Set Of Data Points Research Articles

Mars Ascent Trajectory Optimization Based on Convex Optimization

In this paper, a Mars ascent trajectory optimization algorithm based on convex optimization is designed for the Mars ascent trajectory optimization problem. The main accomplishment of this paper is the trajectory optimization of the second-stage ignition based on the convex optimization. For the first-stage ignition, open-loop guidance is used. In this stage, the ascender flies using the maximum thrust to track the attitude profile, thus completing the ascent flight control. In the presence of errors in the first-stage ignition, the experiment generates multiple sets of first-stage shutdown point data and completes simulation verification experiments based on them. The simulation verifies that the second-stage ignition trajectory optimization based on convex optimization algorithm can achieve high accuracy into orbit despite the large starting point deviation.

Model-free Data-Driven inference in computational mechanics

We extend the model-free Data-Driven computing paradigm to solids and structures that are stochastic due to intrinsic randomness in the material behavior. The behavior of such materials is characterized by a likelihood measure instead of a constitutive relation. We specifically assume that the material likelihood measure is known only through an empirical point-data set in material or phase space. The state of the solid or structure is additionally subject to compatibility and equilibrium constraints. The problem is then to infer the likelihood of a given structural outcome of interest. In this work, we present a Data-Driven method of inference that determines likelihoods of outcomes from the empirical material data and that requires no material or prior modeling. In particular, the computation of expectations is reduced to explicit sums over local material data sets and to quadratures over admissible states, i.e., states satisfying compatibility and equilibrium. The complexity of the material data-set sums is linear in the number of data points and in the number of members in the structure. Efficient population annealing procedures and fast search algorithms for accelerating the calculations are presented. The scope, cost and convergence properties of the method are assessed with the aid selected applications and benchmark tests.

Hyperpower least squares progressive iterative approximation

Geometric iterative methods (GIMs) generate curves or surfaces that fit (interpolate or approximate) given sets of data points. Standard GIMs use the Richardson iteration or gradient descent method, so they converge relatively slowly. This paper investigates the GIMs acceleration method, which is based on the use of more efficient methods for the matrix inverse approximation. That is, the possibility to use hyperpower iterative methods is being explored. As a result, a novel GIM named hyperpower least squares progressive iterative approximation (HPLSPIA) is proposed. The convergence of the HPLSPIA method is analyzed, and the computational complexity is briefly discussed. The method is then applied to tensor product B-spline surface fitting. Several HPLSPIA methods utilizing different hyperpower methods for the matrix inverse approximation are experimentally compared. It is shown that polynomial factorizations used in the hyperpower iterative methods significantly affect the efficiency of the HPLSPIA methods.

Data-driven strain–stress modelling of granular materials via temporal convolution neural network

Machine learning offers a new approach to predicting the path-dependent stress–strain response of granular materials. Recent studies show that temporal convolution neural (TCN) networks, a mutation of the 1D convolution neural network (CNN), have a powerful capability of addressing time-related prediction tasks. In this work, TCN networks are constructed to explore their potential in capturing the constitutive law of granular materials. To train and test the TCN network, three types of numerical experiments are implemented to generate datasets via discrete element modelling. The Bayesian optimisation method is employed to find the optimum architecture of the network. Furthermore, to improve the training accuracy and efficiency, a transfer learning (TL) scheme is innovatively leveraged, which utilises the trained network parameters from a set of shorter time steps and/or coarser data points of the training strain–stress loading curves, as the initial values, to train the network for a longer time step. The prediction ability of the trained TCN network is assessed and compared with a recurrent neural network which has been proved to perform well in predicting constitutive laws of the granular materials. In addition, training datasets with artificially added noise are also used to test and analyse the robustness of TCN networks.

The Effect Of Urban Green Spaces In Reducing Urban Flooding In Lahore, Pakistan, Using Geospatial Techniques

Urban Green Spaces (UGS) curtails all environmental issues and ensure an eco-friendly locale. Similarly, the emergence of UGS is very helpful to cope with emerging urban flooding in cities by setting up the world standard of green space ratio (20 to 25 percent of the area) and green per capita (9m2 ) in a geographical area. Therefore, the present study is conducted to evaluate the causal effect relation of UGS with the frequency of urban flooding. For this purpose, 69 selected union councils are taken as a study area in District Lahore, Pakistan. The relation between UGS and the occurrence of floods is evaluated using geo-statistical and geospatial analysis techniques during the monsoon rainfalls from 2013 to 2019. Furthermore, the data sets of sore points (inundated areas), occurrences of urban flooding (number of event occurrences), green per capita, and green ratio are used. Results revealed that selected union councils in Lahore don’t have enough urban green spaces. There is only a 51 sq km area with adequate UGS that accounts for only 18 percent of the study area. The rest of the area does not meet the world standards of green area. There are some areas including Ravi town, Gulberg town, and Samanabad town with green per capita more than 4 green per capita. On the other hand, there are only 02 union councils including Race Course and Model Town that are comprised of a 20 percent green area. The findings of the study will be helpful for proper urban planning and strategies i.e. with greener structures.

Meta-Active Learning in Probabilistically Safe Optimization

When a robotic system is faced with uncertainty, the system must take calculated risks to gain information as efficiently as possible while ensuring system safety. The need to safely and efficiently gain information in the face of uncertainty spans domains from healthcare to search and rescue. To efficiently learn when data is scarce or difficult to label, active learning acquisition functions intelligently select a data point that, if the label were known, would most improve the estimate of the unknown model. Unfortunately, prior work in active learning suffers from an inability to accurately quantify information-gain, generalize to new domains, and ensure safe operation. To overcome these limitations, we develop Safe MetAL, a probabilistically-safe, active learning algorithm which meta-learns an acquisition function for selecting sample efficient data points in safety critical domains. The key to our approach is a novel integration of meta-active learning and chance-constrained optimization. We (1) meta-learn an acquisition function based on sample history, (2) encode this acquisition function in a chance-constrained optimization framework, and (3) solve for an information-rich set of data points while enforcing probabilistic safety guarantees. We present state-of-the-art results in active learning of the model of a damaged UAV and in learning the optimal parameters for deep brain stimulation. Our approach achieves a 41% improvement in learning the optimal model and a 20% speedup in computation time compared to active and meta-learning approaches while ensuring safety of the system.

PointFace: Point Cloud Encoder-Based Feature Embedding for 3-D Face Recognition

The accuracy of 2D face recognition (FR) has progressed significantly due to the availability of large-scale training data. However, the research of deep learning based 3D FR is still in the early stage. Most of available 3D FR generate 2D maps from 3D data and apply existing 2D CNNs to the generated 2D maps for feature extraction. We propose in this paper a light-weight framework, named PointFace, to directly process point set data for 3D FR. In this framework, two weight-shared encoders are designed to directly extract features from a pair of 3D faces and the distances between embeddings of the same person and different person are minimized and maximized, respectively. The framework also use a feature similarity loss to guide the encoders to obtain discriminative face representations. A pair selection strategy is proposed to generate positive and negative face pairs to further improve the FR performance. Extensive experiments on Lock3DFace and Bosphorus show that the proposed PointFace outperforms state-of-the-art 2D CNN based FR methods.

A GIS Pipeline to Produce GeoAI Datasets from Drone Overhead Imagery

Drone imagery is becoming the main source of overhead information to support decisions in many different fields, especially with deep learning integration. Datasets to train object detection and semantic segmentation models to solve geospatial data analysis are called GeoAI datasets. They are composed of images and corresponding labels represented by full-size masks typically obtained by manual digitizing. GIS software is made of a set of tools that can be used to automate tasks using geo-referenced raster and vector layers. This work describes a workflow using GIS tools to produce GeoAI datasets. In particular, it mentions the steps to obtain ground truth data from OSM and use methods for geometric and spectral augmentation and the data fusion of drone imagery. A method semi-automatically produces masks for point and line objects, calculating an optimum buffer distance. Tessellation into chips, pairing and imbalance checking is performed over the image–mask pairs. Dataset splitting into train–validation–test data is done randomly. All of the code for the different methods are provided in the paper, as well as point and road datasets produced as examples of point and line geometries, and the original drone orthomosaic images produced during the research. Semantic segmentation results performed over the point and line datasets using a classical U-Net show that the semi-automatically produced masks, called primitive masks, obtained a higher mIoU compared to other equal-size masks, and almost the same mIoU metric compared to full-size manual masks.

Visualization and Analysis of Safe Routes to School based on Risk Index using Student Survey Data for Safe Mobility

Risk analysis is important in heterogeneous industrial domains to enable sustainable development. Data is the basis for emphasizing the potential risk elements for improving efficiency, quality, and safety. For supplying safe routes to schools based on risk analysis, the risk assessment of routes is one of the widely used and very effective methodologies to filter the most dangerous roads, intersections, or specific points on roads. This paper presents a visualization and analysis of the risk assessment approach based on the risk index model using geographical information, including routes, danger points, and student survey data. The proposed risk index model is used for deriving a risk index based on geographical information, including danger points and a route's path. The model includes an equation to calculate the distance of danger points to the path using the coordinates of each location. The survey data is mainly comprised of route and survey information that is analyzed and preprocessed for the input data of the risk index model. The survey mainly consists of basic information on the route, survey participants, school route information, and school route coordinates. The data is classified into the school route data set and the school route danger points data set, and these values are applied to the analysis and the risk index model. Also, the risk index model is designed and developed through the analysis of routes.

Comparison of Ground Point Filtering Algorithms for High-Density Point Clouds Collected by Terrestrial LiDAR

Terrestrial LiDAR (light detection and ranging) has been used to quantify micro-topographic changes using high-density 3D point clouds in which extracting the ground surface is susceptible to off-terrain (OT) points. Various filtering algorithms are available in classifying ground and OT points, but additional research is needed to choose and implement a suitable algorithm for a given surface. This paper assesses the performance of three filtering algorithms in classifying terrestrial LiDAR point clouds: a cloth simulation filter (CSF), a modified slope-based filter (MSBF), and a random forest (RF) classifier, based on a typical use-case in quantifying soil erosion and surface denudation. A hillslope plot was scanned before and after removing vegetation to generate a test dataset of ground and OT points. Each algorithm was then tested against this dataset with various parameters/settings to obtain the highest performance. CSF produced the best classification with a Kappa value of 0.86, but its performance is highly influenced by the ‘time-step’ parameter. MSBF had the highest precision of 0.94 for ground point classification but the highest Kappa value of only 0.62. RF produced balanced classifications with the highest Kappa value of 0.75. This work provides valuable information in optimizing the parameters of the filtering algorithms to improve their performance in detecting micro-topographic changes.

An end-to-end lightweight model for grape and picking point simultaneous detection

Grape clusters and their picking point detection (GCPPD) are pivotal in the visual tasks of automatic grape harvesting. In recent years, much progress has been made in increasing the accuracy of GCPPD based on deep learning models. However, GCPPD still has many problems. First, it is inevitable that grape cluster detection requires complex models with many parameters. Second, the prior work on picking point detection can be summarised as the image processing methods using predefined hand-crafted features. This leads to a lack of robustness in the proposed algorithms. To address this, a scheme for the simultaneous detection of grape clusters and their picking points is explored. Due to the superiority of simultaneous detection, the model is constructed as an end-to-end network. Thus, a lightweight end-to-end model called YOLO-GP (YOLO-Grape and Picking points) is proposed. Specifically, YOLO-GP utilises a ghost bottleneck to reduce model parameters. Additionally, this model adds the prediction of picking points using the novel idea, that the picking point follows the bounding box . The Grape-PP (Grape-Picking Point) dataset for model training is constructed, which contains 360 grape images with 4517 grape cluster bounding boxes and picking points. The experiments show that the mean Average Precision (mAP) of grape cluster detection by YOLO-GP is 93.27% with a decrease in the number of weight parameters by at least 10%. The distance error of picking point detection is less than 40 pixels. In summary, YOLO-GP achieves the simultaneous detection of grape clusters and their picking points, and its performance is comparable to that of baseline models. • An idea of the grape cluster and its picking point simultaneous detection is useful. • YOLO-GP can simultaneously detect the grape clusters and their picking points. • Ghost module and heads of picking points prediction added to YOLO-GP. • YOLO-GP is a lightweight end-to-end model. • Grape-PP Dataset can be used to train picking points detection model.

Fitting and filling of 3D datasets with volume constraints using radial basis functions under tension

Given a dataset of 3D points in which there is a hole, i.e., a region with a lack of information, we develop a method providing a surface that fits the dataset and fills the hole. The filling patch is required to fulfill a prescribed volume condition. The fitting–filling function consists of a radial basis functions that minimizes an energy functional involving both, the fitting of the dataset and the volume constraint of the filling patch, as well as the fairness of the function. We give a convergence result and we present some graphical and numerical examples.

A training strategy for hybrid models to break the curse of dimensionality.

Mechanistic/data-driven hybrid modeling is a key approach when the mechanistic details of the processes at hand are not sufficiently well understood, but also inferring a model purely from data is too complex. By the integration of first principles into a data-driven approach, hybrid modeling promises a feasible data demand alongside extrapolation. In this work, we introduce a learning strategy for tree-structured hybrid models to perform a binary classification task. Given a set of binary labeled data, the challenge is to use them to develop a model that accurately assesses labels of new unlabeled data. Our strategy employs graph-theoretic methods to analyze the data and deduce a function that maps input features to output labels. Our focus here is on data sets represented by binary features in which the label assessment of unlabeled data points is always extrapolation. Our strategy shows the existence of small sets of data points within given binary data for which knowing the labels allows for extrapolation to the entire valid input space. An implementation of our strategy yields a notable reduction of training-data demand in a binary classification task compared with different supervised machine learning algorithms. As an application, we have fitted a tree-structured hybrid model to the vital status of a cohort of COVID-19 patients requiring intensive-care unit treatment and mechanical ventilation. Our learning strategy yields the existence of patient cohorts for whom knowing the vital status enables extrapolation to the entire valid input space of the developed hybrid model.

Wi-Learner

Contactless RF-based sensing techniques are emerging as a viable means for building gesture recognition systems. While promising, existing RF-based gesture solutions have poor generalization ability when targeting new users, environments or device deployment. They also often require multiple pairs of transceivers and a large number of training samples for each target domain. These limitations either lead to poor cross-domain performance or incur a huge labor cost, hindering their practical adoption. This paper introduces Wi-Learner, a novel RF-based sensing solution that relies on just one pair of transceivers but can deliver accurate cross-domain gesture recognition using just one data sample per gesture for a target user, environment or device setup. Wi-Learner achieves this by first capturing the gesture-induced Doppler frequency shift (DFS) from noisy measurements using carefully designed signal processing schemes. It then employs a convolution neural network-based autoencoder to extract the low-dimensional features to be fed into a downstream model for gesture recognition. Wi-Learner introduces a novel meta-learner to "teach" the neural network to learn effectively from a small set of data points, allowing the base model to quickly adapt to a new domain using just one training sample. By so doing, we reduce the overhead of training data collection and allow a sensing system to adapt to the change of the deployed environment. We evaluate Wi-Learner by applying it to gesture recognition using the Widar 3.0 dataset. Extensive experiments demonstrate Wi-Learner is highly efficient and has a good generalization ability, by delivering an accuracy of 93.2% and 74.2% - 94.9% for in-domain and cross-domain using just one sample per gesture, respectively.

Using Structure Location Data to Map the Wildland–Urban Interface in Montana, USA

The increasing wildfire activity and rapid population growth in the wildland–urban interface (WUI) have made more Americans exposed to wildfire risk. WUI mapping plays a significant role in wildfire management. This study used the Microsoft building footprint (MBF) and the Montana address/structure framework datasets to map the WUI in Montana. A systematic comparison of the following three types of WUI was performed: the WUI maps derived from the Montana address/structure framework dataset (WUI-P), the WUI maps derived from the MBF dataset (WUI-S), and the Radeloff WUI map derived from census data (WUI-Z). The results show that WUI-S and WUI-P are greater than WUI-Z in the WUI area. Moreover, WUI-S has more WUI area than WUI-P due to the inclusion of all structures rather than just address points. Spatial analysis revealed clusters of high percentage WUI area in western Montana and low percentage WUI area in eastern Montana, which is likely related to a combination of factors including topography and population density. A web GIS application was also developed to facilitate the dissemination of the resulting WUI maps and allow visual comparison between the three WUI types. This study demonstrated that the MBF can be a useful resource for mapping the WUI and could be used in place of a national address point dataset.

Geometric prior of multi-resolution yielding manifolds and the local closest point projection for nearly non-smooth plasticity

Elastoplasticity models often introduce a scalar-valued yield function to implicitly represent the boundary between elastic and plastic material states. This paper introduces a new alternative where the yield envelope is represented by a manifold of which the topology and the geometry are learned from a set of data points in a parametric space (e.g. principal stress space, π-plane). Here, deep geometric learning enables us to reconstruct a highly complex yield envelope by breaking it down into multiple coordinate charts. The global atlas that consists of these coordinate charts in return allows us to represent the yield surface via multiple overlapping patches, each with a specific local parametrization. This setup provides several advantages over the classical implicit function representation approach. For instance, the availability of coordinate charts enables us to introduce an alternative stress integration algorithm where the trial stress may project directly on a local patch and hence circumvent the issues related to non-smoothness and the lack of convexity of yield surfaces. Meanwhile, the local parametric approach also enables us to predict hardening/softening locally in the parametric space, even without complete knowledge of the yield surface. Comparisons between the classical yield function approach on the non-smooth plasticity and anisotropic cam-clay plasticity model are provided to demonstrate the capacity of the models for highly precise yield surface and the feasibility of the implementation of the learned model in the local stress integration algorithm.

A novel neural network and grey correlation analysis method for computation of the heat transfer limit of a loop heat pipe (LHP)

A loop heat pipe (LHP) has the advantages of larger heat transfer capacity and anti-gravity operational performance. The current prediction models for LHP heat transfer capacity have the difficulties in popularization of data volume and determination of accurate parametrical data, leading to the uncertain and varying outcomes that are inconsistent and away from reality. To address these challenges, this paper developed a first-of-its-kind big-data-driven LHP heat transfer limit prediction model by employing the neural network and grey correlation analysis method, which have advantages of high precision and large data volume. A double-layer feedforward neural network with sigmoid hidden neuron and linear output neuron was constructed to predict the heat transfer limit of the LHP. The grey scale analysis is applied to select the variables with correlation coefficient greater than 0.5, thus giving the clear identification of the both input parameters (e.g. refrigerant temperature, filling liquid quantity, height difference between evaporator and condenser, and number of heat pipe array) and output ones (heat transfer limit). The previously validated LHP heat transfer limit calculation model is used to calculate the heat transfer limit corresponding to the selected parameters, thus formulating 1,010,038 sets of data points. Of those calculated datasets, 707,026 (70% of data) are treated as a training set, 151,506 (15% of data) as a verification set, and 151,506 groups of data (15% of data) as the test sets for training. After several optimization and debugging, the number of hidden layer neurons is determined to be 100. The correlation coefficient (R), mean square error (MSE) and mean relative error (MRE) are 0.9997, 52.7 and 0.32% respectively, all of which are within reasonable accuracy range. The results show that the model has good prediction accuracy and consistence and is an effective tool to characterize and optimize the LHP in various application synergies.

Prediction surface tension of ionic liquid–water mixtures using a hybrid group contribution and artificial neural network method

In this work, a database was established to collect relevant surface tension data (4578 sets of data points in total) for 138 ionic liquids (ILs)-H2O hybrid systems. An ANN-GC model for predicting the surface tension of IL-H2O hybrid systems is proposed to be constructed by combining the group contribution (GC) method and artificial neural network (ANN) algorithm. This model correlates surface tension to temperature together with IL structure and is developed by using machine learning algorithms. The results show that the model with 4, 5 and 7 neurons in the hidden layer is able to provide reliable predictions for the surface tension of the IL-H2O hybrid system. When 4, 5 and 7 neurons are used in the hidden layer of the model, the squared correlation coefficients (R2) of the training set are 0.92, 0.93 and 0.93; the MAE are 0.0021, 0.0020 and 0.0020, respectively. While the squared correlation coefficients (R2) of the test set are 0.94, 0.91,0.91 and the MAE are 0.0023, 0.0029, 0.0027, respectively. Meanwhile, the ANN-GC model of the hybrid system was extended and an ANN-GC model of pure ILs was constructed based on 172 surface tension data of pure ionic liquids. The comparison shows that the model with 4 and 5 neurons in the hidden layer can be extended to the pure IL system to provide reliable prediction of the surface tension for the pure IL system.

A novel information processing method based on an ensemble of Auto-Encoders for unsupervised fault detection

The current research paper proposes a novel information processing Deep Learning framework for unsupervised fault detection applications, where during the training process only samples from the normal class are available. Recently, unsupervised fault detection methods based on the Auto-Encoders have been successful and are used extensively. They are supported by the assumption that abnormal unknown samples that don’t belong to the learned manifold of the training dataset of normal points produce higher reconstruction cost than the normal samples. The presented scheme is based on an ensemble of different types of Auto-Encoders; each of them is trained independently for the one-class fault detection task. The final agreement of the normality of the testing sample is made from a soft voting process, where the confidence of each Auto-Encoder about its decision is considered. The significance of each individual Auto-Encoder in the final decision is extracted from a statistical analysis of the independent training process of each one. Simulation results with three widely used fault detection datasets show the effectiveness of the proposed model.

Longitudinal Changes in Resting State FMRI Spectra in Children.

Longitudinal studies can provide more precise measure of brain development, as they focus on within-subject variability, as opposed to cross-sectional studies. In this study, we track longitudinal changes in resting state fMRI data using spectrum of time-courses generated via group independent component analysis (gICA), in a multi time point dataset containing healthy children 8-18 years old, collected on both eyes open and eyes closed resting state conditions. Clinical Relevance - Tracking normal brain development and identifying biomarkers of healthy brain development are critically important to diagnose mental disorders at early ages. We found increased spectral power in low frequencies and decreased spectral power in high frequencies in children with typical development in both the eyes open and eyes closed conditions though the eyes closed condition showed greater changes with development mostly in the visual networks. Results are also replicated on an independent dataset.