Search Results (2,089)

This paper introduces a model that predicts four common smartphone-holding postures, aiming to enhance user interface adaptability. It is unique in being completely independent of platform and hardware, utilizing the inertial measurement unit (IMU) for real-time posture detection based on sensor data collected around tap gestures. The model identifies whether the user is holding and operating the smartphone with one hand or using both hands in different configurations. For model training and validation, sensor time series data undergo extensive feature extraction, including statistical, frequency, magnitude, and wavelet analyses. These features are incorporated into 74 distinct sets, tested across various machine learning frameworks—k-nearest neighbors (KNN), support vector machine (SVM), and random forest (RF)—and evaluated for their effectiveness using metrics such as cross-validation scores, test accuracy, Kappa statistics, confusion matrices, and ROC curves. The optimized model demonstrates a high degree of accuracy, successfully predicting the holding hand with a 95.7% success rate. This approach highlights the potential of leveraging sensor data to improve mobile user experiences by adapting interfaces to natural user interactions. Full article

The aim of this study was to compare barbell velocities at different intensities and estimated 1-RM with actual 1-RM measured with different measuring tools in bench presses and squats. Fourteen resistance-trained athletes (eight men, six women, age 28.1 ± 7.5 years, body mass 78.1 ± 12.2 kg, body height 1.73 ± 0.09 m) performed bench presses and squats at five loads varying from 45 to 85% of one repetition maximum (1-RM), together with 1-RM testing, while measuring mean, mean propulsive, and peak barbell velocity with six different commercially used inertial measurement units (IMUs) and linear encoder software systems attached to the barbell. The 1-RM was also estimated based upon the load–velocity regression, which was compared with the actual 1-RM in the bench press and squat exercises. The main findings were that GymAware revealed the highest reliability along with minimal bias, while Musclelab and Vmaxpro showed moderate reliability with some variability at higher loads. Speed4lifts and PUSH band indicated greater variability, specifically at higher intensities. Furthermore, in relation to the second aim of the study, significant discrepancies were found between actual and estimated 1-RM values, with Speed4lifts and Musclelab notably underestimating 1-RM. These findings underscore the importance of selecting reliable tools for accurate velocity-based training and load prescription. Full article

The purpose of this paper is to describe ongoing research on appropriate instrumentation and analysis techniques to characterize postural stability, postural agility, and dynamic stability, which collectively comprise the postural control spectrum. This study had a specific focus on using emerging sensors to develop protocols suitable for use outside laboratory or clinical settings. First, we examined the optimal number and placement of wearable accelerometers for assessing postural stability. Next, we proposed metrics and protocols for assessing postural agility with the use of a custom force plate-controlled video game. Finally, we proposed a method to quantify dynamic stability during walking tasks using novel frequency-domain metrics extracted from acceleration data obtained with a single body-worn IMU. In each of the three studies, a surrogate for instability was introduced, and the sensors and metrics discussed in this paper show promise for differentiating these trials from stable condition trials. Next steps for this work include expanding the tested population size and refining the methods to even more reliably and unobtrusively characterize postural control status in a variety of scenarios. Full article

The purpose of this study is to develop an optimized system for predicting Knee Adduction Moment (KAM) using wearable Inertial Measurement Unit (IMU) sensors and Long Short-Term Memory (LSTM) RNN. Traditional KAM measurement methods are limited by the need for complex laboratory equipment and significant time and cost investments. This study proposes two systems for predicting Knee Adduction Moment based on wearable IMU sensor data and gait patterns: the Multi-model Gait-based KAM Prediction System and the Single-model KAM Prediction System. The Multi-model system pre-classifies different gait patterns and uses specific prediction models tailored for each pattern, while the Single-model system handles all gait patterns with one unified model. Both systems were evaluated using IMU sensor data and GRF data collected from participants in a controlled laboratory environment. The overall performance of the Multi-model Gait-based KAM Prediction System showed an approximately 20% improvement over the Single-model KAM Prediction System. Specifically, the RMSE for the Multi-model system was 6.84 $\mathrm{N}·\mathrm{m}$, which is lower than the 8.82 $\mathrm{N}·\mathrm{m}$ of the Single-model system, indicating a better predictive accuracy. The Multi-model system also achieved a MAPE of 8.47%, compared with 12.95% for the Single-model system, further demonstrating its superior performance. Full article

Composite robots often encounter difficulties due to changes in illumination, external disturbances, reflective surface effects, and cumulative errors. These challenges significantly hinder their capabilities in environmental perception and the accuracy and reliability of pose estimation. We propose a nonlinear optimization approach to overcome these issues to develop an integrated localization and navigation framework, IIVL-LM (IMU, Infrared, Vision, and LiDAR Fusion for Localization and Mapping). This framework achieves tightly coupled integration at the data level using inputs from an IMU (Inertial Measurement Unit), an infrared camera, an RGB (Red, Green and Blue) camera, and LiDAR. We propose a real-time luminance calculation model and verify its conversion accuracy. Additionally, we designed a fast approximation method for the nonlinear weighted fusion of features from infrared and RGB frames based on luminance values. Finally, we optimize the VIO (Visual-Inertial Odometry) module in the R3LIVE++ (Robust, Real-time, Radiance Reconstruction with LiDAR-Inertial-Visual state Estimation) framework based on the infrared camera’s capability to acquire depth information. In a controlled study, using a simulated indoor rescue scenario dataset, the IIVL-LM system demonstrated significant performance enhancements in challenging luminance conditions, particularly in low-light environments. Specifically, the average RMSE ATE (Root Mean Square Error of absolute trajectory Error) improved by 23% to 39%, with reductions from 0.006 to 0.013. At the same time, we conducted comparative experiments using the publicly available TUM-VI (Technical University of Munich Visual-Inertial Dataset) without the infrared image input. It was found that no leading results were achieved, which verifies the importance of infrared image fusion. By maintaining the active engagement of at least three sensors at all times, the IIVL-LM system significantly boosts its robustness in both unknown and expansive environments while ensuring high precision. This enhancement is particularly critical for applications in complex environments, such as indoor rescue operations. Full article

Snake robots require autonomous localization and mapping capabilities for field applications. However, the characteristics of their motion, such as large turning angles and fast rotation speeds, can lead to issues like drift or even failure in positioning and map building. In response to this situation, this paper starts from the gait motion characteristics of the snake robot itself, proposing an improved gait motion method and a tightly coupled method based on IMU and visual information to solve the problem of poor algorithm convergence caused by head-shaking in snake robot SLAM. Firstly, the adaptability of several typical gaits of the snake robot to SLAM methods was evaluated. Secondly, the serpentine gait was selected as the object of gait improvement, and a head stability control method for the snake robot was proposed, thereby reducing the interference of the snake robot’s motion on the sensors. Thirdly, a visual–inertial tightly coupled SLAM method for the snake robot’s serpentine gait and Arc-Rolling gait was proposed, and the method was verified to enhance the robustness of the visual SLAM algorithm and improve the positioning and mapping accuracy of the snake robot. Finally, experiments proved that the methods proposed in this paper can effectively improve the accuracy of positioning and map building for snake robots. Full article

Due to increasing traffic congestion, travel modeling has gained importance in the development of transportion mode detection (TMD) strategies over the past decade. Nowadays, recent smartphones, equipped with integrated inertial measurement units (IMUs) and embedded algorithms, can play a crucial role in such development. In particular, obtaining much more information on the transportation modes used by users through smartphones is very challenging due to the variety of the data (accelerometers, magnetometers, gyroscopes, proximity sensors, etc.), the standardization issue of datasets and the pertinence of learning methods for that purpose. Reviewing the latest progress on TMD systems is important to inform readers about recent datasets used in detection, best practices for classification issues and the remaining challenges that still impact the detection performances. Existing TMD review papers until now offer overviews of applications and algorithms without tackling the specific issues faced with real-world data collection and classification. Compared to these works, the proposed review provides some novelties such as an in-depth analysis of the current state-of-the-art techniques in TMD systems, relying on recent references and focusing particularly on the major existing problems, and an evaluation of existing methodologies for detecting travel modes using smartphone IMUs (including dataset structures, sensor data types, feature extraction, etc.). This review paper can help researchers to focus their efforts on the main problems and challenges identified. Full article

Background and Purpose: This pilot randomized controlled trial evaluated the effects of 12 sessions of patient-specific adaptive dynamic cycling (PSADC) versus non-adaptive cycling (NA) on motor function and mobility in individuals with Parkinson’s disease (PD), using inertial measurement unit (IMU) sensors for objective assessment. Methods: Twenty-three participants with PD (13 in the PSADC group and 10 in the NA group) completed the study over a 4-week period. Motor function was measured using the Kinesia™ sensors and the MDS-UPDRS Motor III, while mobility was assessed with the TUG test using OPAL IMU sensors. Results: The PSADC group showed significant improvements in MDS-UPDRS Motor III scores (t = 5.165, p < 0.001) and dopamine-sensitive symptoms (t = 4.629, p = 0.001), whereas the NA group did not improve. Both groups showed non-significant improvements in TUG time. IMU sensors provided continuous, quantitative, and unbiased measurements of motor function and mobility, offering a more precise and objective tracking of improvements over time. Conclusions: PSADC demonstrated enhanced treatment effects on PD motor function compared to NA while also reducing variability in individual responses. The integration of IMU sensors was essential for precise monitoring, supporting the potential of a data-driven, individualized exercise approach to optimize treatment outcomes for individuals with PD. Full article

As mining technology advances, intelligent robots in open-pit mining require precise localization and digital maps. Nonetheless, significant pitch variations, uneven highways, and rocky surfaces with minimal texture present substantial challenges to the precision of feature extraction and positioning in traditional visual SLAM systems, owing to the intricate terrain features of open-pit mines. This study proposes an improved SLAM technique that integrates visual and Inertial Measurement Unit (IMU) data to address these challenges. The method incorporates a point–line feature fusion matching strategy to enhance the quality and stability of line feature extraction. It integrates an enhanced Line Segment Detection (LSD) algorithm with short segment culling and approximate line merging techniques. The combination of IMU pre-integration and visual feature restrictions is executed inside a tightly coupled visual–inertial framework utilizing a sliding window approach for back-end optimization, enhancing system robustness and precision. Experimental results demonstrate that the suggested method improves RMSE accuracy by 36.62% and 26.88% on the MH and VR sequences of the EuRoC dataset, respectively, compared to ORB-SLAM3. The improved SLAM system significantly reduces trajectory drift in the simulated open-pit mining tests, improving localization accuracy by 40.62% and 61.32%. The results indicate that the proposed method demonstrates significance. Full article

Without a GNSS (global navigation satellite system) signal, the integrated navigation system in vehicles with a fixed trajectory (e.g., railcars) is limited to the use of micro-electromechanical system-inertial navigation system (MEMS-INS) and odometer (ODO). Due to the significant measurement error of the MEMS inertial device and the inability of ODO to output attitude, the positioning error is generally large. To address this problem, this paper presents a new integrated navigation algorithm based on a dynamically constrained Kalman model. By analyzing the dynamics of a railcar, several new observations have been investigated, including errors of up and lateral velocity, centripetal acceleration, centripetal D-value (difference value), and an up-gyro bias. The state transition matrix and observation matrix for the error state model are represented. To improve navigation accuracy, virtual noise technology is applied to correct errors of up and lateral velocity. The vehicle-running experiment conducted within 240 s demonstrates that the positioning error rate of the dead-reckoning method based on MEMS-INS is 83.5%, whereas the proposed method exhibits a rate of 4.9%. Therefore, the accuracy of positioning can be significantly enhanced. Full article

To suppress pedestrian positioning drift, a velocity constraint commonly known as zero-velocity update (ZUPT) is widely used. However, it cannot correct the error in the non-zero velocity interval (non-ZVI) or observe heading errors. In addition, the positioning accuracy will be further affected when a velocity error occurs in the ZVI (e.g., foot tremble). In this study, the foot, ankle, and shank were regarded as a triangular structure. Consequently, an angle constraint was established by utilizing the sum of the internal angles. Moreover, in contrast to the traditional ZUPT algorithm, a velocity constraint method combined with Coriolis theorem was constructed. Magnetometer measurements were used to correct heading. Three groups of experiments with different trajectories were carried out. The ZUPT method of the single inertial measurement unit (IMU) and the distance constraint method of dual IMUs were employed for comparisons. The experimental results showed that the proposed method had high accuracy in positioning. Furthermore, the constraints built by the lower limb structure were applied to the whole gait cycle (ZVI and non-ZVI). Full article

The rise of mobile communication, low-power chips, and the Internet of Things has made smartwatches increasingly popular. Equipped with inertial measurement units (IMUs), these devices can recognize user activities through artificial intelligence (AI) analysis of sensor data. However, most existing AI-based activity recognition algorithms require significant computational power and storage, making them unsuitable for low-power devices like smartwatches. Additionally, discrepancies between training data and real-world data often hinder model generalization and performance. To address these challenges, we propose LIMUNet and its smaller variant LIMUNet-Tiny—lightweight neural networks designed for human activity recognition on smartwatches. LIMUNet utilizes depthwise separable convolutions and residual blocks to reduce computational complexity and parameter count. It also incorporates a dual attention mechanism specifically tailored to smartwatch sensor data, improving feature extraction without sacrificing efficiency. Experiments on the PAMAP2 and LIMU datasets show that the LIMUNet improves recognition accuracy by 2.9% over leading lightweight models while reducing parameters by 88.3% and computational load by 58.4%. Compared to other state-of-the-art models, LIMUNet achieves a 9.6% increase in accuracy, with a 60% reduction in parameters and a 57.8% reduction in computational cost. LIMUNet-Tiny further reduces parameters by 75% and computational load by 80%, making it even more suitable for resource-constrained devices. Full article

Quadruped robots possess significant mobility in complex and uneven terrains due to their outstanding stability and flexibility, making them highly suitable in rescue missions, environmental monitoring, and smart agriculture. With the increasing use of quadruped robots in more demanding scenarios, ensuring accurate and stable state estimation in complex environments has become particularly important. Existing state estimation algorithms relying on multi-sensor fusion, such as those using IMU, LiDAR, and visual data, often face challenges on non-stationary terrains due to issues like foot-end slippage or unstable contact, leading to significant state drift. To tackle this problem, this paper introduces a state estimation algorithm that integrates an invariant extended Kalman filter (InEKF) with a disturbance observer, aiming to estimate the motion state of quadruped robots on non-stationary terrains. Firstly, foot-end slippage is modeled as a deviation in body velocity and explicitly included in the state equations, allowing for a more precise representation of how slippage affects the state. Secondly, the state update process integrates both foot-end velocity and position observations to improve the overall accuracy and comprehensiveness of the estimation. Lastly, a foot-end contact probability model, coupled with an adaptive covariance adjustment strategy, is employed to dynamically modulate the influence of the observations. These enhancements significantly improve the filter’s robustness and the accuracy of state estimation in non-stationary terrain scenarios. Experiments conducted with the Jueying Mini quadruped robot on various non-stationary terrains show that the enhanced InEKF method offers notable advantages over traditional filters in compensating for foot-end slippage and adapting to different terrains. Full article

Reduced walking endurance is common in people with multiple sclerosis (PwMS), leading to reduced social participation and increased fall risk. This highlights the importance of identifying which gait aspects should be mostly targeted by rehabilitation to maintain/increase walking endurance in this population. A total of 56 PwMS and 24 healthy subjects (HSs) executed the 6 min walk test (6 MWT), a clinical measure of walking endurance, wearing three inertial sensors (IMUs) on their shanks and lower back. Five IMU-based digital metrics descriptive of different gait domains, i.e., double support duration, trunk sway, gait regularity, symmetry, and local dynamic instability, were computed. All metrics demonstrated moderate–high ability to discriminate between HSs and PwMS (AUC: 0.79–0.91) and were able to detect differences between PwMS at minimal (PwMS_mFR) and moderate–high fall risk (PwMS_FR). Compared to PwMS_mFR, PwMS_FR walked with a prolonged double support phase (+100%), larger trunk sway (+23%), lower stride regularity (−32%) and gait symmetry (−18%), and higher local dynamic instability (+24%). Normative cut-off values were provided for all metrics to help clinicians in detecting abnormal scores at an individual level. The five metrics, entered into a multiple linear regression model with 6 MWT distance as the dependent variable, showed that gait regularity and the three metrics most related to dynamic balance (i.e., double support duration, trunk sway, and local dynamic instability) were significant independent contributors to 6 MWT distance, while gait symmetry was not. While double support duration and local dynamic instability were independently associated with walking endurance in both PwMS_mFR and PwMS_FR, gait regularity and trunk sway significantly contributed to 6 MWT distance only in PwMS_mFR and PwMS_FR, respectively. Taken together, the present results allowed us to provide hints for tailored rehabilitation exercises aimed at specifically improving walking endurance in PwMS. Full article

Background/Objectives: This paper proposes a method for managing gait imbalances by integrating the Internet of Things (IoT) and machine learning technologies. Ankle–foot orthosis (AFO) devices are crucial medical braces that align the lower leg, ankle, and foot, offering essential support for individuals with gait imbalances by assisting weak or paralyzed muscles. This research aims to revolutionize medical orthotics through IoT and machine learning, providing a sophisticated solution for managing gait issues and enhancing patient care with personalized, data-driven insights. Methods: The smart ankle–foot orthosis (AFO) is equipped with a surface electromyography (sEMG) sensor to measure muscle activity and an Inertial Measurement Unit (IMU) sensor to monitor gait movements. Data from these sensors are transmitted to the cloud via fog computing for analysis, aiming to identify distinct walking phases, whether normal or aberrant. This involves preprocessing the data and analyzing it using various machine learning methods, such as Random Forest, Decision Tree, Support Vector Machine (SVM), Artificial Neural Network (ANN), Long Short-Term Memory (LSTM), and Transformer models. Results: The Transformer model demonstrates exceptional performance in classifying walking phases based on sensor data, achieving an accuracy of 98.97%. With this preprocessed data, the model can accurately predict and measure improvements in patients’ walking patterns, highlighting its effectiveness in distinguishing between normal and aberrant phases during gait analysis. Conclusions: These predictive capabilities enable tailored recommendations regarding the duration and intensity of ankle–foot orthosis (AFO) usage based on individual recovery needs. The analysis results are sent to the physician’s device for validation and regular monitoring. Upon approval, the comprehensive report is made accessible to the patient, ensuring continuous progress tracking and timely adjustments to the treatment plan. Full article