With the development of continuous fiber-reinforced composites (CFRCs) 3D printing technology, timely, efficient and accurate detection of fiber path defects is essential for ensuring product quality and performance. ...
详细信息
The emergency department (ED) is a safety-critical environment in which healthcare workers (HCWs) are overburdened, overworked, and have limited resources, especially during the COVID-19 pandemic. One way to address t...
详细信息
ISBN:
(纸本)9781728190778
The emergency department (ED) is a safety-critical environment in which healthcare workers (HCWs) are overburdened, overworked, and have limited resources, especially during the COVID-19 pandemic. One way to address this problem is to explore the use of robots that can support clinical teams, e.g., to deliver materials or restock supplies. However, due to EDs being overcrowded, and the cognitive overload HCWs experience, robots need to understand various levels of patient acuity so they avoid disrupting care delivery. In this paper, we introduce the Safety-Critical Deep Q-Network (SafeDQN) system, a new acuity-aware navigation system for mobile robots. SafeDQN is based on two insights about care in EDs: high-acuity patients tend to have more HCWs in attendance and those HCWs tend to move more quickly. We compared SafeDQN to three classic navigation methods, and show that it generates the safest, quickest path for mobile robots when navigating in a simulated ED environment. We hope this work encourages future exploration of social robots that work in safety-critical, human-centered environments, and ultimately help to improve patient outcomes and save lives.
Object tracking is central to robot perception and scene understanding, allowing robots to parse a video stream in terms of moving objects with names. Tracking-by-detection has long been a dominant paradigm for object...
详细信息
ISBN:
(数字)9798350384574
ISBN:
(纸本)9798350384581
Object tracking is central to robot perception and scene understanding, allowing robots to parse a video stream in terms of moving objects with names. Tracking-by-detection has long been a dominant paradigm for object tracking of specific object categories [1], [2]. Recently, large-scale pre-trained models have shown promising advances in detecting and segmenting objects and parts in 2D static images in the wild. This raises the question: can we re-purpose these large-scale pre-trained static image models for open-vocabulary video tracking? In this paper, we combine an open-vocabulary detector [3], segmenter [4], and dense optical flow estimator [5], into a model that tracks and segments any object in 2D videos. Given a monocular video input, our method predicts object and part mask tracks with associated language descriptions, rebuilding the pipeline of Tractor [6] with modern large pre-trained models for static image detection and segmentation: we detect open-vocabulary object instances and propagate their boxes from frame to frame using a flow-based motion model, refine the propagated boxes with the box regression module of the visual detector, and prompt an open-world segmenter with the refined box to segment the objects. We decide the termination of an object track based on the objectness score of the propagated boxes as well as forward-backward optical flow consistency. We re-identify objects across occlusions using deep feature matching. We show that our model achieves strong performance on multiple established benchmarks [7], [8], [9], [10], and can produce reasonable tracks in manipulation data [11]. In particular, our model outperforms previous state-of-the-art in UVO and BURST, benchmarks for open-world object tracking and segmentation, despite never being explicitly trained for tracking. We hope that our approach can serve as a simple and extensible framework for future research and enable imitation learning from videos with unconventional objects.
In this paper, a nonlinear model predictive control considering vehicle jerk dynamics is proposed for improving ride comfort of passengers. Since the vehicle model in prediction phase requires high accuracy dynamics i...
详细信息
In this paper, a nonlinear model predictive control considering vehicle jerk dynamics is proposed for improving ride comfort of passengers. Since the vehicle model in prediction phase requires high accuracy dynamics in order to handle the jerk motion, the approximated wheel load transfer dynamics is introduced. Also, to obtain the control capability for not only jerk but also acceleration, velocity and position, the expanded state space model including these dimensions into the one has been developed. It improves the utility as autonomous vehicle controller. By numerical simulation in assuming cornering driving scene, the effectiveness that jerk and other vehicle states enable to constraint simultaneously by individual torque distribution by electric power train is validated. Further, the principle of the optimized torque distribution by proposed method is analyzed.
A farmland intelligent monitoring and irrigation system based on the Internet of Things and 5G technology is designed. The system mainly includes two templates: intelligent monitoring system and intelligent control sy...
A farmland intelligent monitoring and irrigation system based on the Internet of Things and 5G technology is designed. The system mainly includes two templates: intelligent monitoring system and intelligent control system, in which the intelligent monitoring system is mainly responsible for detecting soil temperature and humidity and trace elements such as nitrogen, phosphorus and potassium, and the intelligent control system is mainly responsible for remote operation of intelligent irrigation. The system measures soil temperature, humidity and trace elements such as nitrogen, phosphorus and potassium through the sensors in the intelligent monitoring section, and uses 5G wireless transmission technology to send the data collected by the sensors to the Internet of Things, which then transmits the collected information to a mobile app. The system can solve the shortcomings of traditional irrigation methods which are cumbersome and seriously inefficient in irrigating agricultural facilities.
As part of this research, software for multimodal biometric authentication using neural networks is developed to improve the efficiency of information system user authorization. The architectures of artificial neural ...
As part of this research, software for multimodal biometric authentication using neural networks is developed to improve the efficiency of information system user authorization. The architectures of artificial neural networks, which are involved in the processes of recognizing a person by facial image and voice, are given. The internationally used databases (DataSet) of images and audio recordings for training of neural networks are considered. The process of training neural networks, the formed database of biometric personal data and the results obtained by the authors are described.
Lovable robots in movies regularly beep, chirp, and whirr, yet robots in the real world rarely deploy such sounds. Despite preliminary work supporting the perceptual and objective benefits of intentionally-produced ro...
详细信息
ISBN:
(纸本)9781728190778
Lovable robots in movies regularly beep, chirp, and whirr, yet robots in the real world rarely deploy such sounds. Despite preliminary work supporting the perceptual and objective benefits of intentionally-produced robot sound, relatively little research is ongoing in this area. In this paper, we systematically evaluate transformative robot sound across multiple robot archetypes and behaviors. We conducted a series of five online video-based surveys, each with N approximate to 100 participants, to better understand the effects of musician-designed transformative sounds on perceptions of personal, service, and industrial robots. Participants rated robot videos with transformative sound as significantly happier, warmer, and more competent in all five studies, as more energetic in four studies, and as less discomforting in one study. Overall, results confirmed that transformative sounds consistently improve subjective ratings but may convey affect contrary to the intent of affective robot behaviors. In future work, we will investigate the repeatability of these results through in-person studies and develop methods to automatically generate transformative robot sound. This work may benefit researchers and designers who aim to make robots more favorable to human users.
Given its wide application in robotics, point cloud registration is a widely researched topic. Conventional methods aim to find a rotation and translation that align two point clouds in 6 degrees of freedom (DoF). How...
详细信息
ISBN:
(数字)9798350384574
ISBN:
(纸本)9798350384581
Given its wide application in robotics, point cloud registration is a widely researched topic. Conventional methods aim to find a rotation and translation that align two point clouds in 6 degrees of freedom (DoF). However, certain tasks in robotics, such as category-level pose estimation, involve non-uniformly scaled point clouds, requiring a 9DoF transform for accurate alignment. We propose HEGN, a novel equivariant graph neural network for 9DoF point cloud registration. HEGN utilizes equivariance to rotation, translation, and scaling to estimate the transformation without relying on point correspondences. Based on graph representations for both point clouds, we extract equivariant node features aggregated in their local, cross-, and global context. In addition, we introduce a novel node pooling mechanism that leverages the cross-context importance of nodes to pool the graph representation. By repeating the feature extraction and node pooling, we obtain a graph hierarchy. Finally, we determine rotation and translation by aligning equivariant features aggregated over the graph hierarchy. To estimate scaling, we leverage scale information in the vector norm of the equivariant features. We evaluate the effectiveness of HEGN through experiments with the synthetic ModelNet40 dataset and the real-world ScanObjectNN dataset. The results show the superior performance of HEGN in 9DoF point cloud registration and its competitive performance in conventional 6DoF point cloud registration.
This study explores various techniques for transforming 1-dimensional time-series data into 2-dimensional images, preparing for the application of machine learning models designed for 2D data. Eight distinct methods a...
This study explores various techniques for transforming 1-dimensional time-series data into 2-dimensional images, preparing for the application of machine learning models designed for 2D data. Eight distinct methods are introduced, including recurrence plots, Markov transition, Gramian angular field, spectrogram, heatmap, direct plot, phase space transformation, and Poincaré plots. These methods are tested using data from a modeled photovoltaic (PV) grid-connected system, specifically simulating a shorted string fault and a no-fault condition. The fault and no-fault responses are captured with a fixed window size of 256 sample points, consistently applied across all methods. All transformation method is tested through python 3 programming using a laptop with minimal computing capability. The generated image of each transformation may contain 1-channel image in grayscale or 3-channel RGB image. Dimension of the generated image can be increase or decrease during saving process. Each method produces a unique visual representation of the shorted string fault and a no-fault, demonstrating diverse perspectives in transforming 1D time-series data into 2D images for subsequent machine learning applications.
Deep Generative Machine Learning models have been growing in popularity across the design community thanks to their ability to learn and mimic complex data distributions. While early works are promising, further advan...
详细信息
ISBN:
(纸本)9780791886236
Deep Generative Machine Learning models have been growing in popularity across the design community thanks to their ability to learn and mimic complex data distributions. While early works are promising, further advancement will depend on addressing several critical considerations such as design quality, feasibility, novelty, and targeted inverse design. We propose the Design Target Achievement Index (DTAI), a differentiable, tunable metric that scores a design's ability to achieve designer-specified minimum performance targets. We demonstrate that DTAI can drastically improve the performance of generated designs when directly used as a training loss in Deep Generative models. We apply the DTAI loss to a Performance-Augmented Diverse GAN (PaDGAN) and demonstrate superior generative performance compared to a set of baseline Deep Generative models including a Multi-Objective PaDGAN and specialized tabular generation algorithms like the Conditional Tabular GAN (CTGAN). We further enhance PaDGAN with an auxiliary feasibility classifier to encourage feasible designs. To evaluate methods, we propose a comprehensive set of evaluation metrics for generative methods that focus on feasibility, diversity, and satisfaction of design performance targets. methods are tested on a challenging benchmarking problem: the FRAMED bicycle frame design dataset featuring mixed-datatype parametric data, heavily skewed and multimodal distributions, and ten competing performance objectives.
暂无评论