Inertial Measurement Unit Sensor Research Articles

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.

A novel method for accurate division of the gait cycle into seven phases using shank angular velocity

BackgroundAccurate detection of gait events is crucial for gait analysis, enabling the assessment of gait patterns and abnormalities. Inertial measurement unit (IMU) sensors have gained traction for event detection, mainly focusing on initial contact (IC) and toe-off (TO) events. However, effective detection of other key events such as heel rise (HR), feet adjacent (FA), and tibia vertical (TBV) is essential for comprehensive gait analysis. Research questionCan a novel IMU-based method accurately detect HR, TO, FA, and TBV events, and how does its performance compare with existing methods? MethodsWe developed and validated an IMU-based method using cumulative mediolateral shank angular velocity (CSAV) for event detection. A dataset of nearly 25,000 gait cycles from healthy adults walking at varying speeds and footwear conditions was used for validation. The method’s accuracy was assessed against force plate and motion capture data and compared with existing TO detection methods. ResultsThe CSAV method demonstrated high accuracy in detecting TO, FA, and TBV events and moderate accuracy in HR event detection. Comparisons with existing TO detection methods showcased superior performance. The method's stability across speed and shoe variations underscored its robustness. SignificanceThis study introduces a highly accurate IMU-based method for detecting gait events needed to divide the gait cycle into seven phases. The effectiveness of the CSAV method in capturing essential events across different scenarios emphasizes its potential applications. Although HR event detection can be further improved, the precision of the CSAV method in TO, FA, and TBV detection advance the field. This study bridges a critical gap in IMU-based gait event detection by introducing a method for subdividing the swing phase into its subphases. Further research can focus on refining HR detection and expanding the method’s utility across diverse gait contexts, thereby enhancing its clinical and scientific significance.

Human motion activity recognition and pattern analysis using compressed deep neural networks

ABSTRACT This work presents an on-device machine learning model with the ability to identify different mobility gestures called human activity recognition (HAR), which includes running, walking, squatting, jumping, and others. The data is collected through an Arduino Nano 33 BLE Sense board with a sampling rate of 119 Hz, which is embedded with an Inertial Measurement Unit (IMU) sensor. The same board is used as a microcontroller to identify human gestures by developing an end-to-end edge computing application. A deep neural network model is trained and then compressed for deployment on the board to create a self-contained, embedded device capable of identifying the type of gesture performed. Three deep learning models, namely Multi-Layer Perceptron (MLP), Convolutional Neural Network – Long Short Term Memory (CNN-LSTM) & CNN-Gated Recurrent Unit (CNN-GRU), are evaluated in the identification of the mobility gestures. The observed accuracy of the models is 96%, 97.1% and 97.8%, respectively, MLP, CNN-LSTM & CNN-GRU across the different gesture categories. The study shows the utility of embedded devices with deep neural network-based models, which can provide low cost, minimal power usage, and meet data privacy requirements in HAR.

Sensorimotor Control Using Adaptive Neuro-Fuzzy Inference for Human-Like Arm Movement

In this study, a sensorimotor controller is designed to characterize the required muscle force to enable a robotics system to perform a human-like circular movement. When the appropriate muscle internal forces are chosen, the arm end-point tracks the desired path via joint-space feedback. An objective function of the least-change rate of muscle forces is determined to find suitable feedback gains. The parameter defining the muscle force is then treated as a learning parameter through an adaptive neuro-fuzzy inference system, incorporating the rate of change of muscle forces. In experimental section, the arm motion of healthy subjects is captured using the inertial measurement unit sensors, and then the image of the drawn path is processed. The inertial measurement unit sensors detect each segment motion’s orientation using quaternions, and the image is employed to identify the exact end-point position. Experimental data on arm movement are then utilized in the control parameter computation. The proposed brain–motor control mechanism enhances motion performance, resulting in a more human-like movement.

Effectiveness of Inertial Measurement Unit Sensor–Based Feedback Assistance in Telerehabilitation of Patients with Diabetes after Total Knee Arthroplasty: A Randomized Controlled Trial

Effectiveness of Inertial Measurement Unit Sensor–Based Feedback Assistance in Telerehabilitation of Patients with Diabetes after Total Knee Arthroplasty: A Randomized Controlled Trial

Risk of using smartphones while walking for digital natives in realistic environments: Effects of cognitive–motor interference

The effect of using smartphones while walking on the cognitive and physical abilities of the “digital native” generation, i.e., individuals who have grown up in a digital media–centric environment, remains poorly understood. This study evaluated the effects of cognitive–motor interference on the use of smartphones while walking in children and young adults. The study involved 50 individuals from the digital age generation, including 24 children and 26 young adults. The study encompassed three experimental conditions, in which participants were instructed to traverse a distance of 60 m. The initial condition functioned as a control, wherein the participants walked without supplementary stimuli. In the second condition, the participants were provided with explicit instructions to grasp the smartphone device and position it in front of their chest by using both hands. This manipulation introduced a postural component into the experimental setup. The third condition required participants to be ambulatory while concurrently engaging in a cognitive task, namely, participating in a game that necessitated focused attention. Gait parameters were obtained by using inertial measurement unit sensors. Subsequently, the acquired gait characteristics were converted into dual-task costs (DTC). In the cognitive condition, children exhibited significantly greater DTC values for gait speed (76%), stride length (79%), stride time (102%), and stride length coefficient of variation (CV) than the young adults (p < 0.025). Moreover, as shown by the increased CV, a significant association exists between poor performance in smartphone games among children and increased variability in stride length. In children, the DTC of stride time CV decreased as smartphone game scores increased (R2 = 16.5%), and the DTC of stride length CV decreased more markedly as smartphone game scores increased (R2 = 28.2%). In conclusion, children are at a higher risk of pedestrian accidents when using smartphones while walking compared to young adults.

Shape Sensing for Continuum Robotics Using Optoelectronic Sensors with Convex Reflectors

Three-dimensional shape sensing in soft and continuum robotics is a crucial aspect for stable actuation and control in fields such as minimally invasive surgery, engine repairs and search and rescue operations, as the estimation of complex curvatures while using continuum robotic tools is required to manipulate through fragile paths. This challenge has been addressed using a range of different sensing techniques, for example, Fibre Bragg grating (FBG) technology, inertial measurement unit (IMU) sensor networks, or stretch sensors. Previously, an optics-based method using optoelectronic sensors was explored, offering a simple and cost-effective solution for shape sensing in a flexible tendon-actuated manipulator in two orientations. This was based on proximity-modulated angle estimation and has been the basis for the shape sensing method addressed in this paper. The improved and miniaturised technique demonstrated in this paper is based on the use of a convex shaped reflector with optoelectronic sensors integrated into a tendon-actuated robotic manipulator. Upgraded sensing capability is achieved using optimisation of the convex reflector shape in terms of sensor range and resolution, and improved calibration is achieved through the integration of spherical bearings for friction-free motion. Shape estimation is achieved in two orientations upon calibration of sensors, with a maximum Root-Mean-Square Error (RMS) of 3.37°.

Effects of core stability and feedback music on upper body mediolateral movements during cycling

BackgroundAsymmetry in involuntary trunk motion during voluntary movements of the lower extremities is a risk factor for musculoskeletal injuries and may be related to core stability. Core stability plays a pivotal role in maintaining postural stability during distal segment movements. Because mediolateral head motion partially represents trunk motion during rhythmic movements, controlling it can help ensure symmetric trunk motion. This study aimed to investigate the relationship between core stability and asymmetric trunk motion during rhythmic movements, and to evaluate the effects of feedback music on mediolateral head motion.MethodsWe developed a system that uses a wireless earbud and a high-resolution inertial measurement unit sensor to measure head angle and provide feedback music. When the head angle exceeds a predefined threshold, the music is muted in the earbud on the side of the head tilt. In our lab-based study, we measured head angles during cycling at 70% of maximum speed using this self-developed system, and compared them between individuals with good (Sahrmann core stability test: 2–5 level) and poor core stability (0–1 level). The amplitude of mediolateral head motion was represented by the difference between the left and right peak angles, and the symmetry in mediolateral head motion was represented by the average of left and right peak angles.ResultsIndividuals with poor core stability demonstrated significantly greater amplitude of, and less symmetry in, mediolateral head motion than those with good core stability. Additionally, feedback music significantly reduced the amplitude of mediolateral head motion in both the good- and poor-core-stability groups.ConclusionOur findings indicate that core stability is crucial for maintaining symmetric head motion during rhythmic movements like cycling. Feedback music could serve as an effective tool for promoting symmetry in head motion and thus preventing musculoskeletal injuries.

3D gait analysis in children using wearable sensors: feasibility of predicting joint kinematics and kinetics with personalized machine learning models and inertial measurement units.

Introduction: Children's walking patterns evolve with age, exhibiting less repetitiveness at a young age and more variability than adults. Three-dimensional gait analysis (3DGA) is crucial for understanding and treating lower limb movement disorders in children, traditionally performed using Optical Motion Capture (OMC). Inertial Measurement Units (IMUs) offer a cost-effective alternative to OMC, although challenges like drift errors persist. Machine learning (ML) models can mitigate these issues in adults, prompting an investigation into their applicability to a heterogeneous pediatric population. This study aimed at 1) quantifying personalized and generalized ML models' performance for predicting gait time series in typically developed (TD) children using IMUs data, 2) Comparing random forest (RF) and convolutional neural networks (CNN) models' performance, 3) Finding the optimal number of IMUs required for accurate predictions. Methodology: Seventeen TD children, aged 6 to 15, participated in data collection involving OMC, force plates, and IMU sensors. Joint kinematics and kinetics (targets) were computed from OMC and force plates' data using OpenSim. Tsfresh, a Python package, extracted features from raw IMU data. Each target's ten most important features were input in the development of personalized and generalized RF and CNN models. This procedure was initially conducted with 7 IMUs placed on all lower limb segments and then performed using only two IMUs on the feet. Results: Findings suggested that the RF and CNN models demonstrated comparable performance. RF predicted joint kinematics with a 9.5% and 19.9% NRMSE for personalized and generalized models, respectively, and joint kinetics with an NRMSE of 10.7% for personalized and 15.2% for generalized models in TD children. Personalized models provided accurate estimations from IMU data in children, while generalized models lacked accuracy due to the limited dataset. Furthermore, reducing the number of IMUs from 7 to 2 did not affect the results, and the performance remained consistent. Discussion: This study proposed a promising personalized approach for gait time series prediction in children, involving an RF model and two IMUs on the feet.

A New Smartphone-Based Method for Remote Health Monitoring: Assessment of Respiratory Kinematics

The remote monitoring of clinical parameters plays a fundamental role in different situations, like pandemic health emergencies and post-surgery conditions. In these situations, the patients might be impeded in their movements, and it could be difficult to have specific health monitoring. In recent years, technological advances in smartphones have opened up new possibilities in this landscape. The present work aims to propose a new method for respiratory kinematics monitoring via smartphone sensors. In particular, a specific application was developed to register inertial measurement unit (IMU) sensor data from the smartphone for respiratory kinematics measurement and to guide the user through a specific acquisition session. The session was defined to allow the monitoring of the respiratory movement in five prescribed positions. The application and the sequence were successfully tested on a given population of 77 healthy volunteers. The resulting accelerometers and gyroscope signals were processed to evaluate the significance of differences according to participants’ sex, vector components, and smartphone positioning and, finally, to estimate the respiratory rate. The statistical differences that emerged revealed the significance of information in the different acquisition positions.

Using In-Shoe Inertial Measurement Unit Sensors to Understand Daily-Life Gait Characteristics in Patients With Distal Radius Fractures During 6 Months of Recovery: Cross-Sectional Study.

A distal radius fracture (DRF) is a common initial fragility fracture among women in their early postmenopausal period, which is associated with an increased risk of subsequent fractures. Gait assessments are valuable for evaluating fracture risk; inertial measurement units (IMUs) have been widely used to assess gait under free-living conditions. However, little is known about long-term changes in patients with DRF, especially concerning daily-life gait. We hypothesized that, in the long term, the daily-life gait parameters in patients with DRF could enable us to reveal future risk factors for falls and fractures. This study assessed the spatiotemporal characteristics of patients with DRF at 4 weeks and 6 months of recovery. We recruited 16 women in their postmenopausal period with DRF as their first fragility fracture (mean age 62.3, SD 7.0 years) and 28 matched healthy controls (mean age 65.6, SD 8.0 years). Daily-life gait assessments and physical assessments, such as hand grip strength (HGS), were performed using an in-shoe IMU sensor. Participants' results were compared with those of the control group, and their recovery was assessed for 6 months after the fracture. In the fracture group, at 4 weeks after DRF, lower foot height in the swing phase (P=.049) and higher variability of stride length (P=.03) were observed, which improved gradually. However, the dorsiflexion angle in the fracture group tended to be lower consistently during 6 months (at 4 weeks: P=.06; during 6 months: P=.07). As for the physical assessments, the fracture group showed lower HGS at all time points (at 4 weeks: P<.001; during 6 months: P=.04), despite significant improvement at 6 months (P<.001). With an in-shoe IMU sensor, we discovered the recovery of spatiotemporal gait characteristics 6 months after DRF surgery without the participants' awareness. The consistently unchanged dorsiflexion angle in the swing phase and lower HGS could be associated with fracture risk, implying the high clinical importance of appropriate interventions for patients with DRF to prevent future fractures. These results could be applied to a screening tool for evaluating the risk of falls and fractures, which may contribute to constructing a new health care system using wearable devices in the near future.

Assessment of Noise of MEMS IMU Sensors of Different Grades for GNSS/IMU Navigation.

Inertial measurement units (IMUs) are key components of various applications including navigation, robotics, aerospace, and automotive systems. IMU sensor characteristics have a significant impact on the accuracy and reliability of these applications. In particular, noise characteristics and bias stability are critical for proper filter settings to perform a combined GNSS/IMU solution. This paper presents an analysis based on the Allan deviation of different IMU sensors that correspond to different grades of micro-electromechanical systems (MEMS)-type IMUs in order to evaluate their accuracy and stability over time. The study covers three IMU sensors of different grades (ascending order): Rokubun Argonaut navigator sensor (InvenSense TDK MPU9250), Samsung Galaxy Note10 phone sensor (STMicroelectronics LSM6DSR), and NovAtel PwrPak7 sensor (Epson EG320N). The noise components of the sensors are computed using overlapped Allan deviation analysis on data collected over the course of a week in a static position. The focus of the analysis is to characterize the random walk noise and bias stability, which are the most critical for combined GNSS/IMU navigation and may differ or may not be listed in manufacturers' specifications. Noise characteristics are calculated for the studied sensors and examples of their use in loosely coupled GNSS/IMU processing are assessed. This work proposes a structured and reproducible approach for working with sensors for their use in navigation tasks in combination with GNSS, and can be used for sensors of different levels to supplement missing or incorrect sensor manufacturers' data.

Improving the underwater navigation performance of an IMU with acoustic long baseline calibration

Underwater acoustic Long-Baseline System (LBL) is an important technique for submarine positioning and navigation. However, the high cost of the seafloor equipment and complex construction of a seafloor network restrict the distribution of the LBL within a small area, making an underwater vehicle difficult for long-distance and high-precision acoustic-based or inertial-based navigation. We therefore propose an acoustic LBL-based Inertial Measurement Unit (IMU) calibration algorithm. When the underwater vehicle can receive the acoustic signal from a seafloor beacon, the IMU is precisely calibrated to reduce the cumulative error of Strapdown Inertial Navigation System (SINS). In this way, the IMU is expected to maintain a certain degree of accuracy by relying solely on SINS when the vehicle reaches out the range of the LBL network and cannot receive the acoustic signal. We present the acoustic LBL-based IMU online calibration model and analyze the factors that affect the accuracy of IMU calibration. The results fulfill the expectation that the gyroscope bias and accelerometer bias are the main error sources that affect the divergence of SINS position errors, and the track line of the underwater vehicle directly affects the accuracy of the calibration results. In addition, we deduce that an optimal calibration trajectory needs to consider the effects of the three-dimensional observability and position dilution of precision. In the experiment, we compare the effects of seven calibration trajectories: straight and diamond-shaped with and without the change of depth, and three sets of curves with the change of depth: circular, S-shaped, and figure-eight. Among them, we find that the figure-eight is the optimal trajectory for acoustic LBL-based IMU online calibration. We take the maintenance period during which the accumulated SINS Three Dimensional (3D) position errors are below 1 km to evaluate the calibration performance. The filed experimental results show that for the Micro-electromechanical Systems-grade IMU sensor, the maintenance period for the IMU calibrated with the proposed algorithm can be increased by 121% and 38.9% compared to the IMU without calibration and with the laboratory default parameter calibration, indicating the effectiveness of the proposed calibration algorithm.

Multi-sensor fusion based optimized deep convolutional neural network for boxing punch activity recognition

Recent advancements in deep learning have significantly enhanced the recognition of player activities in sports by enabling automatic feature extraction. In our proposed work, we focus on recognizing six distinct punches in the context of boxing. We incorporate the sliding window technique during the pre-processing stage to transform the time-series data obtained from Inertial Measurement Unit (IMU) sensors in a boxing punch activity recognition system. Our approach influences a sensor fusion-based Deep Convolutional Neural Network (DCNN) classification model to identify various boxing punches accurately, achieving an impressive F1 score. The system demonstrates proficiency in distinguishing similar activities, such as jab and hook punches where the existing systems made misclassifications due to subtle variations in arm flexion that differentiate the two. Through experimentation, we identify an optimal window size for boxing punch activity recognition, which falls within the range of 15–20 frames (equivalent to 0.15–0.25 s). This window size selection results in a notable reduction in inference time. To evaluate our proposed model, we conduct comparisons with a standard DCNN and an optimized DCNN model. Our proposed optimized DCNN model demonstrates enhanced recognition accuracy, achieving an impressive 99%, coupled with an improved F1 score of 87%. Furthermore, the model displays a remarkable reduction in inference time, clocking in at less than 1 ms. Overall, our research contributes to the field of sports-related player activity recognition by employing the power of deep learning. By expertly combining these techniques, we achieve remarkable accuracy, precision, and efficiency in recognizing various boxing punches.

Distributed IMU Sensors for In-Field Dynamic Measurements on an Alpine Ski.

Modern ski design is an inherently time-consuming process that involves an iterative feedback loop comprised of design, manufacturing and in-field qualitative evaluations. Additionally consumers can only rely on qualitative evaluation for selecting the ideal ski, and due to the variation in skier styles and ability levels, consumers can find it to be an inconsistent and expensive experience. We propose supplementing the design and evaluation process with data from in-field prototype testing, using a modular sensor array that can be ported to nearly any ski. This paper discusses a new distributed Inertial Measurement Unit (IMU) suite, including details regarding the design and operation, sensor validation experiments, and outdoor in-field testing results. Data are collected from a set of spatially distributed IMUs located on the upper surface of the ski. We demonstrate that this system and associated post-processing algorithms provide accurate data at a high rate (>700 Hz), enabling the measurement of both structural and rigid ski characteristics, and are robust to repetitive testing in outdoor winter conditions.

Validation of LEOMO inertial measurement unit sensors with marker-based three-dimensional motion capture during maximum sprinting in track cyclists

ABSTRACT LEOMO™ is a commercial inertial measurement unit system that provides cycling-specific motion performance indicators (MPIs) and offers a mobile solution for monitoring cyclists. We aimed to validate the LEOMO sensors during sprint cycling using gold-standard marker-based three-dimensional (3D) motion technology (Qualisys, AB). Our secondary aim was to explore the relationship between peak power during sprints and MPIs. Seventeen elite track cyclists performed 3 × 15s seated start maximum efforts on a cycle ergometer. Based on intraclass correlation coefficient (ICC3,1), the MPIs derived from 3D and LEOMO showed moderate agreement (0.50 < 0.75) for the right foot angular range (FAR); left foot angular range first quadrant (FARQ1); right leg angular range (LAR); and mean angle of the pelvis in the sagittal plane. Agreement was poor (ICC < 0.50) between MPIs derived from 3D and LEOMO for the left FAR, right FARQ1, left LAR, and mean range of motion of the pelvis in the frontal and transverse planes. Only one LEOMO-derived (pelvic rotation) and two 3D-derived (right FARQ1 and FAR) MPIs showed large positive significant correlations with peak power. Caution is advised regarding use of the LEOMO for short maximal cycling efforts and derived MPIs to inform peak sprint cycling power production.

IMU-Based Approach to Detect Spastic Cerebral Palsy in Infants at Early Stages

INTRODUCTION: Cerebral Palsy (CP) is a non-progressive neurological disorder affecting muscle control in early childhood, leading to permanent alterations in body posture and movement. Early identification is crucial for accurate diagnosis and therapy-based interventions. In recent years, an automated monitoring system has been developed to facilitate the health assessment of infants, enabling early recognition of neurological dysfunctions in high-risk infants. However, the interpretation of these assessments lacks standardization and is subject to examiner bias.&#x0D; OBJECTIVES: Many infants with CP exhibit increased tonic stretch reflexes due to Upper Motor Neuron Syndrome (UMNS), resulting from motor neuron damage that disrupts muscle signalling.&#x0D; METHOD: To detect abnormal muscle reactions, our team employed an Inertial Measurement Unit (IMU) sensor, comprising three tri-axial sensors (accelerometer, gyroscope, magnetometer) that capture movement data continuously and unobtrusively. IMU sensors are compact, cost-effective, and have low processing requirements, requiring attachment to the infant's body to measure inter-body part angles. Our team analyzed muscle activity and posture using IMU sensors, collecting tri-axial data from 43 infants in real-time. Additional factors like age, stride length, and leg length were incorporated into the dataset.&#x0D; RESULTS: Our team has applied various supervised machine learning approaches to predict CP in infants due to the limited dataset size, validating models through k-fold cross-validation. Among the models, Naive Bayes (NB) outperformed Logistic Regression (LR), Decision Tree (DT), Linear Discriminant Analysis (LDA), k-Nearest Neighbors (kNN), and Support Vector Machine (SVM), achieving an accuracy of 88%. CONCLUSION: This research contributes to the early detection and intervention of CP in infants, potentially improving their long-term outcomes.

Robust human locomotion and localization activity recognition over multisensory.

Human activity recognition (HAR) plays a pivotal role in various domains, including healthcare, sports, robotics, and security. With the growing popularity of wearable devices, particularly Inertial Measurement Units (IMUs) and Ambient sensors, researchers and engineers have sought to take advantage of these advances to accurately and efficiently detect and classify human activities. This research paper presents an advanced methodology for human activity and localization recognition, utilizing smartphone IMU, Ambient, GPS, and Audio sensor data from two public benchmark datasets: the Opportunity dataset and the Extrasensory dataset. The Opportunity dataset was collected from 12 subjects participating in a range of daily activities, and it captures data from various body-worn and object-associated sensors. The Extrasensory dataset features data from 60 participants, including thousands of data samples from smartphone and smartwatch sensors, labeled with a wide array of human activities. Our study incorporates novel feature extraction techniques for signal, GPS, and audio sensor data. Specifically, for localization, GPS, audio, and IMU sensors are utilized, while IMU and Ambient sensors are employed for locomotion activity recognition. To achieve accurate activity classification, state-of-the-art deep learning techniques, such as convolutional neural networks (CNN) and long short-term memory (LSTM), have been explored. For indoor/outdoor activities, CNNs are applied, while LSTMs are utilized for locomotion activity recognition. The proposed system has been evaluated using the k-fold cross-validation method, achieving accuracy rates of 97% and 89% for locomotion activity over the Opportunity and Extrasensory datasets, respectively, and 96% for indoor/outdoor activity over the Extrasensory dataset. These results highlight the efficiency of our methodology in accurately detecting various human activities, showing its potential for real-world applications. Moreover, the research paper introduces a hybrid system that combines machine learning and deep learning features, enhancing activity recognition performance by leveraging the strengths of both approaches.

Navigation Path Following Platform for a Greenhouse Shuttle Robot Using the State-flow Method

Localization is an important method for autonomous indoor robots to recognize their positions. Generally, the navigation of a mobile robot is conducted using a camera, Lidar, and global positioning system. However, for an indoor environment, GPS is unavailable. Therefore, a, state-trajectory tracking method is utilized based on a Lidar map. This paper presents the path following of an autonomous indoor mobile robot, that is, a shuttle robot, using a state-flow method via a robot operating system network. MATLAB and Linux high-level computers and an inertial measurement unit sensor are used to obtain the Cartesian coordinate information of a bicycle-type mobile robot. The path following problem can be solved in the state-flow block by setting appropriate time and linear and angular velocity variables. After the predetermined time, the linear and angular velocities are set based on the length of the path and radius of the quarter-circle of the left and right turns in the state-flow block, path planning, which can execute the work effectively, is established using the state-flow algorithm. The state-flow block produces time-series data that are sent to Linux system, which facilitates real-time mobile platform path following scenario. Several cases within the path-following problem of the mobile robot were considered, depending on the linear and angular velocity settings: the mobile robot moved forward and backward, turned in the right and left directions on the circular path. The effectiveness of the method was demonstrated using the desktop-based indoor mobile robot control results. Thus, the paper focuses on the application of the state-flow algorithm to the shuttle robot specifically in the narrow indoor environment.