of electric crawler tractors for the greenhouse planting environment based on LiDAR technology. The navigation hardware system consists of five modules: the information perception module, the control module, the communication module, the motion module, and the power module. The software system is composed of three layers: the application layer, the data processing layer, and the execution layer. The developed navigation system uses LiDAR, Inertial Measurement Unit (IMU) and wheel speed sensor to sense the greenhouse environment and the crawler tractor's information, employs the Gmapping algorithm to build the greenhouse environment map, and utilizes the adaptive Monte Carlo positioning algorithm for positioning. The simulation test of different global path planning algorithms in Matlab shows that the A* algorithm obtains the optimal overall global path. In the scene of map 5, the path planned by the A* algorithm is the most significant, and the number of inflection points is reduced by 40.00% and 87.50%, respectively; meanwhile, the path length is the same as that of the Dijkstra algorithm, but the runtime is reduced by 68.87% and 81.49%, respectively; compared with the RRT algorithm, the path length is reduced by 7.27%. Therefore, the A* algorithm and the Dynamic Window Approach (DWA) method are used for tractor navigation and obstacle avoidance, which ensures global path optimality while also achieving effective local path planning for obstacle avoidance. The test results suggest that the maximum lateral deviation of the built map is 6cm, and the maximum longitudinal deviation is 16cm, which meets the requirement of map accuracy. Additionally, the results of the navigation accuracy test indicate that the maximum lateral deviation of navigation is less than 13cm, the average lateral deviation is less than 7cm, and the standard lateral deviation is less than 8cm. The maximum heading deviation is less than 14°, the average heading deviation is less than 7°, and the standard deviation is less than 8°. These results show that the developed navigation system meets the navigation accuracy requirements of electric crawler tractors in the greenhouse environment.

Quantifying lumbar sagittal plane kinematics using a wrist-worn inertial measurement unit.

Wearable sensors like inertial measurement units (IMUs), and those available as smartphone or smartwatch applications, are increasingly used to quantify lumbar mobility. Currently, wearable sensors have to be placed on the back to measure lumbar mobility, meaning it cannot be used in unsupervised environments. This study aims to compare lumbar sagittal plane angles quantified from a wrist-worn against that of a lumbar-worn sensor. Twenty healthy participants were recruited. An IMU was placed on the right wrist and the L3 spinal level. Participants had to position their right forearm on their abdomen, parallel to the floor. Three sets of three consecutive repetitions of flexion, and extension were formed. Linear mixed models were performed to quantify the effect of region (lumbar vs. wrist) on six outcomes [minimum, maximum, range of motion (ROM) of flexion and extension]. Only flexion ROM was significantly different between the wrist and lumbar sensors, with a mean of 4.54° (95% CI = 1.82°-7.27°). Across all outcomes, the maximal difference between a wrist-worn and lumbar-worn sensor was <8°. A wrist-worn IMU sensor could be used to measure gross lumbar sagittal plane mobility in place of a lumbar-worn IMU. This may be useful for remote monitoring during rehabilitation.

Effects of postural control by personal mobility on human joint movements-prototype of a new alert system for personal mobility devices

Purpose: This study aimed to develop a prototype Personal Mobility (PM) system with human-robot synchronous motion measurement and sensory feedback as an attention reminder to ensure the safe performance of sustained, step-by-step rehabilitation tailored to the daily living conditions of elderly individuals. Methods: Five healthy adults were fitted with a simulation tool to experience the elderly, and inertial measurement unit sensors were placed on the subjects and the PM device to measure joint movements during two postural movements (sitting to supine and supine to standing). In addition, vibration stimulation and voice guidance were implemented as alerts at a certain set threshold of the joint motion angle. We analyzed the deviation of each joint motion between the subject and the PM device regarding the angle measurements of the trunk, hip, knee, and ankle joints. Results: Trunk angle misalignment was high in the sitting-to-supine position. Hip angle misalignment was greater in the supine to standing position. As an alert system, vibration and voice guidance could be presented to the subject with a 0.3-second delay after the set threshold was reached. Conclusion: The misalignment between a human and a motorized wheelchair in contact with the human is more than several tens of degrees when expressed as the angle change value of the joint motion. We believe that a system that can constantly sense and alert the user regarding whether the user and the PM device are in sync is necessary for PM devices being used in rehabilitation or as daily life support.

Deep Learning Approach for Gait Detection for Precise Stimulation of FES to Correct Foot Drop

Automatic detection of foot lift is one of the most important events of Functional Electrical Stimulation (FES). The FES system is used for the correction of Foot Drop (FD). FD is a condition where a person is unable to lift their foot from the ground due to complications that may arise after a stroke or spinal cord injury. It is crucial to accurately detect the patient’s foot lift event when correcting FD through FES as the pulse should only be applied when the person lifts their foot. The FES system applies the electrical pulse based on the input of the foot-lift detection sensor. A conventional FES system employs a sensor that is affixed on the heel to detect the lifting of the foot, but the connecting cables make the patient uncomfortable. To address this problem, IMU (Inertial Measurement Unit)-based sensors have been used, but they have some disadvantages, such as false triggering, low accuracy, and calibration. In this paper, we have presented an algorithm for detecting foot-lift events with high accuracy using a single IMU sensor through the application of deep learning techniques. We have recorded data from 10 healthy people and 10 foot drop patients. We have implemented Artificial Neural Network (ANN), K-Nearest Neighbour (KNN), and Convolutional Neural Network (CNN) models on these data and compared the results of these three models. The proposed algorithm aims to improve the precision of stimulation in the FES system.

Estimation of STEADI performance using inertial measurement unit

IntroductionFalls present a significant public health concern in the United States as a primary cause of unintentional injury-related deaths among older adults. A fall risk assessment toolkit STEADI developed by the CDC has been shown to predict future falls. However, STEADI has issues with accurate evaluations due to the disagreement on cut-off scores in functional assessments and history-taking questionnaires. Wearable sensor technology offers a practical and quantifiable alternative for assessing an individual's movement performance in real-world environments. The use of Inertial Measurement Units (IMUs) offers considerable potential to enhance fall risk screening. PurposeThe primary aim of this study is to test the agreement of STEADI functional assessment performance measured by the IMUs in comparison to human-based measurements. Method27 participants (Age: 74.37 ± 7.21) performed STEADI, including the Four-Stage Balance Test (4SBT), Timed Up & Go Test (TUG), 30-second Chair Stand (30sCS) with IMU placed at the fifth lumbar vertebra which is the proxy location of whole-body Center of Mass. By adopting an equivalent test, the STEADI agreement was tested between the human rater and IMU measurements, giving α = 0.05. ResultBetween the results from evaluators and IMU, the difference in TUG is -0.23 s, and the difference in 30sCS is 0.37, which is equivalent to within 4 % and 8 % for TUG and 30sCS, respectively. The difference in single-leg stance during the 4SBT is 0.59 s; however, the calculated equivalence zone is larger (22.7 %). ConclusionThis study demonstrates the feasibility of using IMU sensors to enhance fall risk screening protocols based on the STEADI. Future refinement may still need to enable broader application and effective screening practices on a larger scale of the population.

Classifying the walking pattern of humans on different surfaces using convolutional features and shallow machine learning classifiers

This study presents a methodology that combines convolution features with shallow classifiers for classifying the walking pattern on different surfaces. At first, convolution features are extracted from six different inertial measurement units (IMU) sensors mounted on the human body. The shallow classifiers namely quadratic SVM, wide neural network, fine KNN, and linear discriminant analysis are trained using convolution features that successfully pass through the global pooling layer of the CNN model. The proposed methodology is also evaluated on the features extracted from both individual IMU sensor and combined IMU sensors. It is observed that proposed methodology performs better for convolution features extracted from all IMU sensors combined together. The proposed methodology is also compared to the CNN model to demonstrate its efficacy. The obtained results show that the shallow classifier (quadratic SVM) achieves the maximum accuracy of 98.2%, whereas the CNN model achieves the accuracy of 90.73% only. Thus, the proposed methodology can be successfully utilised for classifying different walking surfaces based on the gait cycle data. The proposed approach can benefit in automatic gait adjustment of prosthetic foot based on the walking surfaces.

Machine Fault Diagnosis through Vibration Analysis: Time Series Conversion to Grayscale and RGB Images for Recognition via Convolutional Neural Networks

Accurate and timely fault detection is crucial for ensuring the smooth operation and longevity of rotating machinery. This study explores the effectiveness of image-based approaches for machine fault diagnosis using data from a 6DOF IMU (Inertial Measurement Unit) sensor. Three novel methods are proposed. The IMU6DoF-Time2GrayscaleGrid-CNN method converts the time series sensor data into a single grayscale image, leveraging the efficiency of a grayscale representation and the power of convolutional neural networks (CNNs) for feature extraction. The IMU6DoF-Time2RGBbyType-CNN method utilizes RGB images. The IMU6DoF-Time2RGBbyAxis-CNN method employs an RGB image where each channel corresponds to a specific axis (X, Y, Z) of the sensor data. This axis-aligned representation potentially allows the CNN to learn the relationships between movements along different axes. The performance of all three methods is evaluated through extensive training and testing on a dataset containing various operational states (idle, normal, fault). All methods achieve high accuracy in classifying these states. While the grayscale method offers the fastest training convergence, the RGB-based methods might provide additional insights. The interpretability of the models is also explored using Grad-CAM visualizations. This research demonstrates the potential of image-based approaches with CNNs for robust and interpretable machine fault diagnosis using sensor data.

Action Recognition of Taekwondo Unit Actions Using Action Images Constructed with Time-Warped Motion Profiles.

Taekwondo has evolved from a traditional martial art into an official Olympic sport. This study introduces a novel action recognition model tailored for Taekwondo unit actions, utilizing joint-motion data acquired via wearable inertial measurement unit (IMU) sensors. The utilization of IMU sensor-measured motion data facilitates the capture of the intricate and rapid movements characteristic of Taekwondo techniques. The model, underpinned by a conventional convolutional neural network (CNN)-based image classification framework, synthesizes action images to represent individual Taekwondo unit actions. These action images are generated by mapping joint-motion profiles onto the RGB color space, thus encapsulating the motion dynamics of a single unit action within a solitary image. To further refine the representation of rapid movements within these images, a time-warping technique was applied, adjusting motion profiles in relation to the velocity of the action. The effectiveness of the proposed model was assessed using a dataset compiled from 40 Taekwondo experts, yielding remarkable outcomes: an accuracy of 0.998, a precision of 0.983, a recall of 0.982, and an F1 score of 0.982. These results underscore this time-warping technique's contribution to enhancing feature representation, as well as the proposed method's scalability and effectiveness in recognizing Taekwondo unit actions.

Dynamic Stability, Symmetry, and Smoothness of Gait in People with Neurological Health Conditions.

Neurological disorders such as stroke, Parkinson's disease (PD), and severe traumatic brain injury (sTBI) are leading global causes of disability and mortality. This study aimed to assess the ability to walk of patients with sTBI, stroke, and PD, identifying the differences in dynamic postural stability, symmetry, and smoothness during various dynamic motor tasks. Sixty people with neurological disorders and 20 healthy participants were recruited. Inertial measurement unit (IMU) sensors were employed to measure spatiotemporal parameters and gait quality indices during different motor tasks. The Mini-BESTest, Berg Balance Scale, and Dynamic Gait Index Scoring were also used to evaluate balance and gait. People with stroke exhibited the most compromised biomechanical patterns, with lower walking speed, increased stride duration, and decreased stride frequency. They also showed higher upper body instability and greater variability in gait stability indices, as well as less gait symmetry and smoothness. PD and sTBI patients displayed significantly different temporal parameters and differences in stability parameters only at the pelvis level and in the smoothness index during both linear and curved paths. This study provides a biomechanical characterization of dynamic stability, symmetry, and smoothness in people with stroke, sTBI, and PD using an IMU-based ecological assessment.