Non-intrusive Estimation Research Articles

Body Weight Estimation of Cattle in Standing and Lying Postures Using Point Clouds Derived from Unmanned Aerial Vehicle-Based LiDAR

This study aims to explore body weight estimation for cattle in both standing and lying postures, using 3D data. We apply a Unmanned Aerial Vehicle-based (UAV-based) LiDAR system to collect data during routine resting periods between feedings in the natural husbandry conditions of a commercial farm, which ensures minimal interruption to the animals. Ground truth data are obtained by weighing cattle as they voluntarily pass an environmentally embedded scale. We have developed separate models for standing and lying postures and trained them on features extracted from the segmented point clouds of cattle with unique identifiers (UIDs). The models for standing posture achieve high accuracy, with a best-performance model, Random Forest, obtaining an R2 of 0.94, an MAE of 4.72 kg, and an RMSE of 6.33 kg. Multiple linear regression models are trained to estimate body weight for the lying posture, using volume- and posture-wise characteristics. The model used 1 cm as the thickness of the slice-wise volume calculation, achieving an R2 of 0.71, an MAE of 7.71 kg, and an RMSE of 9.56 kg. These results highlight the potential of UAV-based LiDAR data for accurate and non-intrusive estimation of cattle body weight in lying and standing postures, which paves the way for improved management practices in precision livestock farming.

Modeling the Mechanical Properties of Root–Substrate Interaction with a Transplanter Using Artificial Neural Networks

The mechanical properties of a plug seedling substrate determine whether it will crush during the transplantation, thereby affecting the integrity of the root system and the survival rate of transplanted seedlings. In this study, we measured eight morphological parameters of pepper seedlings using machine vision and physical methods, and the corresponding substrate mechanical parameters of the plug seedlings were tested using a texture analyzer. Based on the experimental data, a BPNN framework was constructed to predict the substrate mechanical properties of plug seedlings at different growth stages. The results indicate that the BPNN with a framework of [8, 15, 15, 1] exhibits higher R2 and lower errors. The mean absolute error (MAE), mean squared error (MSE), and mean absolute percentage error (MAPE) values are 7.669, 88.842, and 9.076%, respectively, with an R2 of 0.867. The average prediction accuracy of 20 test data set is 90.472%. Finally, predictions and experimental validations were conducted on the substrate mechanical properties of seedlings grown for 47 days. The results revealed that the BPNN achieved an average prediction accuracy of 93.282%. Additionally, it exhibited faster speed and lower computational costs. This study provides a reference for the non-intrusive estimation of substrate mechanical properties in plug seedlings and the design and optimization of transplanting an end-effector.

Server load estimation by Burr distribution mixture analysis of TCP SYN response time

Server load estimation by Burr distribution mixture analysis of TCP SYN response time

Prediction of oxygen uptake kinetics during heavy-intensity cycling exercise by machine-learning analysis.

Non-intrusive estimation of oxygen uptake (V̇O2) is possible with wearable sensor technology and artificial intelligence. V̇O2 kinetics have been accurately predicted during moderate exercise using easy-to-obtain sensor inputs. However, V̇O2 prediction algorithms for higher intensity exercise with inherent nonlinearities are still being refined. The purpose of this investigation was to test if a machine learning model can accurately predict dynamic V̇O2 across exercise intensities, including slower V̇O2 kinetics normally observed during heavy- compared to moderate-intensity exercise. Fifteen young healthy adults (7 females; peak V̇O2: 42±5 mL·min-1·kg-1) performed three different pseudorandom binary sequence (PRBS) exercise tests ranging in intensity from low-to-moderate, low-to-heavy, and ventilatory threshold-to-heavy work rates. A temporal convolutional network was trained to predict instantaneous V̇O2, with model inputs including heart rate, percent heart rate reserve, estimated minute ventilation, breathing frequency, and work rate. Frequency domain analyses between V̇O2 and work rate were used to evaluate measured and predicted V̇O2 kinetics. Predicted V̇O2 had low bias (-0.017 L·min-1, 95% limits of agreement: [-0.289, 0.254]), and was very strongly correlated (rrm=0.974, p<0.001) with the measured V̇O2. The extracted indicator of kinetics, mean normalized gain (MNG), was not different between predicted and measured V̇O2 responses (main effect: p=0.374, ηp2=0.01), and decreased with increasing exercise intensity (main effect: p<0.001, ηp2=0.64). Predicted and measured V̇O2 kinetics indicators were moderately correlated across repeated measurements (MNG: rrm=0.680, p<0.001). Therefore, the temporal convolutional network accurately predicted slower V̇O2 kinetics with increasing exercise intensity, enabling non-intrusive monitoring of cardiorespiratory dynamics across moderate- and heavy-exercise intensities.

Non-intrusive estimation of model error and discrepancy in dynamics models

Errors in formulating the system physics model, due to inability to rigorously account for nonlinear behavior, boundary conditions, and multi-component and multi-physics interactions result in discrepancies between model predictions of system responses and the corresponding experimentally measured values. In our previous work, we have developed a Bayesian state estimation-based framework for the estimation of these model errors, and the subsequent prediction of model discrepancies at unmeasured locations, for untested inputs, and for untested configurations. In this approach, we represent model errors as external inputs in the dynamic system model, and estimate them using Bayesian state estimation. However, this approach is intrusive in the sense that it requires access to the governing system of equations, the ability to modify the system model by introducing an external system input, and the ability to represent system inputs and responses as probabilistic quantities. In this paper, we develop semi-intrusive and non-intrusive approaches to extend the model error estimation to black-box models. The semi-intrusive method does not need access to the governing equations, but requires the ability to interrupt the simulation at any time instant, update the system responses at that moment, and resume the simulation using the updated value of system responses. For situations where this is not permitted, we propose a fully non-intrusive approach where, a surrogate model is constructed and used in the Bayesian state estimation procedure. The proposed approaches are illustrated with three examples: structural deformation of a Timoshenko beam, heat transfer across a rigid bar, and deformation of a Timoshenko beam subjected to heating. • Estimating errors in dynamics systems using Bayesian state estimation. • Propagating model errors to an untested configuration. • Non-intrusive methodology for use with black-box models.

A Benchmark of Non-intrusive Parametric Audio Quality Estimation Models for Broadcasting Systems and Web-casting Applications

Due to the rising usage of various broadcasting systems and web-casting applications, a measurement of audio quality has become an essential task. This paper presents a benchmark of the parametric models for non-intrusive estimation of the audio quality perceived by the end user. The proposed solution is based on machine learning techniques for broadcasting systems and web-casting applications. The main goal of this study is to assess the performance of the non-intrusive parametric models as well as to evaluate a statistical significance of the performance differences between those models. The paper provides a comparison of several models based on the Support Vector Regression, Genetic Programming, Multigene Symbolic Regression, Neural Networks and Random Forest. The obtained results indicate that among the investigated models the most accurate, although not the fastest ones, are the model based on Random Forest (a broadcast scenario) and the SVR-based model (a web-cast scenario). These models represent promising candidates for non-intrusive parametric audio quality assessment in the context of broadcasting systems and web-casting applications.

Classification of blood pressure in critically ill patients using photoplethysmography and machine learning

The aim of this study was to evaluate the capability of features extracted from photoplethysmography (PPG) based Pulse Rate Variability (PRV) to classify hypertensive, normotensive and hypotensive events, and to estimate mean arterial, systolic and diastolic blood pressure in critically ill patients. Time-domain, frequency-domain and non-linear indices from PRV were extracted from 5-min and 1-min segments obtained from PPG signals. These features were filtered using machine learning algorithms in order to obtain the optimal combination for the classification of hypertensive, hypotensive and normotensive events, and for the estimation of blood pressure. 5-min segments allowed for an improved performance in both classification and estimation tasks. Classification of blood pressure states showed around 70% accuracy and around 75% specificity. The sensitivity, precision and F1 scores were around 50%. In estimating mean arterial, systolic, and diastolic blood pressure, mean absolute errors as low as 2.55 ± 0.78 mmHg, 4.74 ± 2.33 mmHg, and 1.78 ± 0.14 mmHg were obtained, respectively. Bland-Altman analysis and Wilcoxon rank sum tests showed good agreement between real and estimated values, especially for mean and diastolic arterial blood pressures. PRV-based features could be used for the classification of blood pressure states and the estimation of blood pressure values, although including additional features from the PPG waveform could improve the results. PRV contains information related to blood pressure, which may aid in the continuous, noninvasive, non-intrusive estimation of blood pressure and detection of hypertensive and hypotensive events in critically ill subjects.

An empirical correlation between lift and the properties of leading-edge vortices

Using data from numerical simulations, we show that the lift experienced by both impulsively started and surging airfoils correlates well with the sum of the circulation of the leading-edge vortices truncated at the trailing edge. Therefore, we suggest that reasonable estimates of the lift can be obtained using only two vortex parameters, i.e., its circulation and its position. In addition to being convenient for non-intrusive estimation of forces from PIV measurements, we show that this approach can be used to derive low-order models for the analysis of vortex-lift configurations. In particular, we apply this correlation to model high-amplitude surging, which allows us to quantify the effect of wake-capture mechanisms and to determine the flow parameters that drive optimal lift.

Nonintrusive estimation of subsurface geometrical attributes of the melt pool through the sensing of surface oscillations in laser powder bed fusion

Molten pool geometry, whose surface parameters may be extrapolated through direct process observations, has been identified as a fundamental indicator of stability in laser powder bed fusion (LPBF). However, a parameter that cannot be directly measured on industrial systems by means of conventional sensing equipment is the molten pool depth. Indeed, methods based on x-ray imaging demonstrated in the literature have helped to better understand the process. However, retrofitting such solutions to industrial systems does not appear as a viable route currently. Within the present investigation, a nonintrusive sensing method for the indirect measurement of subsurface molten pool geometry based on the detection of surface oscillations is presented. The analysis of frames acquired using a high-speed camera and a secondary illumination light allows the identification of the crests of capillary waves through bright reflections on the surface of the molten pool. The characteristic oscillation frequency of the surface ripples may be correlated with the penetration depth or to other subsurface geometrical parameters. Proof of concept testing of the sensing principle was conducted on two different materials, namely, AISI316L and IN718, by means of single track LPBF depositions. Experiments were conducted at different levels of laser emission power to induce variations in molten pool characteristics. The process was observed by employing an off-axis illumination light and a high-speed camera, which allowed acquisitions with high spatial and temporal resolution. The acquired frames were postprocessed to extract the oscillation indicator, and analysis of the power spectral density of the signal allowed for the identification of the oscillation frequency. Results show that oscillation frequencies range from 3 to 5.5 kHz. Molten pool penetration depth and cross-sectional area could be correlated with the oscillation frequencies for the inline detection of these parameters during LPBF depositions. For both materials, higher oscillation frequencies corresponded to a shallower molten pool and a smaller mass of molten material. Moreover, different characteristic curves of oscillation frequency variations as a function of the melt pool cross-sectional area were determined for IN718 and AISI316L.

Non-Intrusive Estimation of Single-Port Thevenin Equivalents in AC Grids

This paper deals with the estimation of parameters of a Thevenin equivalent in an AC grid. We consider the practically relevant case of measurements coming from a grid-connected inverter. Our focus is on non-intrusive measurements, as we postulate and later experimentally observe transients that occur in normal operation as the primary source of information. We notice that even during quiet, after work hours there are typically enough variations in the grid to enable successful convergence of our procedure. For completeness, we also consider the artificial case of no external variations and show that injection of very small current perturbations suffices for convergence. Our procedure is based on real-time monitoring of the estimated determinant of the regression matrix. Regression matrix properties are also used to evaluate the integrity of the estimates and to initiate a brief sequence of perturbation in rare cases of the exceptionally restful grid. The estimation algorithm is simple enough to be implementable on standard industrial controllers, yet robust and reliable in terms of speed of convergence.

Non-destructive Estimation of Chlorophyll a Content in Red Delicious Apple Cultivar Based on Spectral and Color Data

Non-destructive estimation of the chemical properties of fruit is an important goal of researchers in the food industry, since online operations, such as fruit packaging based on the amount of different chemical properties and determining different stages of handling, are done based on these estimations. In this study, chlorophyll a content in Red Delicious apple cultivar is predicted as a chemical property that is altered by apple ripening stage, using non-destructive spectral and color methods combined. Two artificial intelligence methods based on hybrid Multilayer Perceptron Neural Network - Artificial Bee Colony Algorithm (ANN-ABC) and Partial least squares regression (PLSR) were used in order to obtain a non-destructive estimation of chlorophyll a content. In application of the PLSR method, various pre-processing algorithms were used. In order to statistically properly validate the hybrid ANN-ABC predictive method, 20 runs were performed. Results showed that the best regression coefficient of the PLSR method in predicting chlorophyll a content using spectral data alone was 0.918. At the same time, the average determination coefficient over 20 repetitions in hybrid ANN-ABC in the estimation of chlorophyll a content, using spectral data and color features were higher than 0.92±0.040 and 0.89±0.045, respectively, which to our knowledge is a remarkable non-intrusive estimation result.

On A Posteriori Estimation of the Approximation Error Norm for an Ensemble of Independent Solutions

An ensemble of independent numerical solutions makes it possible to construct a hypersphere around an approximate solution that contains the true solution. The analysis is based on some geometry considerations, such as the triangle inequality and the measure concentration in spaces of large dimensions. As a result, a nonintrusive postprocessor providing error estimation on an ensemble of solutions can be constructed. Some numerical tests for the two-dimensional compressible Euler equations are given to demonstrate the properties of such postprocessing.

Towards Contactless Estimation of Electrodermal Activity Correlates.

This paper presents a proof-of-concept for contactless and nonintrusive estimation of electrodermal activity (EDA) correlates using a camera. RGB video of the palm under three different lighting conditions showed that for a suitably chosen illumination strategy the data from the camera is sufficient to estimate EDA correlates which agree with the measurements done using laboratory grade physiological sensors. The effects we see in the recorded video can be attributed to sweat gland activity, which inturn is known to be correlated with EDA. These effects are so pronounced that simple pixel statistics can be used to quantify them. Such a method benefits from advances in computer vision and graphics research and has the potential to be used in affective computing and psychophysiology research where contact based sensors may not be suitable.

Fast non-intrusive estimation of liquid-phase interface and boundary heat flux in participating media by the Kalman filtering methods

Abstract The non-intrusive inverse heat transfer technique based on the Kalman filtering (KF) method is proposed for on-line retrieving the time-dependent boundary heat flux, internal temperature distribution, and liquid-phase interface of the participating medium simultaneously. To obtain the measured temperature signals, the nonlinear conduction-radiation heat transfer with phase change in the participating medium is solved by the enthalpy method combined with discrete ordinate method. Three different types of retrieval algorithms have been employed: the original Kalman filter (KF), the extended KF (EKF), and the unscented KF (UKF). The ideal participating media, which is assumed to be anisotropic scattering, and opaque and diffuse gray boundary, is employed to verify the reliability and validity of the proposed algorithm. Comparing with the EKF and UKF, the original KF cannot be applied to solve the nonlinear problems. Comparing with the UKF, the EKF can only be employed to solve the weak nonlinear problem. For the inverse coupled conduction-radiation problem with phase change, the UKF is proved to be more robust to estimate the time-dependent heat flux, internal temperature distribution, and liquid-phase interface simultaneously in real time.

An Automatic Non-Destructive Method for the Classification of the Ripeness Stage of Red Delicious Apples in Orchards Using Aerial Video

The estimation of the ripening state in orchards helps improve post-harvest processes. Picking fruits based on their stage of maturity can reduce the cost of storage and increase market outcomes. Moreover, aerial images and the estimated ripeness can be used as indicators for detecting water stress and determining the water applied during irrigation. Additionally, they can also be related to the crop coefficient (Kc) of seasonal water needs. The purpose of this research is to develop a new computer vision algorithm to detect the existing fruits in aerial images of an apple cultivar (of Red Delicious variety) and estimate their ripeness stage among four possible classes: unripe, half-ripe, ripe, and overripe. The proposed method is based on a combination of the most effective color features and a classifier based on artificial neural networks optimized with genetic algorithms. The obtained results indicate an average classification accuracy of 97.88%, over a dataset of 8390 images and 27,687 apples, and values of the area under the ROC (receiver operating characteristic) curve near or above 0.99 for all classes. We believe this is a remarkable performance that allows a proper non-intrusive estimation of ripening that will help to improve harvesting strategies.

Conductimetry technique for the measurement of thin liquid film thickness between two solid surfaces in relative motion: hydrodynamic lubrication

This paper presents an approach for the measurement of a liquid film thickness in contact with a rough surface, based on electrical conductance sensor. This type of sensor consists of two electrodes mounted flush on a wall and using an electrolyte solution representing the liquid film. Assessing of the generated current between a pair of electrode is used as a measure of the film thickness. The liquid film is contained between two parallel surfaces that one of which is coated with a certain roughness, while the other is smooth. The electrical probes are placed on the smooth surface and the facing rough surface is removable allowing a free move of the wall. In this way, a rotational or a sliding motion is imposed on the rough wall allowing a browse of an entire surface relative to the electrode, in order to determine the influence of roughness on the film thickness measurements. In addition, series of signal acquisition were carried out with imposed pressures on the upper plate for the characterization of the effect of the film compression on the measured thicknesses. The principle of this electrochemical technique is briefly explained, as well as how the lateral distance between the electrodes impacts the measuring range limit. The experimental setup is described and used to study the liquid film flow with several configurations of the wall surface. The obtained results demonstrated the feasibility of this type of non-intrusive estimation, as it was possible to estimate the variable film thickness which is depending on the peaks and valleys of the rough surface. This is promising, because measuring this parameter remains difficult. Finally, the results analysis allows to synthesize the advantages and limitations of this method.

Non-intrusive estimation of available throughput for IEEE 802.11 link

Abstract Knowing the available throughput of an IEEE 802.11 link before establishing the connection is beneficial. One usage of such information is the selection of an Access Point (AP) that provides the highest throughput to a user station. In this paper, we propose a scheme that estimates the expected throughput of an 802.11 device in a non-intrusive manner. The proposed scheme does not require any aid from the infrastructure (e.g., AP, router) or other user stations. It does not generate any additional traffic overhead (e.g., probe traffic). Our scheme relies solely on passive sniffing of the on-going traffic for short duration (i.e., less than the typical 802.11 beacon interval). Our scheme works with various 802.11 technologies including 802.11n with frame aggregation and it works even in the situation when it is not possible to sniff complete traffic trace. Our scheme is not affected by the 802.11 security mechanisms, because we only need information from the un-encrypted header part of 802.11 frames. The proposed scheme can be implemented practically in any system and its resource overhead is very low. We have implemented our scheme on a Linux-based notebook and an Android-based smart phone. Experimental evaluation shows that our scheme accurately estimates the expected throughput in various settings.

Magnetic sensor-based simultaneous state and parameter estimation using a nonlinear observer

ABSTRACTThis paper presents an observer design technique for a newly developed non-intrusive position estimation system based on magnetic sensors. Typically, the magnetic field of an object as a function of position needs to be represented by a highly nonlinear measurement equation. Previous results on observer design for nonlinear systems have mostly assumed that the measurement equation is linear, even if the process dynamics are nonlinear. Hence, a new nonlinear observer design method for a Wiener system composed of a linear process model together with a nonlinear measurement equation is developed in this paper. First, the design of a two degree-of-freedom nonlinear observer is proposed that relies on a Lure system representation of the observer error dynamics. To improve the performance in the presence of parametric uncertainty in the measurement model, the nonlinear observer is augmented to estimate both the state and unknown parameters simultaneously. A rigorous nonlinear observability analysis is also presented to show that a dual sensor configuration is a sufficient and necessary condition for simultaneous state and parameter estimation. Finally, the developed observer design technique is applied to non-intrusive position estimation of the piston inside a pneumatic cylinder. Experimental results show that both position and unknown parameters can be reliably estimated in this application.

An Improved Non-Intrusive Induction Motor Efficiency Estimation Technique for Non-Ideal Supply Voltage Conditions

Methods of accurately determining the efficiency of motors in situ without costly plant downtime are required to implement online motor management strategies that can lead to energy and cost savings. An ideal method must perform in practical power system conditions where there is variation in the supply voltage. An improved non-intrusive compensated slip technique utilising vibration signature analysis has been proposed for accurately estimating induction motor efficiency. The performance of this technique is evaluated through experimental data under ideal and non-ideal supply voltage conditions. The proposed method of estimating motor efficiency is found to be accurate and easy to implement. The proposed technique is robust in under-voltage and over-voltage power supply conditions.

A visible-range computer-vision system for automated, non-intrusive assessment of the pH value in Thomson oranges

Abstract Fruit may be classified for the purposes of usage, packaging and marketing based on the pH (potential of hydrogen) value – a numeric scale used to specify the acidity or basicity of an aqueous solution measured in units of moles per liter of hydrogen ions. In this study, a new approach for the automated and non-intrusive estimation of the pH value of the Thomson navel orange (CRC 969, Citrus sinensis) fruit is presented based on visible-range image processing, image feature extraction and with the use of hybrid imperialist competitive algorithm (ICA)-artificial neural network (ANN) regression. Image features studied include length, width, area, eccentricity, perimeter, blue-value, green-value, red-value, width, contrast, texture, roughness and several ratios thereof. Principal component analysis (PCA) is applied to reduce the number of dimensions without loss of important information and a cubic polynomial function of the mean square error (MSE) versus several factors is computed using the response surface methodology (RSM) approach. Results for pH prediction are given and compared with true measured pH values over the entire 100 Thomson orange dataset, including estimated pH scatter regression plots and estimated pH boxplots. Cross validation is performed over 1000 repeated random trial experiments with uniform random train- and test-sample sets (80% training and 20% disjoint test samples). In addition, we provide numerical results based on the levels achieved by response surface methodology (RSM) evaluated over various error coefficients: the sum square error (SSE), the mean absolute error (MAE), the coefficient of determination (R2), the root mean square error (RMSE), and MSE, resulting in R2 = 0.843 ± 0.043, MSE = 0.046 ± 0.022, MAE = 0.166 ± 0.039, SSE = 0.915 ± 0.425, and RMSE = 0.214 ± 0.146, over the test set. The results demonstrate that such an automated pH-based sorting system with machine vision using the hybrid ICA-ANN algorithm can accurately compute the pH value of Thomson oranges without any contact with the fruit, and which has clear potential applications in the food industry.