检索结果-内蒙古大学图书馆

25th International Conference on Pattern Recognition Workshops, ICPR 2020

作者： Her, Paris Manderle, Logan Dias, Philipe A. Medeiros, Henry Odone, Francesca Department of Electrical and Computer Engineering Marquette University MilwaukeeWI United States Department of Informatics Bioengineering Robotics and Systems Engineering University of Genoa Genoa Italy

ISBN: (纸本)9783030687892

Effective assisted living environments must be able to perform inferences on how their occupants interact with their environment. Gaze direction provides strong indications of how people interact with their surroundings. In this paper, we propose a gaze tracking method that uses a neural network regressor to estimate gazes from keypoints and integrates them over time using a moving average mechanism. Our gaze regression model uses confidence gated units to handle cases of keypoint occlusion and estimate its own prediction uncertainty. Our temporal approach for gaze tracking incorporates these prediction uncertainties as weights in the moving average scheme. Experimental results on a dataset collected in an assisted living facility demonstrate that our gaze regression network performs on par with a complex, dataset-specific baseline, while its uncertainty predictions are highly correlated with the actual angular error of corresponding estimations. Finally, experiments on videos sequences show that our temporal approach generates more accurate and stable gaze predictions. © Springer Nature Switzerland AG 2021.

关键词： Assisted living

来源：评论

学校读者我要写书评

暂无评论

A Bi-level Optimization-based Architecture for the Scheduling and Real-Time Control of Microgrids With Hydrogen Production System

引用

IFAC-PapersOnLine 2023年第2期56卷 8284-8289页

作者： Daria Bellotti Yassine Ennassiri Giulio Ferro Loredana Magistri Michela Robba Thermochemical Power Group (TPG) DIME University of Genoa Via Montallegro 1 16145 Genoa Italy DIBRIS—Department of Informatics Bioengineering Robotics and Systems Engineering University of Genova 16145 Italy

This paper proposes a novel bi-level architecture for the scheduling and real-time control of a microgrid with a hydrogen production system. It consists of an economic optimization that performs the operational management of the microgrid at the higher level. At the lower level, a real-time controller, based on a Reference Governor (RG), is designed to control the stack's temperature of a proton exchange membrane (PEM) electrolyzer under the flexible operation imposed by the economic optimization. When switching from one power level to another, the stack's temperature is having a sudden change that might lead to a violation of the operating limits set by the manufacturer. With the RG-based controller added at the lower level, the operating limits of the temperature are respected, and better performances of the system are guaranteed. The effectiveness of the proposed bi-level architecture has been proved through a real case study, in which the sudden changes in the temperatures have significantly been reduced by up to 91% with respect to the optimal operational management of the overall microgrid.

关键词： Optimal control Real-time optimal control Electrolyzer Fuel cell Energy systems

来源：评论

学校读者我要写书评

暂无评论

Path Tracking for Wheeled Mobile Robot Using Non Linear Model Predictive Control in Indoor Environment

Path Tracking for Wheeled Mobile Robot Using Non Linear Mode...

引用

International Conference on Control, Decision and Information Technologies (CoDIT)

作者： Alessandro Bozzi Simone Graffione Michiel M.W. Kockelkoren Roberto Sacile Enrico Zero Department of Informatics Bioengineering Robotics and Systems Engineering (DIBRIS) University of Genoa Genoa Italy Delft Center for Systems and Control (DCSC) Delft University of Technology Delft Netherlands

Wheeled Mobile Robots (WMR) with the assistance of Information Communication Technology (ICT) can navigate and perform some tasks in an uncontrolled environment. They can experience some problems in task management if they can not localize themself in indoor/outdoor environments. Different algorithms can be exploited to evaluate the proper control law to track a default path. In this paper, a path-tracking algorithm for a WMR has been tested for an indoor environment, with a delimited area spanned by localization tags. The WMR must check some waypoints during its path until the last one, where the WMR re-start the simulation. In this work, the control law applied for the experiments is the Non-Linear Model Predictive Controller (NMPC). The control algorithm is tested using two experiments, the first one is based on the robot movements in a simulation environment using the Gazebo tools, and the second one is related to a real context where the mobile robot moves in an indoor environment.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Robotic in-hand manipulation with relaxed optimization

Robotic in-hand manipulation with relaxed optimization

引用

IEEE International Workshop on Robot and Human Communication (ROMAN)

作者： Ali Hammoud Valerio Belcamino Quentin Huet Alessandro Carfi Mahdi Khoramshahi Veronique Perdereau Fulvio Mastrogiovanni Sorbonne Universite Institut des Systemes Intelligents et de Robotique ISIR Paris France The EngineRoom Department of Informatics Bioengineering Robotics and Systems Engineering University of Genoa Genoa Italy

ISBN: (数字)9798350375022

ISBN: (纸本)9798350375039

Dexterous in-hand manipulation is a unique and valuable human skill requiring sophisticated sensorimotor interaction with the environment while respecting stability constraints. Satisfying these constraints with generated motions is essential for a robotic platform to achieve reliable in-hand manipulation skills. Explicitly modelling these constraints can be challenging, but they can be implicitly modelled and learned through experience or human demonstrations. We propose a learning and control approach based on dictionaries of motion primitives generated from human demonstrations. To achieve this, we defined an optimization process that combines motion primitives to generate robot fingertip trajectories for moving an object from an initial to a desired final pose. Based on our experiments, our approach allows a robotic hand to handle objects like humans, adhering to stability constraints without requiring explicit formalization. In other words, the proposed motion primitive dictionaries learn and implicitly embed the constraints crucial to the in-hand manipulation task.

关键词： Training Dictionaries Human anatomy Kinematics Robot sensing systems Stability analysis Trajectory Reliability Optimization

来源：评论

学校读者我要写书评

暂无评论

Runtime Verification for Trustworthy Computing

arXiv

引用

arXiv 2023年

作者： Abela, Robert Colombo, Christian Curmi, Axel Fenech, Mattea Vella, Mark Ferrando, Angelo Department of Computer Science Faculty of ICT University of Malta Msida Malta Department of Informatics Bioengineering Robotics and Systems Engineering University of Genoa Genova16145 Italy

Autonomous and robotic systems are increasingly being trusted with sensitive activities with potentially serious consequences if that trust is broken. Runtime verification techniques present a natural source of inspiration for monitoring and enforcing the desirable properties of the communication protocols in place, providing a formal basis and ways to limit intrusiveness. A recently proposed approach, RV-TEE, shows how runtime verification can enhance the level of trust to the Rich Execution Environment (REE), consequently adding a further layer of protection around the Trusted Execution Environment (TEE). By reflecting on the implication of deploying RV in the context of trustworthy computing, we propose practical solutions to two threat models for the RV-TEE monitoring process: one where the adversary has gained access to the system without elevated privileges, and another where the adversary gains all privileges to the host system but fails to steal secrets from the TEE. © 2023, CC BY.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Discrete-Time Model for Large-Scale Multi-Modal Transport Networks

引用

IFAC-PapersOnLine 2021年第2期54卷 7-12页

作者： C. Pasquale E. Siri S. Sacone S. Siri Department of Informatics Bioengineering Robotics and Systems Engineering University of Genova Italy

The purpose of this paper is to provide a method that can be used to perform resilience analyses on large-scale multi-modal transportation networks. In particular, this work proposes an assignment model through which the allocation of mobility demand on a multimodal transport network is defined at a regional or supra-regional level. The results of the assignment problem are then used as input data of a macroscopic dynamic model specifically developed to represent the dynamics of a multi-modal transport network. Finally, the proposed methodology is applied to represent the behavior of users in the Nguyen-Dupuis test network.

关键词： Multi-modal transport network multi-modal assignment discrete-time network model

来源：评论

学校读者我要写书评

暂无评论

3D Human-Body Modeling for Medical Simulation: Usability and Robustness of Photogrammetry Techniques

3D Human-Body Modeling for Medical Simulation: Usability and...

引用

IEEE Symposium on Nuclear Science (NSS/MIC)

作者： Patrizia Boccacci Manuela Chessa Marco La Iacona Dept. of Informatics Bioengineering Robotics and Systems Engineering University of Genoa Italy

Total skin electron therapy TSET is mainly used in the treatment of mycosis fungoides, a rare cutaneous T-cell lymphoma affecting the first few millimeters of skin. The Skinscan project aims to optimize irradiation by creating a 3D model of the patient’s body surface. This would avoid overdosing or underdosing in different skin areas compared to the prescribed dose and would reduce the inhomogeneity of dose distribution associated with current techniques used during TSET therapy. In SkinScan, we investigate the creation of 3D models of the human body, analyzing the compromise between usability of the software and robustness of the techniques. State-of-the-art techniques for the creation of the 3D models are RGBD and photogrammetry. The first ones require dedicated devices, the other are based on standard photographs. Commercial software to align the different views and to create the models are available, and they can be used by non-experts. The final application would indeed require the creation of ad-hoc models for a personalized medical intervention on the patients. Such models should be created in hospitals, without the intervention of computer scientists. Moreover, accurate 3D modeling often requires specific post-processing to remove noise without adding artifacts on the surfaces. We present the preliminary results to create the 3D models to be used as the input of the SkinScan simulation pipeline. We modeled a flat wall and a box composed of flat surfaces, to evaluate the amount of noise in the 3D reconstruction and the pre-processing of the images to remove it. We then modeled two different body parts, a hand and a shoulder, with the aim of understanding the best compromise between the accuracy of the model and its size, thus the simulation time.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Adversarial Attacks Against Uncertainty Quantification

arXiv

引用

arXiv 2023年

作者： Ledda, Emanuele Angioni, Daniele Piras, Giorgio Fumera, Giorgio Biggio, Battista Roli, Fabio Department of Computer Control and Management Engineering Sapienza University of Rome Italy Department of Electric and Electronic Engineering University of Cagliari Italy Department of Informatics Bioengineering Robotics and Systems Engineering University of Genova Italy

Machine-learning models can be fooled by adversarial examples, i.e., carefully-crafted input perturbations that force models to output wrong predictions. While uncertainty quantification has been recently proposed to detect adversarial inputs, under the assumption that such attacks exhibit a higher prediction uncertainty than pristine data, it has been shown that adaptive attacks specifically aimed at reducing also the uncertainty estimate can easily bypass this defense mechanism. In this work, we focus on a different adversarial scenario in which the attacker is still interested in manipulating the uncertainty estimate, but regardless of the correctness of the prediction;in particular, the goal is to undermine the use of machine-learning models when their outputs are consumed by a downstream module or by a human operator. Following such direction, we: (i) design a threat model for attacks targeting uncertainty quantification;(ii) devise different attack strategies on conceptually different UQ techniques spanning for both classification and semantic segmentation problems;(iii) conduct a first complete and extensive analysis to compare the differences between some of the most employed UQ approaches under attack. Our extensive experimental analysis shows that our attacks are more effective in manipulating uncertainty quantification measures than attacks aimed to also induce misclassifications. © 2023, CC BY.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Adversarial Attacks Against Uncertainty Quantification

Adversarial Attacks Against Uncertainty Quantification

引用

International Conference on Computer Vision Workshops (ICCV Workshops)

作者： Emanuele Ledda Daniele Angioni Giorgio Piras Giorgio Fumera Battista Biggio Fabio Roli Department of Computer Control and Management Engineering Sapienza University of Rome Italy Department of Informatics Bioengineering Robotics and Systems Engineering University of Genova Italy Department of Electric and Electronic Engineering University of Cagliari Italy

Machine-learning models can be fooled by adversarial examples, i.e., carefully-crafted input perturbations that force models to output wrong predictions. While uncertainty quantification has been recently proposed to detect adversarial inputs, under the assumption that such attacks exhibit a higher prediction uncertainty than pristine data, it has been shown that adaptive attacks specifically aimed at reducing also the uncertainty estimate can easily bypass this defense mechanism. In this work, we focus on a different adversarial scenario in which the attacker is still interested in manipulating the uncertainty estimate, but regardless of the correctness of the prediction; in particular, the goal is to undermine the use of machine-learning models when their outputs are consumed by a downstream module or by a human operator. Following such direction, we: (i) design a threat model for attacks targeting uncertainty quantification; (ii) devise different attack strategies on conceptually different UQ techniques spanning for both classification and semantic segmentation problems; (iii) conduct a first complete and extensive analysis to compare the differences between some of the most employed UQ approaches under attack. Our extensive experimental analysis shows that our attacks are more effective in manipulating uncertainty quantification measures than attacks aimed to also induce misclassifications.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Robot-Assisted Navigation for Visually Impaired through Adaptive Impedance and Path Planning

Robot-Assisted Navigation for Visually Impaired through Adap...

引用

IEEE International Conference on robotics and Automation (ICRA)

作者： Pietro Balatti Idil Ozdamar Doganay Sirintuna Luca Fortini Mattia Leonori Juan M. Gandarias Arash Ajoudani HRI2 Lab Istituto Italiano di Tecnologia Genoa Italy Dept. of Informatics Bioengineering Robotics and System Engineering University of Genoa Genoa Italy Systems Engineering and Automation Department Robotics and Mechatronics Lab University of Malaga Malaga Spain

ISBN: (数字)9798350384574

ISBN: (纸本)9798350384581

This paper presents a framework to navigate visually impaired people through unfamiliar environments by means of a mobile manipulator. The Human-Robot system consists of three key components: a mobile base, a robotic arm, and the human subject who gets guided by the robotic arm via physically coupling their hand with the cobot’s end-effector. These components, receiving a goal from the user, traverse a collision-free set of waypoints in a coordinated manner, while avoiding static and dynamic obstacles through an obstacle avoidance unit and a novel human guidance planner. With this aim, we also present a legs tracking algorithm that utilizes 2D LiDAR sensors integrated into the mobile base to monitor the human pose. Additionally, we introduce an adaptive pulling planner responsible for guiding the individual back to the intended path if they veer off course. This is achieved by establishing a target arm end-effector position and dynamically adjusting the impedance parameters in real-time through a impedance tuning unit. To validate the framework we present a set of experiments both in laboratory settings with 12 healthy blindfolded subjects and a proof-of-concept demonstration in a real-world scenario.

关键词： Target tracking Navigation Robot kinematics End effectors Real-time systems Path planning Sensors

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：