Location Prediction Research Articles

Damage visualization and vulnerability assessment of surface ship considering the 3D multihit location of air-explosion threat

Abstract The purpose of this study is to propose a method for rapidly assessing the hit locations and extent of damage to a naval ship attacked by threatening weapons. An integrated method is presented for assessing the vulnerability to the hull and systems caused by the blast or penetrating effects of a weapon at multiple possible impact locations. The proposed method enables the assessment of vulnerability changes for design alternatives during the early design phase of naval vessels. To predict the extent of damage and to visualize the damaged hull and equipment, it is first assumed that a weapon has multiple hit locations. It is shown that a set of possible hit locations was generated by assuming the trajectory for air-explosion (AIREX) weapons by dividing the trajectory into an air-to-ship trajectory and a ship-to-ship trajectory. To account for the non-deterministic nature of weapon hits, a probability distribution approach was used to generate random multiple hit locations for each trajectory while the hit locations of AIREX in all directions were predicted using a multivariate probability distribution that generates three-dimensional random hit points. The extent of damage was then calculated, taking into account the topology of the hull structure and the equipment installed within the hull compartment, along with the damage volume associated with the weapon at each hit point. To identify the damage volume, axis-aligned bounding box (AABB) components were used, which provide a simplified representation of the ship’s geometry, as well as the relative position and dimensions of the hull structure and installed equipment. The damage compartment was defined as the portion of the hull that overlapped the damaged volume. While AABB’s overlap detection algorithm was applied to the damaged hull compartments, the algorithm identified the equipment that overlapped the damaged volume of the hull. Finally, the geometric modeling module, the probabilistic multiple hit location prediction module, and the damage analysis module were developed for damage visualization and vulnerability assessment.

Technical note: Evaluation of the acoustic radiation force imaging for predicting HIFU focus with in vitro and ex vivo experiments.

High-intensity focused ultrasound (HIFU) is currently used for the treatment of various diseases, but it still lacks a reliable technique in the preoperative stage to accurately place its "energy blade" onto diseased targets. Acoustic radiation force imaging (ARFI) was recently introduced to tackle this issue, but its applicability and limitations were not clear. The aim of this study was to evaluate the performance of ARFI method in prediction of HIFU focal location at the preoperative stage. A point spread function (PSF) localization method, which was borrowed from the ultrasound super resolution field, was used to validate the core autocorrelation-based motion estimation algorithm in the ARFI procedure. Accuracy of the ARFI method for estimating the HIFU focus were tested with in vitro and ex vivo experiments with a clinically equivalent HIFU system. Comparisons were made between the estimated focal locations and those of the damaged area after the testing objects were cut open. Results showed that the PSF localization was able to serve as a validating method for motion detection only when the tissue displacement was large. With the ARFI method, location of the HIFU focus could be accurately predicted by a 2D motion map preoperatively, and the axial spatial errors were less than 0.5mm. However, the derived 2D motion maps can only be valuable when the acoustic stimulation in ARFI were strong enough, which was probably due to the fact that the HIFU focal locations were at large depths and the ultrasound imaging signal had low signal to noise ratio. The ARFI method was indeed an accurate technique for preoperatively predicting HIFU focus in vitro and ex vivo. If clinical applications were to be considered, particularly in deep tissues, efforts might need to be made to improve ability of the ultrasound motion estimation technique.

Empirical modeling of optimum tilt angle for flat solar collectors and PV panels.

A new model has been developed to determine the optimal tilt angle for PV panels and solar collectors on a yearly, seasonal, and monthly basis. The model estimates the diffusion component of solar radiation using Orgill and Holland's model, which relates the diffusion fraction of solar radiation to the sky clearness index. Empirical data on the clearness index is used to derive the relationship between the diffusion and direct components of solar radiation at any global latitude on any given day of the year. By maximizing the total amount of diffused and direct radiation, the optimal tilt angle for each month, season, and year is determined relative to the latitude angle. The model is programmed in MATLAB and is freely available for download from the MATLAB file exchange website. The model shows that small deviations from the optimal inclination angle have only a minor effect on overall system yield. The monthly optimal tilt angles predicted by the model are consistent with experimental data and other published model predictions for various locations around the globe. Importantly, unlike some other models, the present model does not predict negative optimal inclination angles for small northern hemisphere latitudes or vice versa.

Hierarchical Wi-Fi Trajectory Embedding for Indoor User Mobility Pattern Analysis

The recent advances in smart building technologies have enabled us to collect massive Wi-Fi network based trajectory data, which provide an unparalleled opportunity for understanding the indoor user mobility pattern and enabling a wide range of business applications. While some previous studies have explored the Wi-Fi positioning of users, there still lacks a systematic and effective solution for indoor user mobility pattern analysis based on Wi-Fi trajectory data. To this end, in this paper, we propose a unified framework for modeling Wi-Fi trajectory data, namely HWTE, which can empower various tasks of indoor user mobility pattern analysis, such as user classification, next location prediction and schedule estimation. Specifically, we first propose a session trajectory construction module to extract the spatio-temporal semantic information from the Wi-Fi trajectories of users. Then, we devise a pre-training module to learn the unified representation of Wi-Fi trajectories. In particular, a session position embedding technique and a position query task is introduced to enhance the representation ability of the whole trajectory. Moreover, we further propose a hierarchical Transformer-based fine-tuning module to support various application tasks with time and space efficiency. Finally, we validate our framework on a real-world dataset with all three kinds of downstream tasks.

Sub-Entity Embedding for inductive spatio-temporal knowledge graph completion

Existing large-scale knowledge graphs generally contain rich spatial and temporal information. Knowledge graph completion has gained wide attention with a massive number of models proposed for inferring missing knowledge. Despite the fact that numerous approaches to knowledge graph completion exist, they often overlook the importance of simultaneously modeling spatial and temporal information, resulting in limited capacity to infer knowledge related to time and location. An intuitive solution to this issue is to use quintuple representation for spatio-temporal facts. To reduce the complexity of learning quintuples, in this paper, we propose sub-entity to tokenize every entity, relation, time stamp and location with a fixed-size vocabulary. Meanwhile, we design Spatio-Temporal Message Passing layer to learn the latent feature vectors of a knowledge graph. We conducted entity link prediction, relation link prediction, time prediction and location prediction experiments. The quantitative results demonstrate the effectiveness of our model in both predicting missing knowledge under both transductive link prediction and inductive link prediction. The visualization results also show that our model can capture meaningful time and location information.

Optimal location and performance prediction of portable air cleaner in composite room shapes using convolutional neural network

Portable air cleaners are commonly used to eliminate hazardous indoor air pollutants such as particulate matters and infectious viruses. The effectiveness of portable air cleaners heavily relies on their installation location, which affects the patterns of airflow carrying the pollutants. Herein, we propose a methodology for optimizing the portable air cleaner placement to maximize its effectiveness using artificial intelligence technology. To build an artificial intelligence model, we obtained the data on airflow patterns for different room shapes and locations of the portable air cleaner through numerical simulations. We utilized floor-plan image data to simplify the handling of complex geometric information related to portable air cleaner placements. The artificial intelligence model extracted geometric features from the floor plan and predicted the effectiveness of the portable air cleaner. The predicted results revealed that the model recommended the location with the highest effectiveness as optimal. The artificial intelligence model achieved a high prediction accuracy of 91.2% and precisely determined the optimal location in rooms with varying shapes. This approach can elucidate the complex relationship between geometric placements and airflow characteristics. Furthermore, the artificial intelligence model can be used as a powerful decision-making tool for indoor air quality management in the building environment.

Differential behavior-related activity of distinct hippocampal interneuron types during odor-associated spatial navigation

Hippocampal pyramidal cells represent an animal's position in space together with specific contexts and events. However, it is largely unknown how distinct types of GABAergic interneurons contribute to such computations. We recorded from the intermediate CA1 hippocampus of head-fixed mice exhibiting odor-to-place memory associations during navigation in a virtual reality (VR). The presence of an odor cue and its prediction of a different reward location induced a remapping of place cell activity in the virtual maze. Based on this, we performed extracellular recording and juxtacellular labeling of identified interneurons during task performance. The activity of parvalbumin (PV)-expressing basket, but not of PV-expressing bistratified cells, reflected the expected contextual change in the working-memory-related sections of the maze. Some interneurons, including identified cholecystokinin-expressing cells, decreased activity during visuospatial navigation and increased activity during reward. Our findings suggest that distinct types of GABAergic interneuron are differentially involved in cognitive processes of the hippocampus.

Continuous monitoring scheduling for moving targets by Earth observation satellites

The studies of earth observation satellite (EOS) scheduling for stationary targets have been increasing rapidly in recent years. However, these studies ignore EOS scheduling for moving targets (EOSSMT), which is urgently needed in many situations. EOSSMT is more complicated due to two factors, 1) location prediction of the moving targets and 2) algorithm design of EOSSMT optimization model for highly-nonlinear characteristics. In this article, we present a novel continuous monitoring scheduling methodology for moving targets by EOSs. Firstly, in order to predict the location of moving targets, a prediction-capture method, which can calculate the capture probability (CP) of the targets, is proposed. Then, based on the concept CP, we regard the EOSSMT as a stochastic integer nonlinear programming problem. Aiming to maximize observation times and duration, two objective functions with low complexity are proposed correspondingly. Finally, to solve our highly-nonlinear optimization model more efficiently, a dispersion-based heuristic (DBH) is proposed. In the computational experiments, a number of instances (scenarios), in which moving targets are successfully observed, verify the correctness and efficiency of our novel methodology for EOSSMT. Computational experiments also show that DBH can achieve better results than the Genetic Algorithm and Greedy Algorithm, with significantly less algorithm complexity.

Utilizing machine learning on freight transportation and logistics applications: A review

This review article explores and locates the current state-of-the-art related to application areas from freight transportation, supply chain and logistics that focuses on arrival time, demand forecasting, industrial processes optimization, traffic flow and location prediction, the vehicle routing problem and anomaly detection on transportation data. This review categorizes the related works according to machine learning methodologies so as to present the methods’ evolution through time, their combinations and their connection with the various applications in the specified fields. Thus, a reader would effortlessly get insights about the current state-of-the-art related to machine learning in freight transportation and related application areas.

Investigating Potential Causes for the Prediction of Spurious Magnetopause Crossings at Geosynchronous Orbit in MHD Simulations

AbstractDuring intense geomagnetic storms, the magnetopause can move in as far as geosynchronous orbit, leaving the satellites in that orbit out in the magnetosheath. Spacecraft operators turn to numerical models to predict the response of the magnetopause to solar wind conditions, but the predictions of the models are not always accurate. This study investigates four storms with a magnetopause crossing by at least one GOES satellite, using four magnetohydrodynamic models at NASA's Community Coordinated Modeling Center to simulate the events, and analyzes the results to investigate the reasons for errors in the predictions. Two main reasons can explain most of the erroneous predictions. First, the solar wind input to the simulations often contains features measured near the L1 point that did not eventually arrive at Earth; incorrect predictions during such periods are due to the solar wind input rather than to the models themselves. Second, while the models do well when the primary driver of magnetopause motion is a variation in the solar wind density, they tend to overpredict or underpredict the integrated Birkeland currents and their effects during times of strong negative interplanetary magnetic field (IMF) Bz, leading to poorer prediction capability. Coupling the MHD codes to a ring current model, when such a coupling is available, generally will improve the predictions but will not always entirely correct them. More work is needed to fully characterize the response of each code under strong southward IMF conditions as it relates to prediction of magnetopause location.

Prediction of brain tumor recurrence location based on multi-modal fusion and nonlinear correlation learning.

Brain tumor is one of the leading causes of cancer death. The high-grade brain tumors are easier to recurrent even after standard treatment. Therefore, developing a method to predict brain tumor recurrence location plays an important role in the treatment planning and it can potentially prolong patient's survival time. There is still little work to deal with this issue. In this paper, we present a deep learning-based brain tumor recurrence location prediction network. Since the dataset is usually small, we propose to use transfer learning to improve the prediction. We first train a multi-modal brain tumor segmentation network on the public dataset BraTS 2021. Then, the pre-trained encoder is transferred to our private dataset for extracting the rich semantic features. Following that, a multi-scale multi-channel feature fusion model and a nonlinear correlation learning module are developed to learn the effective features. The correlation between multi-channel features is modeled by a nonlinear equation. To measure the similarity between the distributions of original features of one modality and the estimated correlated features of another modality, we propose to use Kullback-Leibler divergence. Based on this divergence, a correlation loss function is designed to maximize the similarity between the two feature distributions. Finally, two decoders are constructed to jointly segment the present brain tumor and predict its future tumor recurrence location. To the best of our knowledge, this is the first work that can segment the present tumor and at the same time predict future tumor recurrence location, making the treatment planning more efficient and precise. The experimental results demonstrated the effectiveness of our proposed method to predict the brain tumor recurrence location from the limited dataset.

Retrospective Prediction of Location and Intensity for Two Large Crustal Earthquakes in Iran and India

Retrospective Prediction of Location and Intensity for Two Large Crustal Earthquakes in Iran and India

Environmentally-Aware and Energy-Efficient Multi-Drone Coordination and Networking for Disaster Response

In a disaster response management (DRM) scenario, communication and coordination are limited, and absence of related infrastructure hinders situational awareness. Unmanned aerial vehicles (UAVs) or drones provide new capabilities for DRM to address these barriers. However, there is a dearth of works that address multiple heterogeneous drones collaboratively working together to form a flying ad-hoc network (FANET) with air-to-air and air-to-ground links that are impacted by: (i) environmental obstacles, (ii) wind, and (iii) limited battery capacities. In this paper, we present a novel environmentally-aware and energy-efficient multi-drone coordination and networking scheme that features a Reinforcement Learning (RL) based location prediction algorithm coupled with a packet forwarding algorithm for drone-to-ground network establishment. We specifically present two novel drone location-based solutions (i.e., heuristic greedy, and learning-based) in our packet forwarding approach to support application requirements. These requirements involve improving connectivity (i.e., optimize packet delivery ratio and end-to-end delay) despite environmental obstacles, and improving efficiency (i.e., by lower energy use and time consumption) despite energy constraints. We evaluate our scheme with state-of-the-art networking algorithms in a trace-based DRM FANET simulation testbed featuring rural and metropolitan areas. Results show that our strategy overcomes obstacles and can achieve 81-to-90% of network connectivity performance observed under no obstacle conditions. In the presence of obstacles, our scheme improves the network connectivity performance by 14-to-38% while also providing 23-to-54% of energy savings in rural areas; the same in metropolitan areas achieved an average of 25% gain when compared with baseline obstacle awareness approaches with 15-to-76% of energy savings.

Lightweight APIT with Bat Optimization with Simulated Annealing Localization for Resource-Constrained Sensor Networks

In a wireless sensor network, information processing, and information acquisition, localization technology is the key to making it practically possible application. Approximate Point-in-Triangulation (APIT) is the most widely used localization estimation which has high accuracy in localizing the nodes and ease of deployment of nodes in the real-time environment. Though it has numerous key advantages, some of the drawbacks which make it a little setback in preference are the unevenness in the distribution of nodes. Tracking is more appropriate for mobile sensor nodes than tracking is for static sensor nodes. The two main types of localization algorithms are range-based and range-free techniques. In an indoor setting, the projected range (distance) between an anchor and an unknown node is very inaccurate. By utilizing a large number of already existing access points (APs) in the range-free localization approach, this issue can be overcome to a great extent. The utilization of multisensor data, such as magnetic, inertial, compass, gyroscope, ultrasonic, infrared, visual, and/or odometer, is stressed in recent research to further increase localization accuracy. The tracking system also makes location predictions for the future based on historical location data. To overcome this issue, the proposed localization algorithm of APIT with Bat-SA proves its efficiency. Due to its low localization error, the traditional Bat method is more accurate than APIT. The proposed Bat using the SA algorithm is found to perform better than the traditional APIT algorithm in terms of convergence of computing rate and success rate. In order to mimic the suggested APIT method, it is paired with the Bat-SA localization technique. Simulation evaluation proves the performance efficiency of the proposed algorithm. The performance metrics parameters are latency, node distribution map, positioning error map, and neighbor relationship diagram which are used to evaluate the proposed method.

Identification and Prediction of Fresh Gasoline Locations and Branding Using Newly Targeted Compound Chromatograms with Chemometrics and Machine Learning

Identification and Prediction of Fresh Gasoline Locations and Branding Using Newly Targeted Compound Chromatograms with Chemometrics and Machine Learning

Metal3D: a general deep learning framework for accurate metal ion location prediction in proteins

Metal ions are essential cofactors for many proteins and play a crucial role in many applications such as enzyme design or design of protein-protein interactions because they are biologically abundant, tether to the protein using strong interactions, and have favorable catalytic properties. Computational design of metalloproteins is however hampered by the complex electronic structure of many biologically relevant metals such as zinc . In this work, we develop two tools - Metal3D (based on 3D convolutional neural networks) and Metal1D (solely based on geometric criteria) to improve the location prediction of zinc ions in protein structures. Comparison with other currently available tools shows that Metal3D is the most accurate zinc ion location predictor to date with predictions within 0.70 ± 0.64 Å of experimental locations. Metal3D outputs a confidence metric for each predicted site and works on proteins with few homologes in the protein data bank. Metal3D predicts a global zinc density that can be used for annotation of computationally predicted structures and a per residue zinc density that can be used in protein design workflows. Currently trained on zinc, the framework of Metal3D is readily extensible to other metals by modifying the training data.

Predicting fretting-fatigue endurance of rotating bending shrink-fitted assemblies using a sequential RUIZ-SWT approach: The effect of entrapped debris layer

Fretting-fatigue damage often appears in shaft shrink-fitted assemblies submitted to rotating bending. This paper focuses on the influence of the sleeve edge geometry on such damage. Rotating bending tests have been performed and broken and unbroken specimens have been expertized confirming fretting-fatigue phenomena. A 3D finite element method model was developed to evaluate the sliding and opening lengths. A sequential tribological - fatigue stress analysis strategy allowed the prediction of the maximum crack location ( Ruiz criterion) then the estimation of the fatigue endurance of the assembly using the SWT criterion. The last step of this σSWTXΓmaxapproach considers the presence of a debris layer within the contact to improve the consistency of the service life prediction.

Damage Detection and Assessment of Stiffness and Mass Matrices in Curved Simply Supported Beam Using Genetic Algorithm

In this study, a genetic algorithm (GA) is used to detect damage in curved beam model, stiffness as well as mass matrices of the curved beam elements is formulated using Hamilton's principle. Each node of the curved beam element possesses seven degrees of freedom including the warping degree of freedom. The curved beam element had been derived based on the Kang and Yoo’s thin-walled curved beam theory. The identification of damage is formulated as an optimization problem, binary and continuous genetic algorithms(BGA, CGA) are used to detect and locate the damage using two objective functions (change in natural frequencies, Modal Assurance Criterion MAC). The results show the objective function based on change in natural frequency is the best objective and no error was recorded in prediction of location and small error in detecting damage value. Also the result show that the genetic algorithm method are efficient indicating and quantifying single and multiple damage with high precision, and the prediction error for the CGA are less than corresponding value for the BGA.

Hierarchical Regression and Classification for Accurate Object Detection.

Accurate object detection requires correct classification and high-quality localization. Currently, most of the single shot detectors (SSDs) conduct simultaneous classification and regression using a fully convolutional network. Despite high efficiency, this structure has some inappropriate designs for accurate object detection. The first one is the mismatch of bounding box classification, where the classification results of the default bounding boxes are improperly treated as the results of the regressed bounding boxes during the inference. The second one is that only one-time regression is not good enough for high-quality object localization. To solve the problem of classification mismatch, we propose a novel reg-offset-cls (ROC) module including three hierarchical steps: the regression of the default bounding box, the prediction of new feature sampling locations, and the classification of the regressed bounding box with more accurate features. For high-quality localization, we stack two ROC modules together. The input of the second ROC module is the output of the first ROC module. In addition, we inject a feature enhanced (FE) module between two stacked ROC modules to extract more contextual information. The experiments on three different datasets (i.e., MS COCO, PASCAL VOC, and UAVDT) are performed to demonstrate the effectiveness and superiority of our method. Without any bells or whistles, our proposed method outperforms state-of-the-art one-stage methods at a real-time speed. The source code is available at https://github.com/JialeCao001/HSD.

Numerical simulation of squeeze casting of aluminum alloy flywheel housing with large wall thickness difference and complex shape

The squeeze casting process of ZL104 aluminum alloy flywheel housing with large wall thickness difference and complex shape was simulated by ProCAST software. The results show that the filling process was stable and could be divided into four stages: connection of channels, filling of horizontal zone, vertical direction, and difficult-filling zone. There were six characteristic zones with solidification lag, where shrinkage cavity and shrinkage porosity were obvious. The location prediction of defects was accurate. Through range analysis of defects volume, the optimal combination of process parameters was determined as pouring temperature of 650 °C, specific pressure of 48 MPa, mold temperature of 220 °C, local specific pressure of 800 MPa, and pressure delay time of 10 s (Side A) and 12 s (Side B). The maximum stress occurred in the thin-walled structure with fast solidification and large curvature. The simulation results were verified by the actual process.