56016 research outputs found
Sort by
Expanded linear dynamic affect-expression model for lingering emotional expression in social robot
No full textThe service robot market is growing, and robots are replacing humans in many service industry jobs. Recently, as humans treat robots more emotionally and socially, designing robot emotions to improve human satisfaction in human???robot interaction (HRI) is crucial. Despite the importance of expressing the robot emotions, many robots only respond to the current stimulus when expressing emotions. Just as humans can feel the lingering effects of a strong stimulus after it has passed, social robots might do the same. For example, if a user hits a robot and the robot is very angry, even if the user praises the robot to make it feel better, the anger will not dissipate for some time. In this study, we propose the expanded linear dynamic affect-expression model (e-LDAEM) for expressing lingering emotion. Different intensity of stimuli in the e-LDAEM leads to different results, even if the robot is stimulated to the same emotion. The viscosity matrix and intensity vector show a positive correlation in the process of determining the emotion dynamics. This model enables the implementation of robots with diverse personalities by adjusting the lingering emotions to suit the robot???s form or situation. Through user evaluation results, the e-LDAEM has the proven effect of remaining in emotion for longer periods when the robot is given a strong stimulus. Thus, e-LDAEM is expected to improve the emotional bond between humans and robots
Seismic fragility curve update based on empirical fragility estimates employing automated ERGO platform
No full textTurning Berlin green frameworks into cubic crystals for cathodes with high-rate capability
No full textPrussian blue analogues (PBAs) have been considered as promising host frameworks for charge carriers because of their well-defined diffusion channel along the 100 direction. Among PBA families, Berlin green (BG) would be an ideal cathode platform because the empty carrier ion sites and two redox couples (Fe3+/2+-CN-Fe3+/2+) in the BG framework can deliver high specific capacity during battery operation. Nonetheless, in most solution-based precipitation processes, BG crystals are synthesized in irregular shapes rather than in well-defined cube shapes, thus limiting their capacities at high rate operations. In this work, given the aforementioned challenges, a simple two-step precipitation process to synthesize cubic BG without using any chelating agents and toxic acids was reported. Notably, an intermediate phase was identified as an important stage in converting irregularly shaped BG to cubic BG by releasing crystal water molecules from the framework. Utilizing well-aligned 100 channels in the cubic framework, cubic BG exhibits excellent electrochemical properties as a cathode for lithium-ion batteries, delivering a specific capacity of 107.2 mA h g(-1) at a high current density of 500 mA g(-1). A combined study of in situ X-ray diffraction and X-ray absorption fine structure analyses would provide a comprehensive structure-property relationship of BG cathodes
Artificial neural networks for insights into adsorption capacity of industrial dyes using carbon-based materials
No full textOrganic waste-derived carbon-based materials (CBMs) are commonly applied in sustainable wastewater treatment and waste management. CBMs can remove toxic, non-biodegradable and carcinogenic pollutants such as dyes which include indigo, triphenylmethyl, azo, anthraquinone and phthalocyanine derivatives. Nonetheless, their diverse composition, surface properties, presence of numerous surface functional groups and the altering adsorption experimental conditions to which they are applied against the elimination of organic dyes make it challenging to completely understand the removal mechanism. Herein, a dataset of 1514 data points was compiled from various published peer-reviewed journals along with additional adsorption experiments conducted in this study. Artificial neural networks (ANN) based machine learning (ML) model was compared with other ML and a deep learning model named Tab-Transformer and the findings proposed ANN showed superior prediction performance for adsorption capacity as a function of adsorbent synthesis conditions, adsorbent physical characteristics and adsorption experimental conditions. The hyperparameters of ANN model was optimized using Bayesian optimizer and the batch size, activation and units were proven to be more important than the number of hidden layers and learning rate. The ANN model exhibits a higher coefficient of determination (R2 = 0.98) and lower root mean square error (RMSE = 46.95 mg/g) values for test dataset. Feature importance using SHapley Additive exPlanations (SHAP) analysis suggested that the adsorption characteristics with 51.4% was the most important in the ANN prediction followed by the adsorption experimental condition (31.2%) and adsorbent synthesis condition (17.4%). Moreover, the impact of six most important features were individually analyzed. Finally, a detailed discussion on the environmental impact of the presented ANN model is also included
Analysis and optimization for non-orthogonal pilot sequence sets in massive MIMO systems
No full textIn modern wireless communication systems, orthogonal pilot signals has been generally employed in estimation of the channel state information. However, orthogonal pilot signals are inadequate for supporting the rapidly increasing requirements of communication throughput for 5G-and-beyond wireless environments, owing to the pilot contamination and short coherence times in high-mobility situations. To address these concerns, we present a new strategy for making use of non-orthogonal pilot sequences in channel estimation for multi-cell massive multiple-input multiple-output systems. First, we extend prior pilot assignment strategies based on the orthogonality of pilots to the general case of non-orthogonal pilot signals. Based on the proposed non-orthogonal pilot assignment strategy, we establish the minimal pilot length that fulfills a requirement for the channel estimate error, under a given degree of non-orthogonality. Then, we demonstrate validity of the pilot assignment strategy with the minimal length, which maximizes the entire network throughput. Simulation results show that our proposed method gives a significantly enhanced performance in terms of the net throughput compared to that with orthogonal pilot sequences. The performance gain becomes particularly significant with a higher density of users or shorter coherence time intervals
One camera-based laser keyhole welding monitoring system using deep learning
No full textThe laser-beam absorptance changes dynamically during laser keyhole welding due to unstable keyhole movements, and monitoring the absorptance can provide a deep understanding of the process. Recently, Kim et al. [1,2] developed a deep-learning-based method to monitor the absorptance by detecting the top and bottom keyhole apertures and estimating the absorptance from the reconstructed keyhole shape based on the detected apertures. However, this method was limited in that it required simultaneously observing the top and bottom keyhole apertures using two cameras. In this study, we proposed a novel deep-learning-based method to monitor the laser-beam absorptance in a keyhole using only one camera during laser keyhole welding of Al 5052-H32 alloy. In this method, both the top and bottom keyhole apertures were simultaneously detected from the images coaxially obtained from the top side. Although part of the bottom apertures may be sometimes obscured when viewed from above, this study demonstrated that the predicted absorptance was accurate enough and sufficient for monitoring laser welding processes of aluminum alloys. Using the developed method, changes in welding mode and generation of welding defects were successfully detected
Sparse Multi-Channel Convolutional Neural Network for Multivariate Time Series Classification
No full textIrregularly sampled time series classification using neural stochastic differential equation
No full textBayesian-based uncertainty-aware tool-wear prediction model in end-milling process of titanium alloy
No full textTool wear negatively affects machined surfaces and causes surface cracking, therefore increasing manufacturing costs and degrading product quality. Titanium alloys, which are widely used because of their desirable mechanical properties, have problems associated with tool wear due to poor thermal properties, such as specific heat capacity and thermal conductivity. Therefore, the accurate prediction of tool wear is necessary during the titanium alloy end-milling process to improve product quality and ensure reliability for corrective decisions like tool replacement. To this end, uncertainty-aware tool-wear prediction should be performed. In this study, a deep learning-based tool-wear prediction model based on a Bayesian approach was proposed. First, a convolutional neural network (CNN)-based architecture that integrates multiscale information extracted from raw sensor measurement data, termed deep multiscale CNN (DMSCNN), was proposed. It used different-sized convolutional kernels in parallel to enable various receptive field sizes suitable for machining processes. Second, based on a Bayesian learning approach, DMSCNN was transformed into a probabilistic model that produced a predictive distribution for estimated tool wear. In particular, a variational inference was applied to DMSCNN parameters to provide uncertainty awareness. Experiments were conducted with data collected from an actual end-milling process under three different conditions. The results proved the effectiveness of the proposed DMSCNN for tool-wear prediction. Bayesian DMSCNN showed promising results, as it outperformed existing comparative deterministic methods as well as probabilistic methods for tool-wear prediction. The proposed method is expected to be effectively applied in smart manufacturing as well as other machining processes that require data-driven digital decisions
Symmetry-Mismatched SBU Transformation in MOFs: Postsynthetic Metal Exchange from Zn to Fe and Its Effects on Gas Adsorption and Dye Selectivity
No full textThis research explores the alteration of metal???organic frameworks (MOFs) using a method called postsynthetic metal exchange. We focus on the shift from a Zn-based MOF containing a [Zn4O(COO)6] secondary building unit (SBU) of octahedral site symmetry (ANT-1(Zn)) to a Fe-based one with a [Fe3IIIO(COO)6]+ SBU of trigonal prismatic site symmetry (ANT-1(Fe)). The symmetry-mismatched SBU transformation cleverly maintains the MOF???s overall structure by adjusting the conformation of the flexible 1,3,5-benzenetribenzoate linker to alleviate the framework strain. The process triggers a decrease in the framework volume and pore size alongside a change in the framework???s charge. These alterations influence the MOF???s ability to adsorb gas and dye. During the transformation, core???shell MOFs (ANT-1(Zn@Fe)) are formed as intermediate products, demonstrating unique gas sorption traits and adjusted dye adsorption preferences due to the structural modifications at the core???shell interface. Heteronuclear clusters, located at the framework interfaces, enhance the heat of CO2 adsorption. Furthermore, they also influence the selectivity of the dye size. This research provides valuable insights into fabricating novel MOFs with unique properties by modifying the SBU of a MOF with flexible organic linkers from one site symmetry to another