检索结果-南通市图书馆

Skill enhancement learning with knowledge distillation

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

science china(Information sciences) 2024年第8期67卷 206-220页

作者： Naijun liu fuchun sun Bin FANG huaping liu department of computer science and technology Tsinghua University

Skill learning through reinforcement learning has significantly progressed in recent years. However, it often struggles to efficiently find optimal or near-optimal policies due to the inherent trial-and-error exploration in reinforcement learning. Although algorithms have been proposed to enhance skill learning efficacy, there is still much room for improvement in terms of skill learning performance and training stability. In this paper, we propose an algorithm called skill enhancement learning with knowledge distillation(SELKD), which integrates multiple actors and multiple critics for skill learning. SELKD employs knowledge distillation to establish a mutual learning mechanism among actors. To mitigate critic overestimation bias,we introduce a novel target value calculation method. We also perform theoretical analysis to ensure the convergence of SELKD. Finally, experiments are conducted on several continuous control tasks, illustrating the effectiveness of the proposed algorithm.

关键词： skill learning enhancement learning reinforcement learning knowledge distillation

同方期刊数据库

Data-driven electrical resistance tomography for robotic large-area tactile sensing

在线全文

学校读者我要写书评

暂无评论

science china(Information sciences) 2024年第09期67卷 336-337页

作者： Wendong ZHENG huaping liu Xiaofeng liu fuchun sun School of Electrical Engineering and Automation Tianjin University of Technology department of computer science and technology Tsinghua University Key Laboratory of Maritime Intelligent Cyberspace technology (Hohai university) Ministry of Education

Tactile sensing plays a crucial role in enabling robots to safely interact with objects in dynamic environments [1].Given that potential physical contact can occur at any location during robot interaction, there is a ... 详细信息

关键词： Robot applications

同方期刊数据库

Self-Aligning Multi-Modal Transformer for Oropharyngeal Swab Point Localization

在线全文

学校读者我要写书评

暂无评论

tsinghua science and technology 2024年第4期29卷 1082-1091页

作者： Tianyu liu fuchun sun department of computer science and technology Tsinghua UniversityBeijing 100083China

The oropharyngeal swabbing is a pre-diagnostic procedure used to test various respiratory diseases, including COVID and Influenza A (H1N1). To improve the testing efficiency of testing, a real-time, accurate, and robust sampling point localization algorithm is needed for robots. However, current solutions rely heavily on visual input, which is not reliable enough for large-scale deployment. The transformer has significantly improved the performance of image-related tasks and challenged the dominance of traditional convolutional neural networks (CNNs) in the image field. Inspired by its success, we propose a novel self-aligning multi-modal transformer (SAMMT) to dynamically attend to different parts of unaligned feature maps, preventing information loss caused by perspective disparity and simplifying overall implementation. Unlike preexisting multi-modal transformers, our attention mechanism works in image space instead of embedding space, rendering the need for the sensor registration process obsolete. To facilitate the multi-modal task, we collected and annotate an oropharynx localization/segmentation dataset by trained medical personnel. This dataset is open-sourced and can be used for future multi-modal research. Our experiments show that our model improves the performance of the localization task by 4.2% compared to the pure visual model, and reduces the pixel-wise error rate of the segmentation task by 16.7% compared to the CNN baseline.

关键词： segmentation localization transformer multi-modal perception robotic perception

Multi-objective robust control based on fuzzy singularly perturbed models for hypersonic vehicles

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

science china(Information sciences) 2011年第3期54卷 563-576页

作者： HU YeNan 1 , YUAN Yuan 2 , MIN HaiBo 2,3 & sun fuchun 1 1 department of computer science and technology, tsinghua university, beijing 100084, china 2 department of Automation, Xi’an Research Institute of High-Tech, Xi’an, Shaanxi 710025, china 3 beijing Institute of Astronautical System Engineering, beijing 100076, china department of computer science and technology Tsinghua University Beijing China department of Automation Xi’an Research Institute of High-Tech Xi’an Shaanxi China beijing Institute of Astronautical System Engineering Beijing China

In this paper, we propose a multi-objective robust controller based on fuzzy singularly perturbed models (FSPM) for the longitudinal motion of an air-breathing hypersonic vehicle. The control objective is to provide velocity and altitude tracking in the presence of the uncertainties and unknown nonlinearities in the model caused by variations in flight conditions. By approximating the multi-time-scale model with FSPM, ill-posed matrix calculations are circumvented. Under this framework, a fuzzy multi-objective robust controller is developed, and the linear matrix inequalities (LMI) condition is derived. Numerical simulation using the longitudinal model of a hypersonic aircraft is provided to demonstrate the effectiveness of the proposed strategy.

关键词： FSPM multi-objective robust control intelligent control hypersonic vehicle

COMPLEX NUMBER MODE APPROACH FOR REDUCING VIBRATION OF STRUCTURAL FLEXIBLE REDUNDANT ROBOT MANIPULATORS

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Chinese Journal of Mechanical Engineering 2001年第2期14卷 102-105页

作者： Wu Licheng sun Zengqi sun fuchun department of computer science and technology, Qinghua university Lu Zhen Yu Shouqian Zheng Hong beijing university of Aeronautics and Astronautics 清华大学计算机科学与工程系北京北京航空航天大学北京

A complex number mode analysis approach is proposed for vibration reducing of structural flexible redundant manipulators by utilizing self motion. In the proposed approach, the self motion is evaluated to nullify the modal exciting force of flexural motion, and the approach can be freely used when the degree of freedom of flexural motion is much greater than the available degree of reundancy. The availability and effectiveness of the proposed approach are demonstrated through numerical simulation with a four link spatial robotic manipulator possessing an end flexible link.

关键词： Redundant manipulators Self motion Vibration reducing Complex number mode analysis

维普期刊数据库

Shared Control Teleoperation for Targets Acquisition

在线全文

学校读者我要写书评

暂无评论

Shared Control Teleoperation for Targets Acquisition

Information Fusion-based Mobile Robot Path Control

第24届中国控制与决策会议

作者： WANG Dongli 1 , liu huaping 2 , sun fuchun 2 , DU Yonggui 1 1. Automation college of information engineering, Taiyuan university of technology, Taiyuan 030024, china2. tsinghua national laboratory for information science and technology,state key laboratory of intelligent technology and systems,department of computer science and technology,tsinghua university,beijing 100084,china

It is very difficult to capture the target in the microgravity environment, simply by relying on ground operators. Based on shared control theory, this paper effectively combines the decision-making ability of ground operators and the independent ability of space robot, through making full use of independent control ability of the space robot, and reducing operators’ workload and working time, to achieve more valid capture for targets. Shared control is more accurate capture of the target compared with teleoperation in experimental.

关键词： Space robots Shared control Target acquisition Independent grab

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

Information Fusion-based Mobile Robot Path Control

Fabric Recognition Using Zero-Shot Learning

第24届中国控制与决策会议

作者： Jingyan Wang 1 , huaping liu 2 , Meng Gao 1 , fuchun sun 2 , Wei Xiao 2 1. Shijiazhuang Tiedao university, Shijiazhuang 0500432. department of computer science and technology tsinghua university, beijing, 10084

The paper is combining with the deliberation-reaction hybrid control architecture by using a high-speed DSP chip is used to deal with the motion control and the mobile robot path planning, the technology of multi-sensor information fusion achieve for vertical and horizontal control, to reach precise positioning technology. In the case of local partially unknown environment, the function of obstacle avoidance is completed, and the trajectory coordinates of the mobile robot is showed on the PC real-time visual interface. The experimental results show the good usability of the proposed control method.

关键词： Mobile Robot Information Fusion Path Tracking

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

tsinghua science and technology 2019年第6期24卷 645-653页

作者： Feng Wang huaping liu fuchun sun Haihong Pan department of computer science and technology Tsinghua UniversityBeijing 100084China College of Mechanical Engineering Guangxi UniversityNanning 530003China

In this work, we use a deep learning method to tackle the Zero-Shot Learning(ZSL) problem in tactile material recognition by incorporating the advanced semantic information into a training model. Our main technical contribution is our proposal of an end-to-end deep learning framework for solving the tactile ZSL problem. In this framework, we use a Convolutional Neural Network(CNN) to extract the spatial features and Long Short-Term Memory(LSTM) to extract the temporal features in dynamic tactile sequences, and develop a loss function suitable for the ZSL setting. We present the results of experimental evaluations on publicly available datasets, which show the effectiveness of the proposed method.

关键词： Zero-Shot-Learning (ZSL) fabric recog nition tactile recog nition deep lear ning

维普期刊数据库

A novel multi-modal tactile sensor design using thermochromic material

在线全文

学校读者我要写书评

暂无评论

science china(Information sciences) 2019年第11期62卷 189-191页

作者： fuchun sun Bin FANG Hongxiang XUE huaping liu Haiming HUANG department of computer science and technology Tsinghua University Key Lab of Industrial computer Control Engineering of Hebei Province Yanshan University College of Information Engineering Shenzhen University

Tactile sensors are important for ensuring robotic perception [1], and their performance directly affects a robot’s capability for performing dexterous operations. These sensors mimic the tactile sensing capability of human hands. This is because the skin of human hands is involved in considerably meticulous works and can provide multi-

关键词： A novel multi-modal tactile sensor design using thermochromic material