检索结果-内蒙古大学图书馆

63rd IEEE Conference on Decision and Control, CDC 2024

作者： Graffione, S. Bracco, S. Pasquale, C. Sacone, S. Siri, S. Ferrara, A. University of Genova Department of Informatics Bioengineering Robotics and Systems Engineering Italy University of Genova Department of Electrical Electronics and Telecommunication Engineering and Naval Architecture Italy University of Pavia Department of Electrical Computer and Biomedical Engineering Italy

ISBN: (纸本)9798350316339

This paper proposes a multi-scale control scheme aimed at controlling speed, dwell time, and charging time of electric and automated buses performing the transport service on lines where there are no reserved lanes. The proposed control scheme has two layers characterized by different control objectives, level of detail and time scales. The high-level controller periodically solves a multi-objective optimal control problem formalized to ensure an adequate transport service, while the low-level controller, being based on a more accurate description of the bus dynamics and energy consumption, seeks to follow the high-level control actions while taking into account the effective traffic conditions encountered on the line and the effective dynamic behaviour of the bus. The paper concludes with the application to a case study in order to demonstrate the effectiveness of the proposed control scheme. © 2024 IEEE.

关键词： Bus transportation

来源：评论

学校读者我要写书评

暂无评论

Evaluation of 3D Reconstruction Techniques for the Blending of Real and Virtual Environments

Evaluation of 3D Reconstruction Techniques for the Blending ...

引用

Virtual Reality and 3D User Interfaces Abstracts and Workshops (VRW), IEEE Conference on

作者： Marianna Pizzo Eros Viola Fabio Solari Manuela Chessa Department of Informatics Bioengineering Robotics and Systems Engineering University of Genoa Italy

ISBN: (数字)9798350374490

ISBN: (纸本)9798350374506

Interaction in immersive virtual reality can be enhanced by merging physical objects with virtual stimuli, thus creating extended reality scenarios. Such approaches could be exploited to implement physi-cal obstacle avoidance and to provide the missing haptic feedback when touching, grasping, and manipulating objects. Effective solutions rely on the correct detection and tracking of the 6DOF pose of the objects and the accurate 3D reconstruction of the physical elements. Here, we focus on the latter aspect, analyzing different 3D reconstruction techniques to create and segment the meshes necessary to merge the physical objects with virtual counterparts. We consider six 3D reconstruction software and six different fur-nished rooms, analyzing the time necessary to create the 3D mesh from scratch, the reconstruction success rate, and the percentage of volume shift between the real objects and the reconstructed ones. The quantitative results show several off-the-shelf 3D reconstruction techniques that can be used as input for the considered extended reality system. Although fine 3D reconstruction and meshes are not necessary, the considered techniques can catch the sizes of the real objects in the scene and can be used to create virtual environments coherently aligned with the real ones.

关键词： Three-dimensional displays Extended reality Conferences Merging Virtual environments Grasping User interfaces

来源：评论

学校读者我要写书评

暂无评论

Exploiting LLMs for E-Learning: A Cybersecurity Perspective on AI-Generated Tools in Education

Exploiting LLMs for E-Learning: A Cybersecurity Perspective ...

引用

2024 IEEE International Workshop on Technologies for Defense and Security, TechDefense 2024

作者： Greco, Danilo Chianese, Luca Economics and Industrial Engineering Politecnico di Milano DIG - Department of Management Milan20156 Italy Università degli Studi di Genova Bioengineering Robotics and Systems Engineering DIBRIS - Department of Informatics Genoa16146 Italy

ISBN: (纸本)9798331505585

AI and LLM technologies have emerged rapidly, creating more opportunities to develop and enhance e-learning. This paper provides a detailed account of the tools built by the current and future generations of LLMs and AI technologies for creating complex e-learning applications. It introduces the application of these tools, including innovative learning, context-awareness and intelligent learning systems. Moreover, it delves systematically into the security issues of these AI-integrated e-learning environments. The work applies strict research approaches to determine and describe possible risks in system design, data processing, and AI model deployment. We also undertake deeper penetration tests to assess the level of preparedness of these platforms against different types of attacks from the attackers such as attacks targeting the AI components, injection of unverified data and privacy violations. Our study findings reveal the opportunity to apply AI in e-learning, thus highlighting the importance of addressing these concerns. It is a complete guide for the developers and the educational institutions to integrate the LLMs and AI and keep the security strong. This paper adds to the existing literature on the role of AI in education and cybersecurity, providing essential information to the future advancements of safe and innovative e-learning systems. © 2024 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Real-Time Planning of Platoons Coordination Decisions Based on Traffic Prediction

Real-Time Planning of Platoons Coordination Decisions Based ...

引用

European Control Conference (ECC)

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

ISBN: (数字)9783907144107

ISBN: (纸本)9798331540920

This paper deals with the optimal speed control of two platoons that share part of their freeway routes and need to perform merging and diverging procedures. The proposed control scheme has a centralized nature and is applied periodically by the platoons coordinator which receives in real time the state of the platoons and the traffic measurements on the network, based on which a traffic prediction is made. Based on this information, the coordinator applies specific optimal control algorithms to decide whether the merger is convenient or not and to compute the optimal speed profiles of the two platoons before the merging, during the shared journey and after the diverging phase.

关键词： Corporate acquisitions Merging Velocity control Optimal control Europe Traffic control Prediction algorithms

来源：评论

学校读者我要写书评

暂无评论

A planning algorithm for dispatching and smart charging operations of autonomous delivery robots in pedestrian areas

A planning algorithm for dispatching and smart charging oper...

引用

IEEE International Conference on Automation Science and engineering (CASE)

作者： V. Bachi S. Graffione C. Pasquale S. Siri Department of Informatics Bioengineering Robotics and Systems Engineering University of Genova Italy

ISBN: (数字)9798350358513

ISBN: (纸本)9798350358520

This paper considers a last-mile delivery problem in which autonomous robots are employed for transporting parcels in a pedestrian area. The deliveries to the different buildings are organized through a common collection point, in which couriers unload the parcels and from which a set of autonomous robots move to deliver them to the final customers. The common collection point is located in a building provided with renewable sources and a battery storage and controlled with an energy management system. The proposed planning algorithm acts according to a discrete-time logic, in which the dispatching and charging operations of the robots are planned at each time step. The main objectives are the full exploitation of the robot capacity, the reduction of the delivery delay, the minimization of the travel costs and the smart energy exploitation. This algorithm is tested on a real case study in the Savona Campus of the University of Genova.

关键词： Renewable energy sources Pedestrians Costs Buildings Minimization Dispatching Planning Delays Autonomous robots Smart charging

来源：评论

学校读者我要写书评

暂无评论

Using Visual Feedbacks in an Augmented Video-based Learning Tool 28

Using Visual Feedbacks in an Augmented Video-based Learning ...

引用

28th International DMS Conference on Visualization and Visual Languages, DMSVIVA 2022

作者： Coccoli, Mauro Galluccio, Ilenia Torre, Ilaria Department of Informatics Bioengineering Robotics and Systems Science University of Genoa Italy

ISBN: (纸本)1891706551

In this paper, we face the problem of using video-based learning with multimedia content, which is expected to assume a prominent role in the post-pandemic world. In this respect, we have investigated the possibility of developing new services with suited visual interfaces, to further exploit its potential. Such novel services want to integrate the knowledge extracted from multimedia materials into educational applications. To do this, our approach is that of extracting the theoretical concepts included in a video lesson and describing the prerequisites relations between them, according to the knowledge base of the subject matter of the video itself. Such an addition of knowledge allows creating augmented video lessons and providing learners with new methods to browse videos and perform a non-linear navigation of their learning materials, by means of visual feedback methods. To this aim, we designed a custom web-based video player to support video-based learning, implementing these ideas. From the relevant literature we argue that this approach can be effective from an educational point of view but such effectiveness can be achieved only if the proposed video player is an easy-to-use tool, thus, we made a preliminary evaluation to assess the usability of the proposed system and results are presented. © 2022 by KSI Research Inc. and Knowledge systems Institute, USA.

关键词： Knowledge based systems

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

2022 IEEE Nuclear Science Symposium, Medical Imaging Conference, and Room Temperature Semiconductor Detector Conference, IEEE NSS MIC RTSD 2022

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

ISBN: (纸本)9781665488723

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. © 2022 IEEE.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Robustness-Congruent Adversarial Training for Secure Machine Learning Model Updates

引用

IEEE Transactions on Pattern Analysis and Machine Intelligence 2025年 PP卷 PP页

作者： Angioni, Daniele Demetrio, Luca Pintor, Maura Oneto, Luca Anguita, Davide Biggio, Battista Roli, Fabio University of Cagliari Department of Electrical and Electronic Engineering Italy University of Genova Department of Informatics Bioengineering Robotics and Systems Engineering Italy

Machine-learning models demand periodic updates to improve their average accuracy, exploiting novel architectures and additional data. However, a newly updated model may commit mistakes that the previous model did not make. Such misclassifications are referred to as negative flips, experienced by users as a regression of performance. In this work, we show that this problem also affects robustness to adversarial examples, hindering the development of secure model update practices. In particular, when updating a model to improve its adversarial robustness, previously ineffective adversarial attacks on some inputs may become successful, causing a regression in the perceived security of the system. We propose a novel technique, named robustness-congruent adversarial training, to address this issue. It amounts to fine-tuning a model with adversarial training, while constraining it to retain higher robustness on the samples for which no adversarial example was found before update. We show that our algorithm and, more generally, learning with non-regression constraints, provides a theoretically-grounded framework to train consistent estimators. Our experiments on robust models for computer vision confirm that both accuracy and robustness, even if improved after model update, can be affected by negative flips, and our robustness-congruent adversarial training can mitigate the problem, outperforming competing baseline methods. © 1979-2012 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Optimal control of electric automated buses in intercity lines: scenario-based analysis of the Pareto-optimal solutions

引用

IFAC-PapersOnLine 2024年第10期58卷 37-42页

作者： C. Pasquale S. Sacone S. Siri A. Ferrara Department of Informatics Bioengineering Robotics and Systems Engineering University of Genova Italy Department of Electrical Computer and Biomedical Engineering University of Pavia Italy

This paper deals with the optimal control of electric and automated buses that have to follow an intercity line, where some stops are equipped with a flash-charging infrastructure to charge the batteries, while others are not. In order to control these buses, it is necessary to account for the traffic conditions along the roads of the bus line, and to minimize two objectives related to the deviations from the bus timetable and the lack of energy, at the end of the bus trip, with respect to a desired final energy level. The resulting optimal control problem is a multi-objective linear quadratic problem with both continuous and integer variables. In this paper, we propose and apply two methods based on lexicographic ordering aimed at determining the Pareto-optimal solutions of the control problem in twelve scenarios inspired by a real case.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Grounding Conversational Robots on Vision Through Dense Captioning and Large Language Models

Grounding Conversational Robots on Vision Through Dense Capt...

引用

IEEE International Conference on robotics and Automation (ICRA)

作者： Lucrezia Grassi Zhouyang Hong Carmine Tommaso Recchiuto Antonio Sgorbissa Department of Informatics Bioengineering Robotics and Systems Engineering University of Genoa Genoa Italy

ISBN: (数字)9798350384574

ISBN: (纸本)9798350384581

This work explores a novel approach to empowering robots with visual perception capabilities using textual descriptions. Our approach involves the integration of GPT-4 with dense captioning, enabling robots to perceive and interpret the visual world through detailed text-based descriptions. To assess both user experience and the technical feasibility of this approach, experiments were conducted with human participants interacting with a Pepper robot equipped with visual capabilities. The results affirm the viability of the proposed approach, allowing to perform vision-based conversations effectively, despite processing time limitations.

关键词： Visualization Face recognition Oral communication User experience Delays Face detection Time factors

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

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

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：