Search results

The purpose of this paper is to present fusion of inertial navigation system (INS) and global positioning system (GPS) for estimating position, velocities, attitude and heading of an unmanned aerial vehicle (UAV).

A 15-state extended Kalman filter (EKF) and a split architecture consisting of six-state nonlinear complementary filter (NCF) and nine-state EKF are investigated in detail. In both these fusion architectures GPS and inertial measurement unit consisting of three axis accelerometers, three axis rate gyros and three axis magnetometer have been fused in open loop fashion (loosely coupled) to estimate the navigation states.

These architectures have been implemented in MATLAB/SIMULINK environment and evaluated in closed loop guidance of Black-Kite MAV with software-in-the-loop-simulation (SILS) setup. Furthermore, both the algorithms are validated with flight test data obtained from on-board data logger using an off-the shelf autopilot board (Ardupilot Mega APM-2.5) on SLYBIRD UAV.

The proposed architectures are of high value to accomplish INS/GPS fusion, which plays a vital role in autonomous guidance and navigation of an UAV.

An accelerometer is a device that measures force due to gravity or a change in speed or direction of travel. This paper describes accelerometers and their application in other disciplines and, by way of an example, explores the utility of accelerometers for studying aggression. We end with a discussion of additional ways accelerometers might be used in group processes research.

We first review the use of accelerometers in other disciplines. We then present the results of four studies that demonstrate the use of accelerometers to measure aggression. Study 1 establishes the measure’s concurrent validity. Study 2 concerns its stability and representative reliability. Study 3 seeks to establish the measure’s predictive validity by associating it with an existing measure. Study 4 demonstrates the ability of accelerometers to address a sociological research question.

In Studies 1 and 2, we find that accelerometers can be used to differentiate between distinct levels of aggression. In Study 3, we find that men’s average peak acceleration correlates with a previously validated measure of aggression. Study 4 uses accelerometers to reproduce a well-established finding in the aggression literature.

We conclude that accelerometers are a flexible tool for group processes’ researchers and social scientists more broadly. Our findings should prove useful to social scientists interested in measuring aggression or in employing accelerometers in their work.

In human action recognition based on wearable sensors, most previous studies have focused on a single type of sensor and single classifier. This study aims to use a wearable sensor based on flexible sensors and a tri-axial accelerometer to collect action data of elderly people. It uses a statistical modeling approach based on the ensemble algorithm to classify actions and verify its validity.

Nine types of daily actions were collected by the wearable sensor device from a group of elderly volunteers, and the time-domain features of the action sequences were extracted. The dimensionality of the feature vectors was reduced by linear discriminant analysis. An ensemble learning method based on XGBoost was used to build a model of elderly action recognition. Its performance was compared with the action recognition rate of other algorithms based on the Boosting algorithm, and with the accuracy of single classifier models.

The effectiveness of the method was validated by three experiments. The results show that XGBoost is able to classify nine daily actions of the elderly and achieve an average recognition rate of 94.8 per cent, which is superior to single classifiers and to other ensemble algorithms.

The research could have important implications for health care, including the treatment and rehabilitation of the elderly, and the prevention of falls.

Instead of using a single type of sensor, this research used a wearable sensor to obtain daily action data of the elderly. The results show that, by using the appropriate method, the device can obtain detailed data of joint action at a low cost. Comparing differences in performance, it was concluded that XGBoost is the most suitable algorithm for building a model of elderly action recognition. This method, together with a wearable sensor, can provide key data and accurate feedback information to monitor the elderly in their rehabilitation activities.

Numerical models are being increasingly relied upon to evaluate wind turbine performance by simulating phenomena that are infeasible to measure experimentally. These numerical models, however, require a large number of input parameters that often need to be calibrated against available experiments. Owing to the unavoidable scarcity of experiments and inherent uncertainties in measurements, this calibration process may yield non-unique solutions, i.e. multiple sets of parameters may reproduce the available experiments with similar fidelity. The purpose of this paper is to study the trade-off between fidelity to measurements and the robustness of this fidelity to uncertainty in calibrated input parameters.

Here, fidelity is defined as the ability of the model to reproduce measurements and robustness is defined as the allowable variation in the input parameters with which the model maintains a predefined level of threshold fidelity. These two vital attributes of model predictiveness are evaluated in the development of a simplified finite element beam model of the CX-100 wind turbine blade.

Findings of this study show that calibrating the input parameters of a numerical model with the sole objective of improving fidelity to available measurements degrades the robustness of model predictions at both tested and untested settings. A more optimal model may be obtained by calibration methods considering both fidelity and robustness. Multi-criteria Decision Making further confirms the conclusion that the optimal model performance is achieved by maintaining a balance between fidelity and robustness during calibration.

Current methods for model calibration focus solely on fidelity while the authors focus on the trade-off between fidelity and robustness.

The purpose of this study is to explore the use of smartphone-embedded microelectro-mechanical sensors (MEMS) for accurately estimating rotating machinery speed, crucial for various condition monitoring tasks. Rotating machinery (RM) serves a crucial role in diverse applications, necessitating accurate speed estimation essential for condition monitoring (CM) tasks such as vibration analysis, efficiency evaluation and predictive assessment.

This research explores the utilization of MEMS embedded in smartphones to economically estimate RM speed. A series of experiments were conducted across three test setups, comparing smartphone-based speed estimation to traditional methods. Rigorous testing spanned various dimensions, including scenarios of limited data availability, diverse speed applications and different smartphone placements on RM surfaces.

The methodology demonstrated exceptional performance across low and high-speed contexts. Smartphones-MEMS accurately estimated speed regardless of their placement on surfaces like metal and fiber, presenting promising outcomes with a mere 6 RPM maximum error. Statistical analysis, using a two-sample t-test, compared smartphone-derived speed outcomes with those from a tachometer and high-quality (HQ) data acquisition system.

The research limitations include the need for further investigation into smartphone sensor calibration and accuracy in extremely high-speed scenarios. Future research could focus on refining these aspects.

The societal impact is substantial, offering cost-effective CM across various industries and encouraging further exploration of MEMS-based vibration monitoring.

This research showcases an innovative approach using smartphone-embedded MEMS for RM speed estimation. The study’s multidimensional testing highlights its originality in addressing scenarios with limited data and varied speed applications.

Quantitative gait analysis is an important research area to enable physiotherapist to perform systematic studies and health diagnosis of the lower extremity of patients throughout the rehabilitation. The quantitative measurement of ambulatory gait can be performed by using convenient and economical wearable sensors outside specialized motion laboratories. In this paper, a sensor system consisting of three tri-axial accelerometers, two flex sensors and four force sensors was developed. Subject testing were carried out to obtain temporal and spatial gait parameters. The performance of the sensor suite is compared to results from camera videos analysed by Kinovea motion tracking software.

The purpose of this paper is to develop a highly miniaturized wireless inertial sensor system based on a novel 3D packaging technique using a flexible printed circuit (FPC). The device is very suitable for wearable applications in which small size and lightweight are required such as body area network, medical, sports and entertainment applications.

Modern wireless inertial measurement units are typically implemented on a rigid 2D printed circuit board (PCB). The design concept presented here is based around the use of a novel planar, six‐faceted, crucifix or cross‐shaped FPC instead of a rigid PCB. A number of specific functional blocks (such as microelectromechanical systems gyroscope and accelerometer sensors, microcontroller (MCU), radio transceiver, antenna, etc.) are first assigned to each of the six faces which are each 1 cm² in area. The FPC cross is then developed into a 1 cm³, 3D configuration by folding the cross at each of five bend planes. The result is a low‐volume and lightweight, 1 cm³ wireless inertial sensor that can sense and send motion sensed data wirelessly to a base station. The wireless sensor device has been designed for low power operation both at the hardware and software levels. At the base station side, a radio receiver is connected to another MCU unit, which sends received data to a personal computer (PC) and graphical user interface. The industrial, scientific and medical band (2.45 GHz) is used to achieve half duplex communication between the two sides.

A complete wireless sensor system has been realized in a 3D cube form factor using an FPC. The packaging technique employed during the work is shown to be efficient in fabricating the final cubic system and resulted in a significant saving in the final size and weight of the system. A number of design issues are identified regarding the use of FPC for implementing the 3D structure and the chosen solutions are shown to be successful in dealing with these issues.

Currently, a limitation of the system is the need for an external battery to power the sensor system. A second phase of development would be required to investigate the possibility of the integration of a battery and charging system within the cube structure. In addition, the use of flexible substrate imposes a number of restrictions in terms of the ease of manufacturability of the final system due to the requirement of the required folding step.

The small size and weight of the developed system is found to be extremely useful in different deployments. It would be useful to further explore the system performance in different application scenarios such as wearable motion tracking applications. In terms of manufacturability, component placement needs to be carefully considered, ensuring that there is sufficient distance between the components, bend planes and board edges and this leads to a slightly reduced usable area on the printed circuit.

This paper provides a novel and useful method for realizing a wireless inertial sensor system in a 3D package. The value of the chosen approach is that a significant reduction in the required system volume is achieved. In particular, a 78.5 per cent saving in volume is obtained in decreasing the module size from a 25 to a 15 mm³ size.

The aim of this paper is threefold: first, to review the technological state of the art on tire sensor systems; second, to summarize basic methodologies and explore the potential of tire sensing for intelligent vehicle developments and third, to address challenges in the development of tire sensing systems and inspire future research in this field.

Nowadays, automotive industry is moving toward an intelligent and autonomous driving era with the assistance of sensing technology development, whereas tire-road conditions sensing and utilization are of great interest from the point of view of vehicle dynamics control, vehicle safety and vehicle performance evaluation.

Tire sensing is an emerging technology whereby sensor systems are installed on the tire to provide fundamental insights into tire-road interactions for ground vehicles and wheel robots. In the past two decades, tire sensing systems based on various sensor types have been proposed to offer the possibility to investigate tire-road interactions.

Instrumenting the tire with sensors, especially accelerometers and optical sensors, can sense the tire-road interactions and enhance the vehicle performance. The harsh environment inside tire cavity requires reliable, accurate, low weight, modularized and inexpensive sensors. Challenges, such as the data transmission, power management, lack of physics-based tire models need to be solved before the tire sensor becomes commercially viable for production vehicles.

The paper aims to develop a novel method for the classification of different physical activities of a human being, using fabric sensors. This method focuses mainly on classifying the physical activity between normal action and violent attack on a victim and verifies its validity.

The system is realized as a protective jacket that can be worn by the subject. Stretch sensors, pressure sensors and a 9 degree of freedom accelerometer are strategically woven on the jacket. The jacket has an internal bus system made of conductive fabric that connects the sensors to the Flora chip, which acts as the data acquisition unit for the data generated. Different activities such as still, standing up, walking, twist-jump-turn, dancing and violent action are performed. The jacket in this study is worn by a healthy subject. The main phases which describe the activity recognition method undertaken in this study are the placement of sensors, pre-processing of data and deploying machine learning models for classification.

The effectiveness of the method was validated in a controlled environment. Certain challenges are also faced in building the experimental setup for the collection of data from the hardware. The most tedious challenge is to collect the data without noise and error, created by voltage fluctuations when stretched. The results show that the support vector machine classifier can classify different activities and is able to differentiate normal action and violent attacks with an accuracy of 98.8%, which is superior to other methods and algorithms.

This study leads to an understanding of human physical movement under violent activity. The results show that data compared with normal physical motion, which includes even a form of dance is quite different from the data collected during violent physical motion. This jacket construction with woven sensors can capture every dimension of the physical motion adding features to the data on which the machine learning model will be built.

Unlike other studies, where sensors are placed on isolated parts of the body, in this study, the fabric sensors are woven into the fabric itself to collect the data and to achieve maximum accuracy instead of using isolated wearable sensors. This method, together with a fabric pressure and stretch sensors, can provide key data and accurate feedback information when the victim is being attacked or is in a normal state of action.

This paper seeks to present an inertial motion tracking system for monitoring movements of human upper limbs in order to support a home‐based rehabilitation scheme in which the recovery of stroke patients' motor function through repetitive exercises needs to be continuously monitored and appropriately evaluated.

Two inertial sensors are placed on the upper and lower arms in order to obtain acceleration and turning rates. Then the position of the upper limbs can be deduced by using the kinematical model of the upper limbs that was designed in the previous paper. The tracking system starts from inertial data acquisition and pre‐filtering, followed by a number of processes such as transformation of coordinate systems of sensor data, and kinematical modelling and optimization of position estimation.

The motion detector using the proposed kinematic model only has drifts in the measurements. Fusion of acceleration and orientation data can effectively solve the drift problem without the involvement of a Kalman filter.

The image rendering is not undertaken when the data sampling is performed. This non‐synchronization is applied in order to avoid the breaks in the continuous sampling.

This new motion detector can work in different environments without significant drifts. Also, this system only deploys two inertial sensors but is able to estimate the position of the wrist, elbow and shoulder joints.