Robust Identification Research Articles

Feasibility of Advanced Reflective Cracking Prediction and Detection for Pavement Management Systems Using Machine Learning and Image Detection

This research manuscript presents a comprehensive investigation into the prediction and detection of reflective cracking in pavement infrastructure through a combination of machine learning approaches and advanced image detection techniques. Leveraging machine learning algorithms, reflective cracking prediction models were developed and optimized for accuracy and efficiency. Additionally, the efficacy of image detection methods, particularly utilizing Mask R-CNN, was explored for robust and precise identification of reflective cracking on pavement surfaces. The study not only aims to enhance the predictive capabilities of pavement management systems (PMSs) through machine learning-based models but also seeks to integrate advanced image detection technologies to support real-time monitoring and assessment of pavement conditions. By providing accurate and timely detection of reflective cracking, these methodologies contribute to the optimization of pavement maintenance strategies and the overall improvement of pavement infrastructure management practices. Results indicate that the developed machine learning models achieve an average predictive accuracy of over 85%, with some models achieving accuracies exceeding 90%. Moreover, the utilization of a mask region-based convolutional neural network (Mask R-CNN) for image detection demonstrates exceptional precision, with a detection accuracy of over 95% on average across different pavement types and weather conditions. The results demonstrate the promising performance of the developed machine learning models in predicting reflective cracking, while the utilization of Mask R-CNN showcases exceptional accuracy in the detection of reflective cracking from images. This research underscores the importance of leveraging cutting-edge technologies to address challenges in pavement infrastructure management, ultimately supporting the sustainability and longevity of transportation networks.

Improved individual identification in DNA mixtures of unrelated or related contributors through massively parallel sequencing

DNA mixtures are a common sample type in forensic genetics, and we typically assume that contributors to the mixture are unrelated when calculating the likelihood ratio (LR). However, scenarios involving mixtures with related contributors, such as in family murder or incest cases, can also be encountered. Compared to the mixtures with unrelated contributors, the kinship within the mixture would bring additional challenges for the inference of the number of contributors (NOC) and the construction of probabilistic genotyping models. To evaluate the influence of potential kinship on the individual identification of the person of interest (POI), we conducted simulations of two-person (2 P) and three-person (3 P) DNA mixtures containing unrelated or related contributors (parent-child, full-sibling, and uncle-nephew) at different mixing ratios (for 2 P: 1:1, 4:1, 9:1, and 19:1; for 3 P: 1:1:1, 2:1:1, 5:4:1, and 10:5:1), and performed massively parallel sequencing (MPS) using MGIEasy Signature Identification Library Prep Kit on MGI platform. In addition, in silico simulations of mixtures with unrelated and related contributors were also performed. In this study, we evaluated 1): the MPS performance; 2) the influence of multiple genetic markers on determining the presence of related contributors and inferring the NOC within the mixture; 3) the probability distribution of MAC (maximum allele count) and TAC (total allele count) based on in silico mixture profiles; 4) trends in LR values with and without considering kinship in mixtures with related and unrelated contributors; 5) trends in LR values with length- and sequence-based STR genotypes. Results indicated that multiple numbers and types of genetic markers positively influenced kinship and NOC inference in a mixture. The LR values of POI were strongly dependent on the mixing ratio. Non- and correct-kinship hypotheses essentially did not affect the individual identification of the major POI; the correct kinship hypothesis yielded more conservative LR values; the incorrect kinship hypothesis did not necessarily lead to the failure of POI individual identification. However, it is noteworthy that these considerations could lead to uncertain outcomes in the identification of minor contributors. Compared to length-based STR genotyping, using sequence-based STR genotype increases the individual identification power of the POI, concurrently improving the accuracy of mixing ratio inference using EuroForMix. In conclusion, the MGIEasy Signature Identification Library Prep kit demonstrated robust individual identification power, which is a viable MPS panel for forensic DNA mixture interpretations, whether involving unrelated or related contributors.

Shiba: A unified computational method for robust identification of differential RNA splicing across platforms.

Alternative pre-mRNA splicing (AS) is a fundamental regulatory process that generates transcript diversity and cell type variation. We developed Shiba, a robust method integrating transcript assembly, splicing event identification, read counting, and statistical analysis, to efficiently quantify exon splicing levels across various types of RNA-seq datasets. Compared to existing pipelines, Shiba excels in capturing both annotated and unannotated or cryptic differential splicing events with superior accuracy, sensitivity, and reproducibility. Furthermore, Shiba's unique consideration of junction read imbalance and exon-body read coverage reduces false positives, essential for downstream functional analyses. We have further developed scShiba for single-cell/nucleus (sc/sn) RNA-seq data, enabling the exploration of splicing variations in heterogeneous cell populations. Both simulated and real data demonstrate Shiba's robustness across multiple sample sizes, including n=1 datasets and individual cell clusters from scRNA-seq. Application of Shiba on single replicates of RNA-seq identified new AS-NMD targets, and scShiba on snRNA-seq revealed intricate temporal AS regulation in dopaminergic neurons. Both Shiba and scShiba are provided in Docker/Singularity containers and Snakemake pipeline, enhancing accessibility and reproducibility. The comprehensive capabilities of Shiba and scShiba allow systematic and robust quantification of alternative splicing events, laying a solid foundation for mechanistic exploration of functional complexity in RNA splicing.

Fisher discrimination multiple kernel dictionary learning for robust identification of nonlinear features in machinery health monitoring

In sparse representation-based classification, Fisher discrimination dictionary learning (FDDL) has attracted widespread attention due to its advantages such as fewer parameters and good dictionary discriminability. However, due to its linear nature, it is difficult to handle nonlinear features. Kernel-based nonlinear transformation enables dictionary learning to capture nonlinear features embedded in samples, but traditional single-kernel methods have limited performance. In this paper, a Fisher discrimination multiple kernel dictionary learning (FDMKDL) method is proposed, in which Fisher discriminative criterion is imposed on both high-dimensional samples and coding coefficients. Specifically, we derive the kernel version of FDDL, i.e., Fisher discrimination kernel dictionary learning (FDKDL) to learn nonlinear features and promote dictionary discriminability. Meanwhile, to avoid the problem of poor adaptability and possible manual selection caused by single-kernel methods, we further derive a flexible multiple kernel learning (MKL) framework, which utilizes the complementary information of multiple kernel functions to adaptively obtain the optimal weights and synthesize a discriminative kernel space. Finally, FDMKDL combines FDKDL with this MKL framework to obtain a more discriminative dictionary. Experimental evaluations conducted on two datasets demonstrate the effectiveness and robustness of the proposed FDMKDL method in learning nonlinear features for machinery health monitoring when compared to several state-of-the-art methods.

An unsupervised H&E-based machine-learning approach for precise prediction of tumor microenvironment subtypes.

e13593 Background: Precision oncology increasingly relies on accurate classification of the tumor microenvironment (TME), which influences therapeutic responses and prognostication in cancer. While histological observation via H&E staining remains the gold standard due to the detailed spatial context and morphological information provided, its lack of reproducibility limits its use in clinics. Next-generation sequencing (NGS) reliably provides valuable insights into the molecular landscape of tumors. Here we debut an unsupervised machine-learning approach that leverages both the rich detail in H&E slides and the reproducibility of NGS to enhance TME classification. Methods: An ML-based image analysis algorithm was developed using a convolutional neural network (CNN) deep learning model. Using over 600 images from the TCGA COAD and READ datasets and from 139 colorectal cancer (CRC) samples in an internal cohort, we trained the CNN model to curate features from small patches of whole slide images (WSIs). Then, clusters were created from these H&E WSIs without prior labeling. The clusters were validated by experienced pathologists to confirm cluster homogeneity and filter artifacts. We trained a classifier to categorize molecular features from the WSIs into TME subtypes. Performance was measured using the area under the receiver operating characteristic curve (AUC-ROC), based on the model's ability to correctly classify the Fibrotic (F) and Immune Desert (D) TME subtypes (1), as well as MSI status identified by whole exome sequencing. The F subtype corresponds with high stromal content, low immune infiltration, and the worst prognosis. The D subtype corresponds with a cold immune TME and an intermediate prognosis. Progression-free survival (PFS) data from TCGA were used to examine prognostic significance. Results: We obtained 0.86, 0.79, 0.72, and 0.71 AUC scores for predicting the ID and F TMEs, MSI status, and KRAS mutation status, respectively. PFS data analysis showed our H&E-based approach to yield similar or better prognostic significance compared to NGS-based TME classification (ID vs others, p=0.05 (H&E) and 0.2 (NGS); F vs others, p=0.07 (H&E) and 0.09 (NGS)). Our findings showed a notable correlation with RNA- and DNA-based predictions of ID and F TMEs and MSI, with enhanced precision and more robust subtype identification while achieving high embedding consistency. Conclusions: By harnessing the inherently rich detail in histological images, our H&E-based approach complements NGS-based TME analysis. Our findings advocate for the integration of advanced image analysis into the standard TME characterization workflow to refine prognostic and therapeutic strategies for CRC. We will expand this algorithm for use with other cancer types to realize its potential as a promising adjunct to NGS for precise TME classification. 1. Bagaev et al., 2021.

Direct elections and trust in state and political institutions: Evidence from Indonesia's election reform

People's trust in state and political institutions is a key foundation of a well-functioning economy. The question of whether direct elections affect trust remains open, especially in developing countries, due to a scarcity of studies with robust identification strategies. We estimate the impact of direct elections on people's trust in state and political institutions, using a major political reform in Indonesia as the source of exogenous variation. Pre-2005, regents, mayors, and governors were elected by the local legislative assembly. Starting in 2005, they must compete in an open election where voters directly choose their preferred leader. The historically and institutionally driven staggered implementation of these local direct elections enables us to identify the causal impact of the reform. We find that direct elections increase trust in all state and political institutions, except for the police and political parties. We find that trust only increased in districts where the elections were held with no or low hostility, implying that trust is strongly influenced by political situations during direct elections. We find that elections did not really improve well-being or satisfaction, implying that the absence of election hostility is a stronger explanation for the higher trust.

Threshing cylinder unbalance detection using a signal extraction method based on parameter-adaptive variational mode decomposition

The threshing cylinder will wear and deform during the threshing process, causing dynamic balance problems. The combine harvester has multiple vibration excitation sources and a complex vibration environment, making it challenging to extract weak unbalanced signals from strong background noise. A novel three-step filtering framework is proposed in this paper. A zero phase filter was used as the pre-processing layer to filter out the high frequency components in the original signal and reduce the number of parameter-adaptive variational mode decompositions (PAVMD) needed. The PAVMD was used to decompose the non-stationary vibration signal before Adaptive Neuron Linear (Adaline) function was used to fit sinusoidal signal parameters. A measurement index, termed the correlation amplitude (CA) index, is constructed. The parameterisation of PAVMD was guided by the CA index, and the modal component of the unbalanced fault features were located. The simulation and real cylinder signals proved that the proposed method could effectively extract unbalanced signals under noise interference, and the unbalance was identified accurately by the influence coefficient method. Experiments on a threshing cylinder showed that the amplitude identification error was <24 g in single-sided unbalance identification, and the amplitude identification error was <27 g in double-sided unbalance identification. The proposed method had high robustness and small identification error, especially under short-time working conditions, compared with other similar approaches.

Structural variant calling and clinical interpretation in 6224 unsolved rare disease exomes

Structural variants (SVs), including large deletions, duplications, inversions, translocations, and more complex events have the potential to disrupt gene function resulting in rare disease. Nevertheless, current pipelines and clinical decision support systems for exome sequencing (ES) tend to focus on small alterations such as single nucleotide variants (SNVs) and insertions-deletions shorter than 50 base pairs (indels). Additionally, detection and interpretation of large copy-number variants (CNVs) are frequently performed. However, detection of other types of SVs in ES data is hampered by the difficulty of identifying breakpoints in off-target (intergenic or intronic) regions, which makes robust identification of SVs challenging. In this paper, we demonstrate the utility of SV calling in ES resulting in a diagnostic yield of 0.4% (23 out of 5825 probands) for a large cohort of unsolved patients collected by the Solve-RD consortium. Remarkably, 8 out of 23 pathogenic SV were not found by comprehensive read-depth-based CNV analysis, resulting in a 0.13% increased diagnostic value.

Parameter identification of Yoshida–Uemori combined hardening model by using a variable step size firefly algorithm

Abstract The material behavior under cyclic loading is more complex than under monotonic loading and the usage of the sophisticated constitutive models is required to accurately define the elastoplastic behaviors of the advanced high-strength steels and aluminum alloys. These models involve the numerous material parameters that are determined from cyclic tests and accurate calibration of the variables has a great influence on the description of the material response. Therefore, the development of a precise and robust identification method is needed to obtain reliable results. In this study, a systematic methodology depending upon the firefly algorithm (FA) with variable step size has been developed and Yoshida–Uemori combined hardening model parameters of a dual-phase steel (DP980) and an aluminum alloy (AA6XXX-T4) are determined. The identified parameters are verified based on comparisons between the finite element simulations of the cyclic uniaxial tension-compression tests and experimental data and also the search performance of the variable FA is evaluated by comparing it with the standard FA. It is seen from these comparisons that variable FA can easily find and rapidly converge to the global optimum solutions.

Efficient Faba Bean Leaf Disease Identification through Smart Detection using Deep Convolutional Neural Networks

Background: Legumes, such as lentils, field peas, Faba beans and chickpeas, are high in vitamins, fiber, important minerals and protein and can help avoid obesity and cardiovascular illnesses. They also contribute to ecosystem services, such as nitrogen fixation and resilience to environmental stresses. Despite a 60% increase in global pulse production from 2000 to 2021, a demand-supply gap, especially in South Asia, raises concerns about nutritional access. Since illnesses are currently an issue to the food security of faba beans, machine learning is required for efficient disease identification. Methods: This research employs Convolutional Neural Networks (CNNs) for robust Faba bean leaf disease identification. The CNN model is trained with diverse images representing specific diseases. The study focuses on diseases like Chocolate Spot, Faba Bean Gall, Rust and Healthy leaves. Image processing involves resizing, grayscale conversion and labeling. The CNN architecture includes eight convolutional layers, four max-pooling layers and three dropout layers. The model is trained using 80% of the dataset, validated using 20% and tested for accuracy. Result: The CNN model achieves an accuracy of 99.37% during training and 89.69% during validation after 75 epochs. Confusion matrix and classification report illustrate the model’s performance. It shows high precision, recall and F1 scores for each class, indicating balanced performance. Chocolate Spot and Rust exhibit the highest precision and F1 scores. The overall accuracy is 91%, comparable to other studies on Faba bean disease detection. The study presents a CNN-based disease identification system for Faba beans, demonstrating high accuracy and balanced performance across different diseases. The model’s effectiveness is comparable to other advanced techniques. The research highlights the potential of machine learning in optimizing disease management for Faba beans. Future work could explore a broader range of diseases and incorporate hybrid machine learning algorithms for further improvement.

MTG_CD: Multi-scale learnable transformation graph for fault classification and diagnosis in microservices

The rapid advancement of microservice architecture in the cloud has led to the necessity of effectively detecting, classifying, and diagnosing run failures in microservice applications. Due to the high dynamics of cloud environments and the complex dependencies between microservices, it is challenging to achieve robust real-time system fault identification. This paper proposes an interpretable fault diagnosis framework tailored for microservice architecture, namely Multi-scale Learnable Transformation Graph for Fault Classification and Diagnosis(MTG_CD). Firstly, we employ multi-scale neural transformation and graph structure adjacency matrix learning to enhance data diversity while extracting temporal-structural features from system monitoring metrics Secondly, a graph convolutional network (GCN) is utilized to fuse the extracted temporal-structural features in a multi-feature modeling approach, which helps to improve the accuracy of anomaly detection. To identify the root cause of system faults, we finally conduct a coarse-grained level diagnosis and exploration after obtaining the results of classifying the fault data. We evaluate the performance of MTG_CD on the microservice benchmark SockShop, demonstrating its superiority over several baseline methods in detecting CPU usage overhead, memory leak, and network delay faults. The average macro F1 score improves by 14.05%.

ResNet1D-Based Personal Identification with Multi-Session Surface Electromyography for Electronic Health Record Integration.

Personal identification is an important aspect of managing electronic health records (EHRs), ensuring secure access to patient information, and maintaining patient privacy. Traditionally, biometric, signature, username/password, photo identity, etc., are employed for user authentication. However, these methods can be prone to security breaches, identity theft, and user inconvenience. The security of personal information is of paramount importance, particularly in the context of EHR. To address this, our study leverages ResNet1D, a deep learning architecture, to analyze surface electromyography (sEMG) signals for robust identification purposes. The proposed ResNet1D-based personal identification approach using the sEMG signal can offer an alternative and potentially more secure method for personal identification in EHR systems. We collected a multi-session sEMG signal database from individuals, focusing on hand gestures. The ResNet1D model was trained using this database to learn discriminative features for both gesture and personal identification tasks. For personal identification, the model validated an individual's identity by comparing captured features with their own stored templates in the healthcare EHR system, allowing secure access to sensitive medical information. Data were obtained in two channels when each of the 200 subjects performed 12 motions. There were three sessions, and each motion was repeated 10 times with time intervals of a day or longer between each session. Experiments were conducted on a dataset of 20 randomly sampled subjects out of 200 subjects in the database, achieving exceptional identification accuracy. The experiment was conducted separately for 5, 10, 15, and 20 subjects using the ResNet1D model of a deep neural network, achieving accuracy rates of 97%, 96%, 87%, and 82%, respectively. The proposed model can be integrated with healthcare EHR systems to enable secure and reliable personal identification and the safeguarding of patient information.

An ultrafast and robust structural damage identification framework enabled by an optimized extreme learning machine

Artificial intelligence (AI) has recently been implemented in structural health monitoring (SHM) systems for damage detection and identification. However, existing AI methods often involve a high number of parameters, resulting in the laborious workload during model training and implementation. In this study, we develop a lightweight damage identification framework embedded with an optimized extreme learning machine (ELM) using a significantly reduced number of parameters from frequency features of the structural vibrational signals. Coupled with the ensemble empirical mode decomposition (EEMD), the frequency features were abstracted from the structural vibrational signals in the data pre-processing step. Then, a metaheuristic algorithm (chaos game optimization, CGO) was chosen to optimize the model weight of the ELM to guarantee damage identification accuracy. Once the model is trained, the structural damages can be precisely identified despite the noise-contaminated data. We validated the efficiency and accuracy of the proposed framework with a public database and compare its performance with several common metaheuristic algorithms. With the proven capability, the proposed framework shows a promising future as a handy and reliable AI-assisted digital tool for robust development in next-generation structural monitoring systems.

Uncertainty-guided label correction with wavelet-transformed discriminative representation enhancement

Label noises, categorized into closed-set noise and open-set noise, are prevalent in real-world scenarios and can seriously hinder the generalization ability of models. Identifying noise is challenging because noisy samples closely resemble true positives. Existing approaches often assume a single noise source, oversimplify closed-set noise, or treat open-set noise as toxic and eliminate it, resulting in limited practical effects. To address these issues, we present a novel approach named uncertainty-guided label correction with wavelet-transformed discriminative representation enhancement (Ultra), designed to mitigate the effects of mixed noise. Specifically, our approach considers a more practical noise setting. To achieve robust mixed-noise identification, we initially look into a learnable wavelet filter for obtaining discriminative features and filtering spurious cues automatically at the representation level. Subsequently, we introduce a two-fold uncertainty estimation to stably locate noise within the corrupted supervised signal level. These insights pave the way for a simple yet potent label correction technique, enabling comprehensive utilization of open-set noise, which can be rendered non-toxic in a specific manner, in contrast to harmful closed-set noise. Experimental validation on datasets with synthetic mixed noise, web noise corruption, and a real-world dataset confirms the effectiveness and generality of Ultra. Furthermore, our approach enhances the application of efficient techniques (e.g., supervised contrastive learning) within label noise scenarios.

ARGNet: using deep neural networks for robust identification and classification of antibiotic resistance genes from sequences

BackgroundEmergence of antibiotic resistance in bacteria is an important threat to global health. Antibiotic resistance genes (ARGs) are some of the key components to define bacterial resistance and their spread in different environments. Identification of ARGs, particularly from high-throughput sequencing data of the specimens, is the state-of-the-art method for comprehensively monitoring their spread and evolution. Current computational methods to identify ARGs mainly rely on alignment-based sequence similarities with known ARGs. Such approaches are limited by choice of reference databases and may potentially miss novel ARGs. The similarity thresholds are usually simple and could not accommodate variations across different gene families and regions. It is also difficult to scale up when sequence data are increasing.ResultsIn this study, we developed ARGNet, a deep neural network that incorporates an unsupervised learning autoencoder model to identify ARGs and a multiclass classification convolutional neural network to classify ARGs that do not depend on sequence alignment. This approach enables a more efficient discovery of both known and novel ARGs. ARGNet accepts both amino acid and nucleotide sequences of variable lengths, from partial (30–50 aa; 100–150 nt) sequences to full-length protein or genes, allowing its application in both target sequencing and metagenomic sequencing. Our performance evaluation showed that ARGNet outperformed other deep learning models including DeepARG and HMD-ARG in most of the application scenarios especially quasi-negative test and the analysis of prediction consistency with phylogenetic tree. ARGNet has a reduced inference runtime by up to 57% relative to DeepARG.ConclusionsARGNet is flexible, efficient, and accurate at predicting a broad range of ARGs from the sequencing data. ARGNet is freely available at https://github.com/id-bioinfo/ARGNet, with an online service provided at https://ARGNet.hku.hk.7nR91jX7ySRMRPZrE5tsTyVideo

A novel AI-powered method for robust identification of operational phases in refrigerators

Refrigerators are essential appliances in our daily lives, while the accurate identification of their operational phase is crucial for various applications, such as energy management and maintenance. This paper introduces a novel method to predict the operational phase of refrigerators based on power consumption data, exploiting artificial intelligence (AI) and pattern recognition techniques. Specifically, the proposed method initially identifies the driving speed of the compressor (fixed or variable speed) utilizing an Adaptive Boosting, (AdaBoost) classifier. Following that, a Long Short-Term Memory (LSTM) classifier was designed for each type of compressor based on extracted statistical features. This classifier is used to categorize daily power consumption into the different operational phases. Moreover, the proposed AI-driven diagnostic tool facilitates operational-based prescriptive maintenance by addressing periodic and phase-specific maintenance actions while providing homeowners with informed suggestions on alternative refrigerator models that offer greater cost savings and reduced annual energy consumption. The proposed method was tested on real and simulated data for over 20 years, with the AdaBoost classifier achieving an accuracy close to 97% and an F1 score close to 97% and the LSTM nodes achieving an accuracy of approximately 95% in all cases. Finally, the method demonstrated robustness when applied to new datasets, showing comparable performance on unknown data, rendering it a powerful tool for refrigerator management and maintenance.

A robust self-supervised image hashing method for content identification with forensic detection of content-preserving manipulations

Image content identification systems have many applications in industry and academia. In particular, a hash-based content identification system uses a robust image hashing function that computes a short binary identifier summarizing the perceptual content in a picture and is invariant against a set of expected manipulations while being capable of differentiating between different pictures. A common approach to designing these algorithms is crafting a processing pipeline by hand. Unfortunately, once the context changes, the researcher may need to define a new function to adapt. A deep hashing approach exploits the feature learning capabilities in deep networks to generate a feature vector that summarizes the perceptual content in the image, achieving outstanding performance for the image retrieval task, which requires measuring semantic and perceptual similarity between items. However, its application to robust content identification systems is an open area of opportunity. Also, image hashing functions are valuable tools for image authentication. However, to our knowledge, its application to content-preserving manipulation detection for image forensics tasks is still an open research area. In this work, we propose a deep hashing method exploiting the metric learning capabilities in contrastive self-supervised learning with a new modular loss function for robust image hashing. Moreover, we propose a novel approach for content-preserving manipulation detection for image forensics through a sensitivity component in our loss function. We validate our method through extensive experimentation in different data sets and configurations, validating the generalization properties in our work.

Enhancements in Immediate Speech Emotion Detection: Harnessing Prosodic and Spectral Characteristics

Speech is essential to human communication for expressing and understanding feelings. Emotional speech processing has challenges with expert data sampling, dataset organization, and computational complexity in large-scale analysis. This study aims to reduce data redundancy and high dimensionality by introducing a new speech emotion recognition system. The system employs Diffusion Map to reduce dimensionality and includes Decision Trees and K-Nearest Neighbors(KNN)ensemble classifiers. These strategies are suggested to increase voice emotion recognition accuracy. Speech emotion recognition is gaining popularity in affective computing for usage in medical, industry, and academics. This project aims to provide an efficient and robust real-time emotion identification framework. In order to identify emotions using supervised machine learning models, this work makes use of paralinguistic factors such as intensity, pitch, and MFCC. In order to classify data, experimental analysis integrates prosodic and spectral information utilizing methods like Random Forest, Multilayer Perceptron, SVM, KNN, and Gaussian Naïve Bayes. Fast training times make these machine learning models excellent for real-time applications. SVM and MLP have the highest accuracy at 70.86% and 79.52%, respectively. Comparisons to benchmarks show significant improvements over earlier models.

A robust model predictive control unified framework for autonomous rendezvous and docking with a tumbling target

During autonomous rendezvous and docking (AR&D) with a tumbling target, multiple unfavorable engineering constraints exist, such as constraints on the control, velocity, or collision avoidance. Meanwhile, many sources of uncertainty exist, including incomplete and inaccurate measurements, control deviation, dynamic uncertainty, and inertial parameter uncertainty of the tumbling target, which significantly complicate the design of the guidance and control algorithm. Considering these challenges, this paper proposes a robust identification, planning, and control unified framework for AR&D with a tumbling target, which incorporates an extended Kalman filter (EKF) with a robust tube-based model predictive control for tracking (TMPT) controller. First, the EKF is designed to estimate the system state and inertial parameters of the target and then predict the future behavior of the tumbling target. Additionally, the prediction information of the target is transmitted to the designed TMPT controller as a reference signal. This TMPT controller unites trajectory planning and control in a single layer that can ensure the stability and recursive feasibility of the algorithm with a short prediction horizon. Moreover, it can guarantee that the closed-loop system is robustly and asymptotically convergent to the reachable set whose center is the optimal reachable periodic trajectory concerning the docking port of the tumbling target. The unified framework represents a robustness-to-optimality control strategy that can accomplish the AR&D mission efficiently while considering multiple engineering constraints and uncertainties. Finally, the efficiency of the proposed control framework is verified through a numerical simulation.

Intelligent Structure Identification and Robust Control Implementation

This study outlines a comprehensive strategy for designing and implementing robust controllers tailored for intelligent structures. This study presents a robust control-based structural identification technique that uses the input/output data of the system to construct a state-space mode and frequency domain. To reduce vibrations, a robust controller is created using the control Simulink model. The identification and robust control of smart structures using Simulink involve a combination of system identification techniques and control design within the MATLAB Simulink environment. The key challenge is dealing with uncertainties and variations in system dynamics. Robust control methods have been employed to suppress the vibrations during dynamic disturbances. These methods are important for mechanical systems operating under stochastic loading conditions.