Search

In this paper, a LQR controller in combination with compound disturbance observer was designed and applied for a two-wheeled inverted pendulum robot (TWIR) in laboratory. The experimental results showed that the designed controller was able to handle with model's uncertainty and unknown input disturbance.

Detecting and understanding interactions between students and teachers in classroom is an important criterion for computer vision-based educational assistive systems. Recently, deep long-range spatial dependencies modeling techniques, such as non-local networks, have been proven to be very effective for such tasks. Yet, regarding global context generation, we analyze that the non-local operation only compares pixels using their values, which cannot pertain to structural information. In this paper, we first extend the non-local module to corporate locality attributes. We further observe that each query is treated uniformly to generate the attention map. Hence, we incorporate distance-wise representations with an efficient implementation into the non-local formulas. The proposed locality and relative distance-aware non-local module is integrated into an object detection architecture namely Libra-RCNN and is evaluated through our experiments on a pre-access hand-raising gesture dataset. Our straightforward modification achieves 0.8% and 2.8% higher performance compared to the baseline Libra-RCNN model, in terms of mAP 0.5 and mAP 0.75 respectively.

Video or image-based people counting in real-time has multiple applications in intelligent transportation, density estimation or class management, and so on. This problem is usually carried out by detecting people using conventional detectors. However, this approach can be failed when people stay in various postures or are occluded by each other. In this paper, we notice that even a main part of human body is occluded, their face and head are still observable. We then propose a method that counts people based on face and head detection and pairing. Instead of deploying only face or head detector, we apply both detectors as in many cases the human does not turn his/her face to camera then head detector takes advantage. Otherwise, face detector produces reliable results. The fact of combining both head and face detection results will lead to duplicated responses for one person. We then propose a simple yet effective alignment technique to pair a face with a head of a person. Subsequently, the remaining heads and faces which are not paired with any other faces or heads will be added to our people counter to increase the true positive rate. We evaluate our proposed method on four datasets (Hollywood, Casablanca, Wider Face, and our own dataset). The experimental results show an improvement of average precision and recall comparing to the original head or face detectors.

Nowadays, while modeling environments provide users with facilities to specify different kinds of artifacts, e.g., metamodels, models, and transformations, the possibility of learning from previous modeling experiences and being assisted during modeling tasks remains largely unexplored. In this paper, we propose MORGAN, a recommender system based on a graph neural network (GNN) to assist modelers in performing the specification of metamodels and models. The (meta)model being specified, and the training data are encoded in a graph-based format by exploiting natural language processing (NLP) techniques. Afterward, a graph kernel function uses the extracted graphs to provide modelers with relevant recommendations to complete the partially specified (meta)models. We evaluated MORGAN on real-world datasets using various quality metrics, i.e., precision, recall, and F-measure. The experimental results are encouraging and demonstrate the feasibility of our tool to support modelers while specifying metamodels and models.

Need for comprehensive development of photovoltaic (PV) based micro-inverter interface for various rural applications is observed. This paper demonstrates the design, development, and real-time field experimentation of an energy-efficient, PV based single-phase micro-inverter for rural/remote applications. The proposed dual-stage single-phase inverter takes the advantage of interleaved DC/DC boost converter for cancellation of 180 ο phase-shifted harmonics and application of voltage doubler circuit results in a reduction of the transformer turns ratio with low flux leakage and thus endeavoring for better conversion efficiency at a reduced size of magnetics. Further, high-frequency switching at the first leg and low switching frequency at another leg of the full-bridge converter ensures lower switching losses at light load conditions. Experimental results are carried out at field location (latitude: 26.431°, longitude: 74.707°) in Rajasthan, India. The interleaved boost DC/DC converter efficiency is found approximately 91.73% at full load and 77.74% at 10% load. Micro-inverter overall efficiency is observed in between 57.34% to 91.6%.

Antonymy is one of the fundamental relations shaping the organization of the semantic lexicon. Therefore, automatic detection of antonymy can be leveraged to make contributions to different NLP tasks, such as Machine Translation, Sentiment Analysis, and Information Retrieval. Currently, most prior studies just focus on discriminating between antonyms and synonyms. However, not only synonymy but other semantic relations, such as hypernymy, co-hyponyms, which also get high similarities thereby making it hard to discriminate. Therefore, it is necessary to make a thorough research on identifying antonyms from a wide variety of other semantic relations. In this paper, we aim to identify Vietnamese antonyms pairs according to the vector semantics approach. Specifically, we build up specialized word embedding models by incorporating lexical-semantic resource and distributional information. In addition, we propose specialized Vietnamese features and utilize mutual information between words in order to integrate with word embedding vectors. This aims to generate more meaningful feature vectors for supervised classifiers solving antonym detection problems. Furthermore, we construct three reliable Vietnamese testing datasets consisting of AntSynlOOO, AntHyplOOO, and AntMixlOOO, for this task. Experimental results conducted on the datasets demonstrated that our model performs effectively.

Categorical data clustering has been attracted a lot of attention recently due to its necessary in the real-world applications. Many clustering methods have been proposed for categorical data. However, most of the existing algorithms require the predefined number of clusters which is usually unavailable in real-world problems. Only a few works focused on automatic clustering, but mainly handled for numerical data. This study develops a novel automatic fuzzy clustering using non-dominated sorting particle swarm optimization (AFC-NSPSO) algorithm for categorical data. The proposed AFC-NSPSO algorithm can automatically identify the optimal number of clusters and exploit the clustering result with the corresponding selected number of clusters. In addition, a new technique is investigated to identify the maximum number of clusters in a dataset based on the local density. To select a final solution in the first Pareto front, some internal validation indices are used. The performance of the proposed AFC-NSPSO on the real-world datasets collected from the UCI machine learning repository exhibits effectiveness compared with some other existing automatic categorical clustering algorithms. Besides, this study also applies the proposed algorithm to analyze a real-world case study with an unknown number of clusters.

Information provenance seeks to model how data evolves and is manipulated within a system, the processes and actions that affect the data, and who or what is responsible for these actions. Systems can generate large amounts of provenance data depending on the granularity and abstractions modeled for that system. Methods developed for managing and performing analytics over provenance data sets were created before the standardization of the ontology and data model. This initial work focuses on revisiting older provenance analytics methods and applying them to the new standard while also bringing them into a introducing big data platforms and infrastructure to tackle the issue of big provenance.

Neural Machine Translation recently has become the state-of-the-art approach in Machine Translation. One of the more advanced techniques concerning this approach, the attention model, tends to not use alignments from past translation steps and selects the context word purely using the devised attention score. Unfortunately, this sometimes leads to repetition and omission of important words in translations. To solve this problem, we propose a simple approach using coverage techniques that can be used in conjunction with a diverse number of attention models. Our experiments show that our improved technique increases the quality of translation on both English -Vietnamese and Japanese - Vietnamese language pairings.

In this paper we propose an simple digital learning platform for flexible energy detection using data with fine granularity. The platform is empowered with artificially intelligent methods aiming to quantify the uncertainty of building energy consumption at building level, as well as at the aggregated level. Two major learning tasks are perform in this context: prediction and classification. Firstly, the building energy prediction with various time steps resolution are perform using methods such as Fully Connected Neural Networks (FCNN), Long short-term memory (LSTM), and Decision Trees (DT). Secondly, a Support Vector Machine (SVM) method is used to unlock the building energy flexibility by performing classification assuming three different levels of flexibility. Further on, a collaborative task is integrate within the platform to improve the multi-class classification accuracy. Through the end, we argue that this approach can be considered a solid integrated and automated basic block able to incorporate future AI models in (near) real-time to explore the benefits at the synergy between built environment and emerging smart grid technologies and applications.

Semantic parsing in Natural language processing (NLP) is a challenging task, which has been studied for many years. The main purpose is to model the language as a logical form like a machine translation task. Recently, an approach which uses a Neural network with Sequence to sequence model (Seq2seq) has achieved positive results. However, there are many challenges which have not been solved thoroughly yet, especially in the problem of rare words. Rare words in a natural sentence are usually the name of an object, a place or number, time, etc. Although these words are very various and difficult for the model to capture meaning, it holds a key information role in human communication (for example: name all the rivers in colorado ?). There are some methods to solve this problem such as using Attention or using Copy mechanism. However, these methods still difficult to copy phrase rare words, especially in case these phrases are variable in size. This paper proposes a novel approach to solve this problem, namely Marking mechanism in Seq2seq. The main idea is to label special words which are rare-words in a sentence by the encoder (marking step) and the decoder represents the logical form based on those labels (transforming step). Our experiments demonstrate that this approach works effectively, achieved a competitive result with old methods on all 3 datasets Geo, Atis, Jobs and special outperformed on our Artificial dataset.

In this article, we propose a combinatorial testing technique for mobile application, thereby suggesting a combination of test-driven data generation techniques as well as developing a support tool named CTGen which using the IPO algorithm to generate test data with input model files, the tool also supports code generation for JUnit testing. The experiment results were compared in terms of test coverage, effort to build a test case with the PICT tool for positive results. Depending on the purpose of developer or tester, testing effort or reliability of application to use appropriate approaches and tools.

The number of individuals suffering from disability caused by congenital limb deficiency or amputation is on the rise and most commercial prostheses are high in cost, low in dexterity and transradial despite the high demand for transhumeral ones. Studies show that the rejection rate is high and that recipients are dissatisfied with the activity variations and respective reachable performance standards. This paper considers the current limitations of prosthesis and proposes an affordable transhumeral design based on bio-signals and active vision system. The overall mechanical design has 10 degrees of freedom, is fully 3D printed and modular. The system is distributed and support multimodal interaction based on biosignals and active vision system. The prosthesis has configurable level of automation and could autonomously choose the most suitable grip pattern for object handling from a wide range of supported objects.

The objective of this paper is to assess parameter robustness of a controller design based on feedback linearisation applied to a wind turbine with a permanent magnet synchronous generator (PMSG). To study robust local stability, we employ an analysis of the eigenvalues of the small signal model. Some controller parameters are deduced based on measured system variables. It is therefore of interest to know bounds on measurement errors that are guaranteed to preserve local stability. Finally, simulations of a 2MW direct drive PMSG wind turbine system illustrate the effectiveness of the method.

Hardware security primitives such as physical obfuscated keys (POKs) allow tamper-resistant storage of random keys based on manufacturing or physical variability. The output bits of existing POK designs need to be first corrected due to measurement noise using error correction methods and then de-correlated by privacy amplification processes. These additional requirements increase the hardware overhead and reduce the efficiency of the system. In this work, we propose an intrinsically reliable POK design capable of generating random bits by exploiting the limits of the lithographic process for a given technology. Our design does not require any error correction and requires only XOR circuits for privacy amplification which reduces the hardware overhead of the whole system.

International audience; Software developers interact with APIs on a daily basis and, therefore, often face the need to learn how to use new APIs suitable for their purposes. Previous work has shown that recommending usage patterns to developers facilitates the learning process. Current approaches to usage pattern recommendation, however, still suffer from high redundancy and poor run-time performance. In this paper, we reformulate the problem of usage pattern recommendation in terms of a collaborative-filtering recommender system. We present a new tool, FOCUS, which mines open-source project repositories to recommend API method invocations and usage patterns by analyzing how APIs are used in projects similar to the current project. We evaluate FOCUS on a large number of Java projects extracted from GitHub and Maven Central and find that it outperforms the state-of-the-art approach PAM with regards to success rate, accuracy, and execution time. Results indicate the suitability of context-aware collaborative-filtering recommender systems to provide API usage patterns.

Clustering is an important technique which is used to discover the data structure. Clustering is applied in many areas, such as customer segmentation, image recognition, social science, and so on. However, most of the existing clustering methods suffer from two major drawbacks including 1) the susceptibility of clustering result due to the randomly initial centers and 2) the sensitivity of outliers and noise data. To solve these two problems, this study proposes a new algorithm named genetic algorithm-based fuzzy c-ordered-means algorithm (GA-FCOM). Herein, the fuzzy c-ordered-means algorithm (FCOM) can deal with noise and outliers data while the genetic algorithm is employed to obtain the optimal initial centroids efficiently during the clustering process. An experiment is conducted using the benchmark datasets collected from the UCI machine repository to validate the proposed algorithm. The computational results indicate that the proposed GA-FCOM outperforms fuzzy c-means algorithm (FCM) and FCOM in terms of both accuracy and objective function values.

Based on the green wave theory, we proposed a novel algorithm to tackle congested condition in Vietnam urban areas. While green bandwidth of priority routes were still maximized, the dynamic nature of traffic status indicators were firstly considered in the design stage. Furthermore, we constructed a hardware system, employing a PLC S7-1200 at each intersection, to control traffic lights. Communication between the host machine at an operation center and each PLC was implemented via a Profinet port embedded in the PLC and via TCP/IP protocol.

Automatic Test Equipment (ATE) obsolescence drives a requirement for rehosting Test Program Sets (TPSs). However, due to the many variations of programming languages, in most cases, the target language of the new ATE platform differs from the legacy language. This paper discusses a flexible, modular software approach for TPS rehost efforts. The new approach can be used to convert code while also allowing for the automated creation of additional output products (e.g., Test Requirements Document (TRD), Automatic Test Markup Language (ATML) files, fault universe, and ATE instrument requirements). Furthermore, the modular design and nonspecific, intermediate data structure significantly improves code reusability for rehosting between different ATE platforms. Additionally, as a case study, this paper discusses the benefits achieved using this approach.