检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Wang, Yufei Sun, Zhanyi Zhang, Jesse Xian, Zhou Bıyık, Erdem Held, David Erickson, Zackory Robotics Institute Carnegie Mellon University United States Department of Computer Science University of Southern California United States

Reward engineering has long been a challenge in Reinforcement Learning (RL) research, as it often requires extensive human effort and iterative processes of trial-and-error to design effective reward functions. In this paper, we propose RL-VLM-F, a method that automatically generates reward functions for agents to learn new tasks, using only a text description of the task goal and the agent’s visual observations, by leveraging feedbacks from vision language foundation models (VLMs). The key to our approach is to query these models to give preferences over pairs of the agent’s image observations based on the text description of the task goal, and then learn a reward function from the preference labels, rather than directly prompting these models to output a raw reward score, which can be noisy and inconsistent. We demonstrate that RL-VLM-F successfully produces effective rewards and policies across various domains — including classic control, as well as manipulation of rigid, articulated, and deformable objects — without the need for human supervision, outperforming prior methods that use large pretrained models for reward generation under the same assumptions. Videos can be found on our project website: https://***/. Copyright © 2024, The Authors. All rights reserved.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

ICE-SEARCH: A LANGUAGE MODEL-DRIVEN FEATURE SELECTION APPROACH

arXiv

引用

arXiv 2024年

作者： Yang, Tianze Yang, Tianyi Lyu, Fuyuan Liu, Shaoshan Liu, Xue School of Computer Science McGill University Canada MILA Canada Shenzhen Institute of Artificial Intelligence and Robotics for Society China

This study unveils the In-Context Evolutionary Search (ICE-SEARCH) method, which is among the first works that melds large language models (LLMs) with evolutionary algorithms for feature selection (FS) tasks and demonstrates its effectiveness in Medical Predictive Analytics (MPA) applications. ICE-SEARCH harnesses the crossover and mutation capabilities inherent in LLMs within an evolutionary framework, significantly improving FS through the model's comprehensive world knowledge and its adaptability to a variety of roles. Our evaluation of this methodology spans three crucial MPA tasks: stroke, cardiovascular disease, and diabetes, where ICE-SEARCH outperforms traditional FS methods in pinpointing essential features for medical applications. ICE-SEARCH achieves State-of-the-Art (SOTA) performance in stroke prediction and diabetes prediction;the Decision-Randomized ICE-SEARCH ranks as SOTA in cardiovascular disease prediction. The study emphasizes the critical role of incorporating domain-specific insights, illustrating ICE-SEARCH's robustness, generalizability, and convergence. This opens avenues for further research into comprehensive and intricate FS landscapes, marking a significant stride in the application of artificial intelligence in medical predictive analytics. Copyright © 2024, The Authors. All rights reserved.

关键词： Predictive analytics

来源：评论

学校读者我要写书评

暂无评论

A Research on Scattering Removal Technology Using Compact GPU Machine for Real-Time Visualization 23

A Research on Scattering Removal Technology Using Compact GP...

引用

23rd International Conference on Control, Automation and Systems, ICCAS 2023

作者： Ono, Seiya Kim, Hyun-Woo Cho, Myungjin Lee, Min-Chul Kyushu Institute of Technology Department of Computer Science and Networks Fukuoka Iizuka820-8502 Japan School of ICT Robotics and Mechanical Engineering Iitc Hankyong National University Kyonggi-do Anseong17579 Korea Republic of

ISBN: (纸本)9788993215274

In this paper, we propose a system that enables visualization under the situation of scattering media such as fog and smoke by Peplography which is scattering media removal system, on a small GPU machine. Compared to conventional haze removal techniques, peplography technology is capable of removing more dense scattering media and can handle cases where the density of fog varies in the image. On the other hand, the conventional peplography process is performed on a desktop PC, and no system has yet been developed to perform peplography in an outside environment. Thus, we have developed a peplography system using the Jetson AGX Xavier, a single-board computer equipped with a GPU, which is superior to conventional methods in terms of portability, execution speed, and power consumption. We also demonstrate the performance of this system by conducting experiments using this system under a simulated scattering media scene and evaluating the obtained results in terms of FPS and the image evaluation metric. © 2023 ICROS.

关键词： Graphics processing unit

来源：评论

学校读者我要写书评

暂无评论

CoCap: Coordinated motion Capture for multi-actor scenes in outdoor environments

arXiv

引用

arXiv 2024年

作者： Rauniyar, Aditya Corah, Micah Scherer, Sebastian Robotics Institute School of Computer Science Carnegie Mellon University PittsburghPA United States Department of Computer Science The Colorado School of Mines GoldenCO United States

Motion capture has become increasingly important, not only in computer animation but also in emerging fields like the virtual reality, bioinformatics, and humanoid training. Capturing outdoor environments offers extended horizon scenes but introduces challenges with occlusions and obstacles. Recent approaches using multi-drone systems to capture multiple actor scenes often fail to account for multi-view consistency and reasoning across cameras in cluttered environments. Coordinated motion Capture (CoCap), inspired by Conflict-Based Search (CBS), addresses this issue by coordinating view planning to ensure multi-view reasoning during conflicts. In scenarios with high occlusions and obstacles, where the likelihood of inter-robot collisions increases, CoCap demonstrates performance that approaches the ideal outcomes of unconstrained planning, outperforming existing sequential planning methods. Additionally, CoCap offers a single-robot view search approach for real-time applications in dense environments. Copyright © 2024, The Authors. All rights reserved.

关键词： Robot programming

来源：评论

学校读者我要写书评

暂无评论

DiffTORI: differentiable trajectory optimization for deep reinforcement and imitation learning 24

DiffTORI: differentiable trajectory optimization for deep re...

引用

Proceedings of the 38th International Conference on Neural Information Processing Systems

作者： Weikang Wan Ziyu Wang Yufei Wang Zackory Erickson David Held Computer Science and Engineering Department University of California San Diego Institute for Interdisciplinary Information Sciences Tsinghua University Robotics Institute Carnegie Mellon University

ISBN: (纸本)9798331314385

This paper introduces DiffTORI, which utilizes Differentiable Trajectory Optimization as the policy representation to generate actions for deep Reinforcement and Imitation learning. Trajectory optimization is a powerful and widely used algorithm in control, parameterized by a cost and a dynamics function. The key to our approach is to leverage the recent progress in differentiable trajectory optimization, which enables computing the gradients of the loss with respect to the parameters of trajectory optimization. As a result, the cost and dynamics functions of trajectory optimization can be learned end-to-end. DiffTORI addresses the "objective mismatch" issue of prior model-based RL algorithms, as the dynamics model in DiffTORI is learned to directly maximize task performance by differentiating the policy gradient loss through the trajectory optimization process. We further benchmark DiffTORI for imitation learning on standard robotic manipulation task suites with high-dimensional sensory observations and compare our method to feed-forward policy classes as well as Energy-Based Models (EBM) and Diffusion. Across 15 model-based RL tasks and 35 imitation learning tasks with high-dimensional image and point cloud inputs, DiffTORI outperforms prior state-of-the-art methods in both domains.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Hybrid Model and Learning-Based Force Estimation Framework for Surgical Robots

A Hybrid Model and Learning-Based Force Estimation Framework...

引用

IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

作者： Hao Yang Haoying Zhou Gregory S. Fischer Jie Ying Wu Department of Computer Science Vanderbilt University TN USA Department of Robotics Engineering Worcester Polytechnic Institute MA USA

ISBN: (数字)9798350377705

ISBN: (纸本)9798350377712

Haptic feedback to the surgeon during robotic surgery would enable safer and more immersive surgeries but estimating tissue interaction forces at the tips of robotically controlled surgical instruments has proven challenging. Few existing surgical robots can measure interaction forces directly and the additional sensor may limit the life of instruments. We present a hybrid model and learning-based framework for force estimation for the Patient Side Manipulators (PSM) of a da Vinci Research Kit (dVRK). The model-based component identifies the dynamic parameters of the robot and estimates free-space joint torque, while the learning-based component compensates for environmental factors, such as the additional torque caused by trocar interaction between the PSM instrument and the patient’s body wall. We evaluate our method in an abdominal phantom and achieve an error in force estimation of under 10% normalized root-mean-squared error. We show that by using a model-based method to perform dynamics identification, we reduce reliance on the training data covering the entire workspace. Although originally developed for the dVRK, the proposed method is a generalizable framework for other compliant surgical robots. The code is available at https://***/vu-maple-lab/dvrk_force_estimation.

关键词： Learning systems Torque Medical robotics Instruments Force Dynamics Estimation Surgery Robot sensing systems Trajectory optimization

来源：评论

学校读者我要写书评

暂无评论

Laser-scribed graphene for sensors:preparation,modification,applications,and future prospects

引用

Light(Advanced Manufacturing) 2023年第2期4卷 67-91页

作者： Xing Liu Fangyi Zhang Qiwen Zhang Zhengfen Wan Xi Chen Institute of Photonic Chips University of Shanghai for Science and TechnologyShanghai 200093China Centre for Artificial-Intelligence Nanophotonics School of Optical-Electrical and Computer EngineeringUniversity of Shanghai for Science and TechnologyShanghai 200093China Queensland University of Technology(QUT)Centre for Robotics BrisbaneQLD 4000Australia

Sensors are widely used to acquire biological and environmental information for medical diagnosis,and health and environmental *** is a promising new sensor material that has been widely used in sensor fabrication in recent *** with many other existing graphene preparation methods,laser-scribed graphene(LSG)is simple,low-cost,environmentally friendly,and has good conductivity and high thermal stability,making it widely used in the sensor *** paper summarizes existing LSG methods for sensor *** LSG preparation methods and their variants are introduced first,followed by a summary of LSG modification methods designed explicitly for sensor ***,the applications of LSG in stress,bio,gas,temperature,and humidity sensors are summarized with a particular focus on multifunctional integrated ***,the current challenges and prospects of LSG-based sensors are discussed.

关键词： Graphene Graphene oxide Laser-scribed graphene Sensors

来源：评论

学校读者我要写书评

暂无评论

Analysis of Accurate Focal Points and Automatic Method of Digital Holographic Microscopy (DHM) for Obtaining an Accurate 3D Profile 23

Analysis of Accurate Focal Points and Automatic Method of Di...

引用

23rd International Conference on Control, Automation and Systems, ICCAS 2023

作者： Ono, Taishi Kim, Hyun-Woo Cho, Myungjin Lee, Min-Chul Kyushu Institute of Technology Department of Computer Science and Networks Fukuoka Iizuka820-8502 Japan School of ICT Robotics and Mechanical Engineering Iitc Hankyong National University Kyonggi-do Anseong17579 Korea Republic of

ISBN: (纸本)9788993215274

Currently, Digital Holographic Microscopy(DHM) is used for researching disease diagnosis or microbes. It cannot obtain the correct three-dimensional (3D) profile for a little noise because biological cells are microscopic. Then we need images of the focal point of DHM for reducing noise. However, in DHM, the depth of field becomes smaller influenced by the objective lens. In addition, focusing to obtain hologram images of biological cells in conventional DHM systems is entirely manual, because the decision of the focal point is made by the photographer's senses, which makes it even more difficult to focus on the object. These factors make it difficult for the conventional DHM system to acquire images of the focal points. Therefore, in this paper, we propose the analysis of accurate focusing points using a noise map. This eliminates the illusion or misperception that is a drawback of the manual. We improved an accurate 3D profile. 3D profiles were created at the focal point by the proposed method and at the positions before and after the focal points, and evaluated for SNR. From this, the proposed method can be used to analyze biological cells and microstructures. © 2023 ICROS.

关键词： Holograms

来源：评论

学校读者我要写书评

暂无评论

Towards Collaborative Simultaneous Localization and Mapping: A Survey of the Current Research Landscape

Field Robotics

引用

Field robotics 2025年 2卷 971-1000页

作者： Pierre-Yves Lajoie Benjamin Ramtoula Fang Wu Giovanni Beltrame Department of Computer and Software Engineering Polytechnique Montréal Canada Oxford Robotics Institute Department of Engineering Science University of Oxford United Kingdom Desay SV Automotive Singapore

Motivated by the tremendous progress, we witnessed in recent years, this paper presents a survey of the scientific literature on the topic of Collaborative Simultaneous Localization and Mapping (C-SLAM), also known as multi-robot SLAM. With fleets of self-driving cars on the horizon and the rise of multi-robot systems in industrial applications, we believe that Collaborative SLAM will soon become a cornerstone of future robotic applications. In this survey, we introduce the basic concepts of C-SLAM and present a thorough literature review. We also outline the major challenges and limitations of C-SLAM in terms of robustness, communication, and resource management. We conclude by exploring the area's current trends and promising research avenues.

关键词： Robots Simultaneous localization and mapping Robot kinematics Robot sensing systems Collaboration Surveys Multi-robot systems Sensors Market research Trajectory

来源：评论

学校读者我要写书评

暂无评论

Improvement of Depth Mapping and Occlusion Removal Method Using Integral Imaging 23

Improvement of Depth Mapping and Occlusion Removal Method Us...

引用

23rd International Conference on Control, Automation and Systems, ICCAS 2023

作者： Sakanoue, Haruka Kim, Hyun-Woo Cho, Myungjin Lee, Min-Chul Kyushu Institute of Technology Department of Computer Science and Networks Fukuoka Iizuka820-8502 Japan School of ICT Robotics and Mechanical Engineering Iitc Hankyong National University Kyonggi-do Anseong17579 Korea Republic of

ISBN: (纸本)9788993215274

In this paper, we propose the depth mapping and occlusion removal method using integral imaging (InIm). InIm is a method to obtain 3D information by capturing and reconstructing an object from multiple viewpoints. InIm can remove occlusion using parallax obtained by photographing an object from multiple viewpoints. However, occlusion removal uses a unit matrix to calculate the number of overlaps, which results in an overall darkening of the image. To solve this problem, this paper aims to improve occlusion removal by using the matrix of the mask image to calculate the number of overlaps. InIm can provide depth mapping that shows the distance from an object as pixel values. However, when there is an occlusion in front of the object, it is difficult to provide depth mapping object due to the influence of the occlusion. To solve this problem, we propose a method of depth mapping and occlusion removal method using InIm. Our method is expected to make great progress in improving safety in the field of autonomous driving. © 2023 ICROS.

关键词： Mapping

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：