检索结果-内蒙古大学图书馆

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Lee, Younghwan Luu, Tung M. Lee, Donghoon Yoo, Chang D. Electrical Engineering KAIST Daejeon Korea Republic of Robotics Program KAIST Daejeon Korea Republic of

ISBN: (纸本)9798350368741

In offline reinforcement learning (RL), learning from fixed datasets presents a promising solution for domains where real-time interaction with the environment is expensive or risky. However, designing dense reward signals for offline dataset requires significant human effort and domain expertise. Reinforcement learning with human feedback (RLHF) has emerged as an alternative, but it remains costly due to the human-in-the-loop process, prompting interest in automated reward generation models. To address this, we propose Reward Generation via Large Vision-Language Models (RG-VLM), which leverages the reasoning capabilities of LVLMs to generate rewards from offline data without human involvement. RG-VLM improves generalization in long-horizon tasks and can be seamlessly integrated with the sparse reward signals to enhance task performance, demonstrating its potential as an auxiliary reward signal. © 2025 IEEE.

关键词： Large Vision-Language Model Offline Reinforcement Learning Reward Labeling

来源：评论

学校读者我要写书评

暂无评论

Smartphone-Based Orientation Estimation for Assisting the Visually Impaired in Lines 23

Smartphone-Based Orientation Estimation for Assisting the Vi...

引用

23rd International Conference on Machine Learning and Cybernetics, ICMLC 2024

作者： Tokuda, Haruto Bilguunmaa, Myagmardulam Takagi, Noboru Fuse, Yotato Motoyoshi, Tatsuo Sawai, K.E.I. Masuta, Hiroyuki Graduate School of Intelligent Robotics Program Toyama Prefectural University Japan Toyama Prefectural University Department of Data Science Japan Toyama Prefectural University Department of Intelligent Robotics Japan

ISBN: (纸本)9798331528041

Standing in line poses a significant challenge for the visually impaired. Existing queue guidance systems, which recognize line locations and estimate personal orientation, require highspecification computers, making them impractical for everyday use. This study proposes a smartphone-based queue guidance system that leverages the device's camera and sensors to estimate personal orientation efficiently. By optimizing computationally efficient algorithms for smartphone hardware, we aim to create an accessible, cost-effective solution. This system will enhance the independence and mobility of visually impaired individuals, providing real-time guidance to navigate queues seamlessly. Initial testing indicates promising usability and accuracy, paving the way for broader implementation. © 2024 IEEE.

关键词： Smartphones

来源：评论

学校读者我要写书评

暂无评论

Sample Efficient Reinforcement Learning via Large Vision Language Model Distillation

Sample Efficient Reinforcement Learning via Large Vision Lan...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Lee, Donghoon Luu, Tung M. Lee, Younghwan Yoo, Chang D. Robotics Program KAIST Daejeon Korea Republic of Electrical Engineering KAIST Daejeon Korea Republic of

ISBN: (纸本)9798350368741

Recent research highlights the potential of multimodal foundation models in tackling complex decision-making challenges. However, their large parameters make real-world deployment resource-intensive and often impractical for constrained systems. Reinforcement learning (RL) shows promise for task-specific agents but suffers from high sample complexity, limiting practical applications. To address these challenges, we introduce LVLM to Policy (LVLM2P), a novel framework that distills knowledge from large vision-language models (LVLM) into more efficient RL agents. Our approach leverages the LVLM as a teacher, providing instructional actions based on trajectories collected by the RL agent, which helps reduce less meaningful exploration in the early stages of learning, thereby significantly accelerating the agent's learning progress. Additionally, by leveraging the LVLM to suggest actions directly from visual observations, we eliminate the need for manual textual descriptors of the environment, enhancing applicability across diverse tasks. Experiments show that LVLM2P significantly enhances the sample efficiency of baseline RL algorithms. The code is available at https://***/i22024/LVLM2P. © 2025 IEEE.

关键词： Knowledge Distillation Reinforcement Learning Vision Language Model

来源：评论

学校读者我要写书评

暂无评论

Docking Assistance System Between Trailer and Tractor Using UWB Measuring Module

引用

Journal of Institute of Control, robotics and Systems 2024年第1期30卷 59-65页

作者： Park, Jong Hak Heo, Jeong Woong Moon, Hee Chang Industrial Convergence Interdepartmental Program Autonomous Vehicle･Intelligent Robotics The Graduate School of Hongik University Korea Republic of Mechanical and System Design Engineering Hongik University Korea Republic of

Trailers, which are used to load equipment such as containers and construction materials, are towed by the power of tractors to transport the equipment. The tractor, the towing vehicle, and the trailer, the towed vehicle, are independent systems, and the tractor can combine and separate by replacing the trailer for various purposes. To dock the trailer and tractor, the tractor should be backward with the posture and position of the rear of the tractor in line with the front of the trailer. The tractor driver shall be mounted based on the driver’s cognitive judgment, with a rear-view camera and mirror to determine the matching position of the trailer. We implemented and explored a system that could determine the length of this combination of systems using UWB, a positioning system, rather than driver recognition judgment, and used it to determine the position and posture of the trailer reference tractor using a square geometric approach. © ICROS 2024.

关键词： Light trailers

来源：评论

学校读者我要写书评

暂无评论

Micro-electrode array (MEA) electrophysiological signal analysis by descriptive stochastic tests of the probability density structure

引用

Research on Biomedical Engineering 2025年第2期41卷 1-13页

作者： Faria, Vinícius Naves Rezende Ramos, Marina Abadia Martinoia, Sérgio Destro-Filho, João-Batista Postgraduate Program in Electrical Engineering Universidade Federal de MG Uberlândia Brazil Postgraduate Program in Biomedical Sciences Universidade Federal de MG Uberlândia Brazil Division of Informatics Bioengineering Robotics and System Engineering Università Degli Studi Di Genova Genova Italy

Purpose: The primary objective of this study is to present, analyze, and quantitatively characterize the electrical signals obtained from neuronal cultures coupled to micro-electrode arrays (MEAs). Special attention is given to the temporal evolution of electrical activity across distinct geometric regions of the MEA devices. Methods: The proposed methodology involves signal acquisition, classical analysis based on spike and burst detection, and, critically, the characterization of the probability density structure of the signals. This is achieved through the quantification of Gaussianity—expressed as the percentage of segments non-Gaussian (PSNG)—and stationarity— expressed as the percentage of segments non-stationary (PSNE). Results: Initial analyses showed that first-order power spectral density quantifiers lacked the statistical robustness required to effectively describe the complex components of MEA signals. In contrast, PSNG and PSNE demonstrated strong potential in capturing the different stages of neuronal culture development. Cluster analysis further supported the discriminative capacity of these quantifiers. Conclusion: The phase corresponding to decreased neuronal activity and onset of cell death—a biologically critical stage—exhibited the strongest correlation among the studied variables. These results suggest a discernible biological organization and highlight the relevance of PSNG and PSNE as promising tools for assessing neuronal culture dynamics. © The Author(s), under exclusive licence to The Brazilian Society of Biomedical Engineering 2025.

关键词： Cell death

来源：评论

学校读者我要写书评

暂无评论

Design of an Automatic Price Tag System for Web-Based Retail Business 9

Design of an Automatic Price Tag System for Web-Based Retail...

引用

9th International Conference on Information Management and Technology, ICIMTech 2024

作者： Liawatimena, Suryadiputra Gunawan, Devina Bina Nusantara University Automotive & Robotics Program Computer Engineering Department BINUS ASO School of Engineering Computer Science Deparment BINUS Graduate Program Master of Computer Science Jakarta11480 Indonesia Bina Nusantara University Automotive & Robotics Program Computer Engineering Department BINUS ASO School of Engineering Jakarta11480 Indonesia

ISBN: (纸本)9798350390025

Modern retail businesses face a significant challenge with the inefficiency of manually changing price labels on shelves. This manual process not only consumes valuable time and resources but also increases the likelihood of errors, leading to potential inaccuracies in pricing and a less streamlined shopping experience for customers. The proposed solution is the implementation of an Electronic Shelf Label (ESL) system that automatically displays the prices of goods on retail business shelves. This system connects a website (front end and back end) to e-paper via a Wi-Fi network and microcontroller, allowing retail business owners to update prices more easily. Additionally, buyers can search for desired items through the website. The results are the time to send data from the website to the e-paper, namely, to know the performance of the e-paper used. The average time required from 10 attempts to send data from the website to the e-paper is 20.935 seconds. Since the data is connected to the server and will be updated automatically, using this system will be more efficient than manually changing the price label, although it takes time to transfer data from the website to the e-paper. © 2024 IEEE.

关键词： Sales

来源：评论

学校读者我要写书评

暂无评论

Investigation of Robot's Eye Gaze Affecting Human Pedestrian Behavior by Robot in Virtual Space 23

Investigation of Robot's Eye Gaze Affecting Human Pedestrian...

引用

23rd International Conference on Machine Learning and Cybernetics, ICMLC 2024

作者： Takata, R.Y.O. Fuse, Yotaro Takagi, Noboru Myagmardulam, Bilguunmaa Motoyoshi, Tatsuo Sawai, K.E.I. Masuta, Hiroyuki Graduate School of Intelligent Robotics Program Toyama Prefectural University Toyama Japan Toyama Prefectural University Department of Data Science Toyama Japan Toyama Prefectural University Department of Intelligent Robotics Toyama Japan

ISBN: (纸本)9798331528041

We investigate the effects of robot's gaze on human behavior when a human and a robot pass each other in a virtual reality space (VR). Nonverbal communication, particularly eye contact, influences physical proximity between individuals in human-to-human interaction. Results of a study indicate that the robot's gaze affects the physical distance between a human and a robot in human-robot communication. Studies on robot's gaze tend to focus on experimental scenarios in which robots and humans communicate. However, robot's gaze in environments with interactions, such as crossing paths, has not been extensively investigated. Therefore, we hypothesize that the robot's gaze exhibits an influence on the physical distance between humans and robots during encounters. In our experiment, we compare three conditions: a robot with a rotating neck and gaze directed at the subject, a faceless robot with a rotating neck, and a faceless robot with a rotating neck and gaze directed at the subject. We analyze the subject's paths of movement and physical distance between the subjects and the robot. Results suggest that the hypothesis that 'intimacy equilibrium model' works when humans and robot's cross paths in the VR environment is not supported. © 2024 IEEE.

关键词： Human robot interaction

来源：评论

学校读者我要写书评

暂无评论

Quantitative Analysis of Ankle Spasticity Through Electromyography Signal

Quantitative Analysis of Ankle Spasticity Through Electromyo...

引用

2024 Medical Technologies Congress, TIPTEKNO 2024

作者： Gurluk, Ayse Bilal, Hafiz Muhammad Hocaoglu, Elif Living Robotics Laboratory Graduate School of Health Sciences Physiotherapy and Rehabilitation Program Istanbul Medipol University Istanbul Turkey Living Robotics Laboratory Graduate School of Engineering and Natural Sciences Electrical-Electronics Engineering and Cyber-Systems Program Istanbul Medipol University Istanbul Turkey Living Robotics Laboratory Department of Electrical and Electronics Engineering Istanbul Medipol University Istanbul Turkey

ISBN: (纸本)9798331529819

This study presents a more accurate evaluation method for assessing ankle spasticity compared to conventional goniometric techniques by utilizing measurements obtained from Inertial Measurement Units (IMU) and surface electromyography (sEMG) devices. Integrating IMU and EMG sensors provides substantial benefits for accurately determining the range of motion (ROM) associated with spasticity during targeted movements. In this study, the proposed method provides precise measurement and an objective evaluation of onset time detection of antagonistic muscle co-contraction, and accordingly the ROM of the actuated joint. Moreover, the implementation of cost-effective wearable sensors enables the monitoring of muscle spasticity in clinical environments and can be particularly valuable in regions with limited access to experienced therapists, providing a more efficient approach to tracking spasticity levels. © 2024 IEEE.

关键词： Electromyography

来源：评论

学校读者我要写书评

暂无评论

Consideration of Constructing a Weed Detection Method for an Agricultural Robot in Perilla Fields 23

Consideration of Constructing a Weed Detection Method for an...

引用

23rd International Conference on Machine Learning and Cybernetics, ICMLC 2024

作者： Sugie, Tatsuya Takagi, Noboru Motoyoshi, Tatsuo Sawai, Kei Masuta, Hiroyuki Graduate School of Intelligent Robotics Program Toyama Prefectural University Toyama Japan Toyama Prefectural University Department of Data Science Toyama Japan Toyama Prefectural University Department of Intelligent Robotics Toyama Japan

ISBN: (纸本)9798331528041

In recent years, the number of farmers in Japan has been decreasing, and the automation of agriculture using robots, AI, and IoT, known as 'smart agriculture,' is being promoted. In this context, Toyama city is promoting a project to build a robot for automatic weeding between plants in a perilla fields. In this research, we are in charge of the system part that detects weeding regions and informs the weeding mechanism. In the field, a wide variety of weeds form vegetation, and it is necessary to prevent food crops from being cut. Therefore, we detect crop regions and the rest of the area should be the weeding regions. In this study, we construct a system to detect crop regions by transfer learning the object detection model YOLO-v8, which is a convolutional neural network (CNN), and evaluate the system in terms of detection accuracy and processing speed. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Separation of microalgae from bacterial contaminants using spiral microchannel in the presence of a chemoattractant

引用

Bioresources and Bioprocessing 2024年第1期11卷 511-525页

作者： Leticia F.Ngum Y.Matsushita Samir F.El-Mashtoly Ahmed M.R.Fath El-Bab Ahmed L Abdel-Mawgood Institute of Basic and Applied Sciences Biotechnology ProgramEgypt-Japan University of Science and TechnologyAlexandria 21934Egypt Institute of Basic and Applied Sciences Nanoscience ProgramEgypt-Japan University of Science and TechnologyAlexandria 21934Egypt Mechatronics and Robotics Engineering Department Egypt-Japan University of Science and TechnologyAlexandria 21934Egypt.

Cell separation using microfluidics has become an effective method to isolate biological contaminants from bodily fluids and cell cultures,such as isolating bacteria contaminants from microalgae cultures and isolating bacteria contaminants from white blood *** this study,bacterial cells were used as a model contaminant in microalgae culture in a passive microfluidics device,which relies on hydrodynamic forces to demonstrate the separation of microalgae from bacteria contaminants in U and W-shaped cross-section spiral microchannel fabricated by defocusing CO_(2) laser *** a flow rate of 0.7 ml/min in the presence of glycine as bacteria chemoattractant,the spiral microfluidics devices with U and W-shaped cross-sections were able to isolate microalgae(Desmodesmus sp.)from bacteria(***)with a high separation efficiency of 92%and 96%*** the same flow rate,in the absence of glycine,the separation efficiency of microalgae for U-and W-shaped cross-sections was 91%and 96%,*** was found that the spiral microchannel device with a W-shaped cross-section with a barrier in the center of the channel showed significantly higher separation *** microchannel chips with U-or W-shaped cross-sections were easy to fabricate and exhibited high *** these advantages,these devices could be widely applicable to other cell separation applications,such as separating circulating tumor cells from blood.

关键词： W-shaped cross-section CO_(2)laser ablation Glycine Separation efficiency Removal ratio

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：