sensors-logo

Journal Browser

Journal Browser

Special Issue "Advances in Machine Learning for Intelligent Engineering Systems and Applications"

A special issue of Sensors (ISSN 1424-8220). This special issue belongs to the section "Intelligent Sensors".

Deadline for manuscript submissions: closed (30 November 2019).

Special Issue Editors

Dr. Anastasios Doulamis
Website
Guest Editor
Photogrammetry and Computer Vision Lab., National Technical University of Athens, Athens 15773, Greece
Interests: image processing; computer vision; robotic systems; deep machine learning; machine learning; Markovian models; signal processing and pattern analysis
Special Issues and Collections in MDPI journals
Dr. Nikolaos Doulamis
Website
Guest Editor
National Technical University of Athens, 9 Heroon Polytechniou Str., Athens 15773, Greece
Interests: pattern recognition; machine learning; signal processing; image /hyper-spectral sensors
Special Issues and Collections in MDPI journals
Prof. Dr. Athanasios Voulodimos
Website
Guest Editor
Department of Informatics and Computer Engineering, University of West Attica, Agiou Spiridonos 28, Egaleo 122 43, Greece
Interests: machine learning; artificial intelligence; multimedia; intelligent systems; pervasive computing
Special Issues and Collections in MDPI journals

Special Issue Information

Dear Colleagues,

Latest advances in machine learning have contributed to great developments in many areas of interest to the engineering community. Data-driven or domain-oriented engineering applications can benefit significantly from the latest developments in machine learning theories and methods (including deep, reinforcement, transfer and extreme learning), but may also promote the development of learning algorithms, optimization approaches, fusion techniques for multimodal data, novel hardware and network architectures. The rapid development in these fields has also stimulated new research on sensors and sensor networks.

The purpose of this Special Issue is to provide a forum for engineers, data scientists, researchers and practitioners to present new academic research and industrial development on machine learning for engineering applications. The Special Issue aims at original research papers in the field, covering new theories, algorithms, systems, as well as new implementations and applications incorporating state-of-the-art machine learning techniques. Emphasis will be given on systems that incorporate new sensors and the configuration of them. Review articles and works on performance evaluation and benchmark datasets are also solicited.

Indicative domains of application of interest to the Special Issue include:

  • Research on sensors for new critical engineering applications
  • Sensors networks and drones to survey critical infrastructures
  • Software and hardware architectures for new sensorial systems in managing critical infrastructures
  • Electrical and mechanical engineering, production management and optimization, manufacturing, failure detection, energy management, smart grid
  • Robotics and automation, computer vision and pattern recognition applications, critical infrastructure protection
  • Civil engineering, construction management and optimization, structural health monitoring, earthquake engineering, urban planning
  • Transportation, hydraulics, water power and environmental engineering
  • Surveying and geospatial engineering, spatial planning, and remote sensing
  • Materials science and engineering
  • Biomedical engineering

Dr. Anastasios Doulamis
Dr. Nikolaos Doulamis
Dr. Athanasios Voulodimos
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All papers will be peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Sensors is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2000 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

  • Engineering Applications
  • Machine Learning
  • Deep Learning
  • Intelligent Systems

Published Papers (18 papers)

Order results
Result details
Select all
Export citation of selected articles as:

Research

Jump to: Other

Open AccessArticle
Standing-Posture Recognition in Human–Robot Collaboration Based on Deep Learning and the Dempster–Shafer Evidence Theory
Sensors 2020, 20(4), 1158; https://doi.org/10.3390/s20041158 - 20 Feb 2020
Cited by 3
Abstract
During human–robot collaborations (HRC), robot systems must accurately perceive the actions and intentions of humans. The present study proposes the classification of standing postures from standing-pressure images, by which a robot system can predict the intended actions of human workers in an HRC [...] Read more.
During human–robot collaborations (HRC), robot systems must accurately perceive the actions and intentions of humans. The present study proposes the classification of standing postures from standing-pressure images, by which a robot system can predict the intended actions of human workers in an HRC environment. To this end, it explores deep learning based on standing-posture recognition and a multi-recognition algorithm fusion method for HRC. To acquire the pressure-distribution data, ten experimental participants stood on a pressure-sensing floor embedded with thin-film pressure sensors. The pressure data of nine standing postures were obtained from each participant. The human standing postures were discriminated by seven classification algorithms. The results of the best three algorithms were fused using the Dempster–Shafer evidence theory to improve the accuracy and robustness. In a cross-validation test, the best method achieved an average accuracy of 99.96%. The convolutional neural network classifier and data-fusion algorithm can feasibly classify the standing postures of human workers. Full article
Show Figures

Figure 1

Open AccessArticle
Lie Group Methods in Blind Signal Processing
Sensors 2020, 20(2), 440; https://doi.org/10.3390/s20020440 - 13 Jan 2020
Cited by 2
Abstract
This paper deals with the use of Lie group methods to solve optimization problems in blind signal processing (BSP), including Independent Component Analysis (ICA) and Independent Subspace Analysis (ISA). The paper presents the theoretical fundamentals of Lie groups and Lie algebra, the geometry [...] Read more.
This paper deals with the use of Lie group methods to solve optimization problems in blind signal processing (BSP), including Independent Component Analysis (ICA) and Independent Subspace Analysis (ISA). The paper presents the theoretical fundamentals of Lie groups and Lie algebra, the geometry of problems in BSP as well as the basic ideas of optimization techniques based on Lie groups. Optimization algorithms based on the properties of Lie groups are characterized by the fact that during optimization motion, they ensure permanent bonding with a search space. This property is extremely significant in terms of the stability and dynamics of optimization algorithms. The specific geometry of problems such as ICA and ISA along with the search space homogeneity enable the use of optimization techniques based on the properties of the Lie groups O ( n ) and S O ( n ) . An interesting idea is that of optimization motion in one-parameter commutative subalgebras and toral subalgebras that ensure low computational complexity and high-speed algorithms. Full article
Show Figures

Figure 1

Open AccessArticle
A Novel Methodology for Series Arc Fault Detection by Temporal Domain Visualization and Convolutional Neural Network
Sensors 2020, 20(1), 162; https://doi.org/10.3390/s20010162 - 26 Dec 2019
Cited by 4
Abstract
AC arc faults are one of the most important causes of residential electrical wiring fires, which may produce extremely high temperatures and easily ignite surrounding combustible materials. The global interest in machine learning-based methods for arc fault diagnosis applications is increasing due to [...] Read more.
AC arc faults are one of the most important causes of residential electrical wiring fires, which may produce extremely high temperatures and easily ignite surrounding combustible materials. The global interest in machine learning-based methods for arc fault diagnosis applications is increasing due to continuous challenges in efficiency and accuracy. In this paper, a temporal domain visualization convolutional neural network (TDV-CNN) methodology is proposed. The current transformer and high-speed data acquisition system are used to collect the current of a series of arc faults, then the signal is filtered by a digital filter and converted into a gray image in time sequence before being fed into TDV-CNN. Five different electric loads were selected for experimental validation with various signal characteristics, including vacuum cleaner, fluorescent lamp, dimmer, heater, and desktop computer. The experimental results confirm that the classification accuracy of the five loads’ work states in the ten categories could reach 98.7% or even higher by adjusting parameters perfectly. The methodology is believed to be reliable for series arc detection with relatively high accuracy and also has important potential applications in other fault diagnosis fields. Full article
Show Figures

Figure 1

Open AccessArticle
Intelligent Identification for Rock-Mineral Microscopic Images Using Ensemble Machine Learning Algorithms
Sensors 2019, 19(18), 3914; https://doi.org/10.3390/s19183914 - 11 Sep 2019
Cited by 1
Abstract
It is significant to identify rock-mineral microscopic images in geological engineering. The task of microscopic mineral image identification, which is often conducted in the lab, is tedious and time-consuming. Deep learning and convolutional neural networks (CNNs) provide a method to analyze mineral microscopic [...] Read more.
It is significant to identify rock-mineral microscopic images in geological engineering. The task of microscopic mineral image identification, which is often conducted in the lab, is tedious and time-consuming. Deep learning and convolutional neural networks (CNNs) provide a method to analyze mineral microscopic images efficiently and smartly. In this research, the transfer learning model of mineral microscopic images is established based on Inception-v3 architecture. The four mineral image features, including K-feldspar (Kf), perthite (Pe), plagioclase (Pl), and quartz (Qz or Q), are extracted using Inception-v3. Based on the features, the machine learning methods, logistic regression (LR), support vector machine (SVM), random forest (RF), k-nearest neighbors (KNN), multilayer perceptron (MLP), and gaussian naive Bayes (GNB), are adopted to establish the identification models. The results are evaluated using 10-fold cross-validation. LR, SVM, and MLP have a significant performance among all the models, with accuracy of about 90.0%. The evaluation result shows LR, SVM, and MLP are the outstanding single models in high-dimensional feature analysis. The three models are also selected as the base models in model stacking. The LR model is also set as the meta classifier in the final prediction. The stacking model can achieve 90.9% accuracy, which is higher than all the single models. The result also shows that model stacking effectively improves model performance. Full article
Show Figures

Figure 1

Open AccessArticle
A Method of Short Text Representation Based on the Feature Probability Embedded Vector
Sensors 2019, 19(17), 3728; https://doi.org/10.3390/s19173728 - 28 Aug 2019
Cited by 2
Abstract
Text representation is one of the key tasks in the field of natural language processing (NLP). Traditional feature extraction and weighting methods often use the bag-of-words (BoW) model, which may lead to a lack of semantic information as well as the problems of [...] Read more.
Text representation is one of the key tasks in the field of natural language processing (NLP). Traditional feature extraction and weighting methods often use the bag-of-words (BoW) model, which may lead to a lack of semantic information as well as the problems of high dimensionality and high sparsity. At present, to solve these problems, a popular idea is to utilize deep learning methods. In this paper, feature weighting, word embedding, and topic models are combined to propose an unsupervised text representation method named the feature, probability, and word embedding method. The main idea is to use the word embedding technology Word2Vec to obtain the word vector, and then combine this with the feature weighted TF-IDF and the topic model LDA. Compared with traditional feature engineering, the proposed method not only increases the expressive ability of the vector space model, but also reduces the dimensions of the document vector. Besides this, it can be used to solve the problems of the insufficient information, high dimensions, and high sparsity of BoW. We use the proposed method for the task of text categorization and verify the validity of the method. Full article
Show Figures

Figure 1

Open AccessArticle
Real-Time Water Surface Object Detection Based on Improved Faster R-CNN
Sensors 2019, 19(16), 3523; https://doi.org/10.3390/s19163523 - 12 Aug 2019
Cited by 5
Abstract
In this paper, we consider water surface object detection in natural scenes. Generally, background subtraction and image segmentation are the classical object detection methods. The former is highly susceptible to variable scenes, so its accuracy will be greatly reduced when detecting water surface [...] Read more.
In this paper, we consider water surface object detection in natural scenes. Generally, background subtraction and image segmentation are the classical object detection methods. The former is highly susceptible to variable scenes, so its accuracy will be greatly reduced when detecting water surface objects due to the changing of the sunlight and waves. The latter is more sensitive to the selection of object features, which will lead to poor generalization as a result, so it cannot be applied widely. Consequently, methods based on deep learning have recently been proposed. The River Chief System has been implemented in China recently, and one of the important requirements is to detect and deal with the water surface floats in a timely fashion. In response to this case, we propose a real-time water surface object detection method in this paper which is based on the Faster R-CNN. The proposed network model includes two modules and integrates low-level features with high-level features to improve detection accuracy. Moreover, we propose to set the different scales and aspect ratios of anchors by analyzing the distribution of object scales in our dataset, so our method has good robustness and high detection accuracy for multi-scale objects in complex natural scenes. We utilized the proposed method to detect the floats on the water surface via a three-day video surveillance stream of the North Canal in Beijing, and validated its performance. The experiments show that the mean average precision (MAP) of the proposed method was 83.7%, and the detection speed was 13 frames per second. Therefore, our method can be applied in complex natural scenes and mostly meets the requirements of accuracy and speed of water surface object detection online. Full article
Show Figures

Figure 1

Open AccessArticle
Full-Waveform LiDAR Point Clouds Classification Based on Wavelet Support Vector Machine and Ensemble Learning
Sensors 2019, 19(14), 3191; https://doi.org/10.3390/s19143191 - 19 Jul 2019
Cited by 2
Abstract
Light Detection and Ranging (LiDAR) produces 3D point clouds that describe ground objects, and has been used to make object interpretation in many cases. However, traditional LiDAR only records discrete echo signals and provides limited feature parameters of point clouds, while full-waveform LiDAR [...] Read more.
Light Detection and Ranging (LiDAR) produces 3D point clouds that describe ground objects, and has been used to make object interpretation in many cases. However, traditional LiDAR only records discrete echo signals and provides limited feature parameters of point clouds, while full-waveform LiDAR (FWL) records the backscattered echo in the form of a waveform, which provides more echo information. With the development of machine learning, support vector machine (SVM) is one of the commonly used classifiers to deal with high dimensional data via small amount of samples. Ensemble learning, which combines a set of base classifiers to determine the output result, is presented and SVM ensemble is used to improve the discrimination ability, owing to small differences in features between different types of data. In addition, previous kernel functions of SVM usually cause under-fitting or over-fitting that decreases the generalization performance. Hence, a series of kernel functions based on wavelet analysis are used to construct different wavelet SVMs (WSVMs) that improve the heterogeneity of ensemble system. Meanwhile, the parameters of SVM have a significant influence on the classification result. Therefore, in this paper, FWL point clouds are classified by WSVM ensemble and particle swarm optimization is used to find the optimal parameters of WSVM. Experimental results illustrate that the proposed method is robust and effective, and it is applicable to some practical work. Full article
Show Figures

Figure 1

Open AccessArticle
A Machine-Learning Approach to Distinguish Passengers and Drivers Reading While Driving
Sensors 2019, 19(14), 3174; https://doi.org/10.3390/s19143174 - 19 Jul 2019
Cited by 3
Abstract
Driver distraction is one of the major causes of traffic accidents. In recent years, given the advance in connectivity and social networks, the use of smartphones while driving has become more frequent and a serious problem for safety. Texting, calling, and reading while [...] Read more.
Driver distraction is one of the major causes of traffic accidents. In recent years, given the advance in connectivity and social networks, the use of smartphones while driving has become more frequent and a serious problem for safety. Texting, calling, and reading while driving are types of distractions caused by the use of smartphones. In this paper, we propose a non-intrusive technique that uses only data from smartphone sensors and machine learning to automatically distinguish between drivers and passengers while reading a message in a vehicle. We model and evaluate seven cutting-edge machine-learning techniques in different scenarios. The Convolutional Neural Network and Gradient Boosting were the models with the best results in our experiments. Results show accuracy, precision, recall, F1-score, and kappa metrics superior to 0.95. Full article
Show Figures

Figure 1

Open AccessArticle
An Industrial Micro-Defect Diagnosis System via Intelligent Segmentation Region
Sensors 2019, 19(11), 2636; https://doi.org/10.3390/s19112636 - 11 Jun 2019
Cited by 1
Abstract
In the field of machine vision defect detection for a micro workpiece, it is very important to make the neural network realize the integrity of the mask in analyte segmentation regions. In the process of the recognition of small workpieces, fatal defects are [...] Read more.
In the field of machine vision defect detection for a micro workpiece, it is very important to make the neural network realize the integrity of the mask in analyte segmentation regions. In the process of the recognition of small workpieces, fatal defects are always contained in borderline areas that are difficult to demarcate. The non-maximum suppression (NMS) of intersection over union (IOU) will lose crucial texture information especially in the clutter and occlusion detection areas. In this paper, simple linear iterative clustering (SLIC) is used to augment the mask as well as calibrate the score of the mask. We propose an SLIC head of object instance segmentation in proposal regions (Mask R-CNN) containing a network block to learn the quality of the predict masks. It is found that parallel K-means in the limited region mechanism in the SLIC head improved the confidence of the mask score, in the context of our workpiece. A continuous fine-tune mechanism was utilized to continuously improve the model robustness in a large-scale production line. We established a detection system, which included an optical fiber locator, telecentric lens system, matrix stereoscopic light, a rotating platform, and a neural network with an SLIC head. The accuracy of defect detection is effectively improved for micro workpieces with clutter and borderline areas. Full article
Show Figures

Figure 1

Open AccessArticle
Collaborative Representation Using Non-Negative Samples for Image Classification
Sensors 2019, 19(11), 2609; https://doi.org/10.3390/s19112609 - 08 Jun 2019
Cited by 3
Abstract
Collaborative representation based classification (CRC) is an efficient classifier in image classification. By using l 2 regularization, the collaborative representation based classifier holds competitive performances compared with the sparse representation based classifier using less computational time. However, each of the elements calculated from [...] Read more.
Collaborative representation based classification (CRC) is an efficient classifier in image classification. By using l 2 regularization, the collaborative representation based classifier holds competitive performances compared with the sparse representation based classifier using less computational time. However, each of the elements calculated from the training samples are utilized for representation without selection, which can lead to poor performances in some classification tasks. To resolve this issue, in this paper, we propose a novel collaborative representation by directly using non-negative representations to represent a test sample collaboratively, termed Non-negative Collaborative Representation-based Classifier (NCRC). To collect all non-negative collaborative representations, we introduce a Rectified Linear Unit (ReLU) function to perform filtering on the coefficients obtained by l 2 minimization according to CRC’s objective function. Next, we represent the test sample by using a linear combination of these representations. Lastly, the nearest subspace classifier is used to perform classification on the test samples. The experiments performed on four different databases including face and palmprint showed the promising results of the proposed method. Accuracy comparisons with other state-of-art sparse representation-based classifiers demonstrated the effectiveness of NCRC at image classification. In addition, the proposed NCRC consumes less computational time, further illustrating the efficiency of NCRC. Full article
Show Figures

Figure 1

Open AccessFeature PaperArticle
Using Random Forests on Real-World City Data for Urban Planning in a Visual Semantic Decision Support System
Sensors 2019, 19(10), 2266; https://doi.org/10.3390/s19102266 - 16 May 2019
Cited by 2
Abstract
The constantly increasing amount and availability of urban data derived from varying sources leads to an assortment of challenges that include, among others, the consolidation, visualization, and maximal exploitation prospects of the aforementioned data. A preeminent problem affecting urban planning is the appropriate [...] Read more.
The constantly increasing amount and availability of urban data derived from varying sources leads to an assortment of challenges that include, among others, the consolidation, visualization, and maximal exploitation prospects of the aforementioned data. A preeminent problem affecting urban planning is the appropriate choice of location to host a particular activity (either commercial or common welfare service) or the correct use of an existing building or empty space. In this paper, we propose an approach to address these challenges availed with machine learning techniques. The proposed system combines, fuses, and merges various types of data from different sources, encodes them using a novel semantic model that can capture and utilize both low-level geometric information and higher level semantic information and subsequently feeds them to the random forests classifier, as well as other supervised machine learning models for comparisons. Our experimental evaluation on multiple real-world data sets comparing the performance of several classifiers (including Feedforward Neural Networks, Support Vector Machines, Bag of Decision Trees, k-Nearest Neighbors and Naïve Bayes), indicated the superiority of Random Forests in terms of the examined performance metrics (Accuracy, Specificity, Precision, Recall, F-measure and G-mean). Full article
Show Figures

Figure 1

Open AccessArticle
Classifying Image Stacks of Specular Silicon Wafer Back Surface Regions: Performance Comparison of CNNs and SVMs
Sensors 2019, 19(9), 2056; https://doi.org/10.3390/s19092056 - 02 May 2019
Cited by 4
Abstract
In this work, we compare the performance of convolutional neural networks and support vector machines for classifying image stacks of specular silicon wafer back surfaces. In these image stacks, we can identify structures typically originating from replicas of chip structures or from grinding [...] Read more.
In this work, we compare the performance of convolutional neural networks and support vector machines for classifying image stacks of specular silicon wafer back surfaces. In these image stacks, we can identify structures typically originating from replicas of chip structures or from grinding artifacts such as comets or grinding grooves. However, defects like star cracks are also visible in those images. To classify these image stacks, we test and compare three different approaches. In the first approach, we train a convolutional neural net performing feature extraction and classification. In the second approach, we manually extract features of the images and use these features to train support vector machines. In the third approach, we skip the classification layers of the convolutional neural networks and use features extracted from different network layers to train support vector machines. Comparing these three approaches shows that all yield an accuracy value above 90%. With a quadratic support vector machine trained on features extracted from a convolutional network layer we achieve the best compromise between precision and recall rate of the class star crack with 99.3% and 98.6%, respectively. Full article
Show Figures

Graphical abstract

Open AccessArticle
A Hybrid Multi-Objective Optimization Model for Vibration Tendency Prediction of Hydropower Generators
Sensors 2019, 19(9), 2055; https://doi.org/10.3390/s19092055 - 02 May 2019
Cited by 5
Abstract
The hydropower generator unit (HGU) is a vital piece of equipment for frequency and peaking modulation in the power grid. Its vibration signal contains a wealth of information and status characteristics. Therefore, it is important to predict the vibration tendency of HGUs using [...] Read more.
The hydropower generator unit (HGU) is a vital piece of equipment for frequency and peaking modulation in the power grid. Its vibration signal contains a wealth of information and status characteristics. Therefore, it is important to predict the vibration tendency of HGUs using collected real-time data, and achieve predictive maintenance as well. In previous studies, most prediction methods have only focused on enhancing the stability or accuracy. However, it is insufficient to consider only one criterion (stability or accuracy) in vibration tendency prediction. In this paper, an intelligence vibration tendency prediction method is proposed to simultaneously achieve strong stability and high accuracy, where vibration signal preprocessing, feature selection and prediction methods are integrated in a multi-objective optimization framework. Firstly, raw sensor signals are decomposed into several modes by empirical wavelet transform (EWT). Subsequently, the refactored modes can be obtained by the sample entropy-based reconstruction strategy. Then, important input features are selected using the Gram-Schmidt orthogonal (GSO) process. Later, the refactored modes are predicted through kernel extreme learning machine (KELM). Finally, the parameters of GSO and KELM are synchronously optimized by the multi-objective salp swarm algorithm. A case study and analysis of the mixed-flow HGU data in China was conducted, and the results show that the proposed model performs better in terms of predicting stability and accuracy. Full article
Show Figures

Figure 1

Open AccessArticle
Research on Modeling and Analysis of Generative Conversational System Based on Optimal Joint Structural and Linguistic Model
Sensors 2019, 19(7), 1675; https://doi.org/10.3390/s19071675 - 08 Apr 2019
Cited by 1
Abstract
Generative conversational systems consisting of a neural network-based structural model and a linguistic model have always been considered to be an attractive area. However, conversational systems tend to generate single-turn responses with a lack of diversity and informativeness. For this reason, the conversational [...] Read more.
Generative conversational systems consisting of a neural network-based structural model and a linguistic model have always been considered to be an attractive area. However, conversational systems tend to generate single-turn responses with a lack of diversity and informativeness. For this reason, the conversational system method is further developed by modeling and analyzing the joint structural and linguistic model, as presented in the paper. Firstly, we establish a novel dual-encoder structural model based on the new Convolutional Neural Network architecture and strengthened attention with intention. It is able to effectively extract the features of variable-length sequences and then mine their deep semantic information. Secondly, a linguistic model combining the maximum mutual information with the foolish punishment mechanism is proposed. Thirdly, the conversational system for the joint structural and linguistic model is observed and discussed. Then, to validate the effectiveness of the proposed method, some different models are tested, evaluated and compared with respect to Response Coherence, Response Diversity, Length of Conversation and Human Evaluation. As these comparative results show, the proposed method is able to effectively improve the response quality of the generative conversational system. Full article
Show Figures

Figure 1

Open AccessArticle
The Actor-Dueling-Critic Method for Reinforcement Learning
Sensors 2019, 19(7), 1547; https://doi.org/10.3390/s19071547 - 30 Mar 2019
Cited by 3
Abstract
Model-free reinforcement learning is a powerful and efficient machine-learning paradigm which has been generally used in the robotic control domain. In the reinforcement learning setting, the value function method learns policies by maximizing the state-action value (Q value), but it suffers from [...] Read more.
Model-free reinforcement learning is a powerful and efficient machine-learning paradigm which has been generally used in the robotic control domain. In the reinforcement learning setting, the value function method learns policies by maximizing the state-action value (Q value), but it suffers from inaccurate Q estimation and results in poor performance in a stochastic environment. To mitigate this issue, we present an approach based on the actor-critic framework, and in the critic branch we modify the manner of estimating Q-value by introducing the advantage function, such as dueling network, which can estimate the action-advantage value. The action-advantage value is independent of state and environment noise, we use it as a fine-tuning factor to the estimated Q value. We refer to this approach as the actor-dueling-critic (ADC) network since the frame is inspired by the dueling network. Furthermore, we redesign the dueling network part in the critic branch to make it adapt to the continuous action space. The method was tested on gym classic control environments and an obstacle avoidance environment, and we design a noise environment to test the training stability. The results indicate the ADC approach is more stable and converges faster than the DDPG method in noise environments. Full article
Show Figures

Figure 1

Open AccessArticle
A Deep Neural Network Based Model for a Kind of Magnetorheological Dampers
Sensors 2019, 19(6), 1333; https://doi.org/10.3390/s19061333 - 17 Mar 2019
Cited by 7
Abstract
In this paper, a deep neural network based model for a set of small-scale magnetorheological dampers (MRD) is developed where relevant parameters that have a physical meaning are inputs to the model. An experimental platform and a 3D-printing rapid prototyping facility provided a [...] Read more.
In this paper, a deep neural network based model for a set of small-scale magnetorheological dampers (MRD) is developed where relevant parameters that have a physical meaning are inputs to the model. An experimental platform and a 3D-printing rapid prototyping facility provided a set of different conditions including MRD filled with two different MR fluids, which were used to train a Deep Neural Network (DNN), which is the core of the proposed model. Testing results indicate the model could forecast the hysteretic response of magnetorheological dampers for different load conditions and various physical configurations. Full article
Show Figures

Figure 1

Open AccessArticle
Anchor Generation Optimization and Region of Interest Assignment for Vehicle Detection
Sensors 2019, 19(5), 1089; https://doi.org/10.3390/s19051089 - 03 Mar 2019
Cited by 5
Abstract
Region proposal network (RPN) based object detection, such as Faster Regions with CNN (Faster R-CNN), has gained considerable attention due to its high accuracy and fast speed. However, it has room for improvements when used in special application situations, such as the on-board [...] Read more.
Region proposal network (RPN) based object detection, such as Faster Regions with CNN (Faster R-CNN), has gained considerable attention due to its high accuracy and fast speed. However, it has room for improvements when used in special application situations, such as the on-board vehicle detection. Original RPN locates multiscale anchors uniformly on each pixel of the last feature map and classifies whether an anchor is part of the foreground or background with one pixel in the last feature map. The receptive field of each pixel in the last feature map is fixed in the original faster R-CNN and does not coincide with the anchor size. Hence, only a certain part can be seen for large vehicles and too much useless information is contained in the feature for small vehicles. This reduces detection accuracy. Furthermore, the perspective projection results in the vehicle bounding box size becoming related to the bounding box position, thereby reducing the effectiveness and accuracy of the uniform anchor generation method. This reduces both detection accuracy and computing speed. After the region proposal stage, many regions of interest (ROI) are generated. The ROI pooling layer projects an ROI to the last feature map and forms a new feature map with a fixed size for final classification and box regression. The number of feature map pixels in the projected region can also influence the detection performance but this is not accurately controlled in former works. In this paper, the original faster R-CNN is optimized, especially for the on-board vehicle detection. This paper tries to solve these above-mentioned problems. The proposed method is tested on the KITTI dataset and the result shows a significant improvement without too many tricky parameter adjustments and training skills. The proposed method can also be used on other objects with obvious foreshortening effects, such as on-board pedestrian detection. The basic idea of the proposed method does not rely on concrete implementation and thus, most deep learning based object detectors with multiscale feature maps can be optimized with it. Full article
Show Figures

Figure 1

Other

Jump to: Research

Open AccessCase Report
Crack Detection on a Retaining Wall with an Innovative, Ensemble Learning Method in a Dynamic Imaging System
Sensors 2019, 19(21), 4784; https://doi.org/10.3390/s19214784 - 03 Nov 2019
Cited by 1
Abstract
In this study, an innovative, ensemble learning method in a dynamic imaging system of an unmanned vehicle is presented. The feasibility of the system was tested in the crack detection of a retaining wall in a climbing area or a mountain road. The [...] Read more.
In this study, an innovative, ensemble learning method in a dynamic imaging system of an unmanned vehicle is presented. The feasibility of the system was tested in the crack detection of a retaining wall in a climbing area or a mountain road. The unmanned vehicle can provide a lightweight and remote cruise routine with a Geographic Information System sensor, a Gyro sensor, and a charge-coupled device camera. The crack was the target to be tested, and the retaining wall was patrolled through the drone flight path setting, and then the horizontal image was instantly returned by using the wireless transmission of the system. That is based on the cascade classifier, and the feature comparison classifier was designed further, and then the machine vision correlation algorithm was used to analyze the target type information. First, the system collects the target image and background to establish the samples database, and then uses the Local Binary Patterns feature extraction algorithm to extract the feature values for classification. When the first stage classification is completed, the classification results are target features, and edge feature comparisons. The innovative ensemble learning classifier was used to analyze the image and determine the location of the crack for risk assessment. Full article
Show Figures

Figure 1

Back to TopTop