检索结果-内蒙古大学图书馆

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

作者： Yuting Xie Xianda Guo Cong Wang Kunhua Liu Long Chen School of Computer Science and Engineering Sun Yat-Sen University School of Computer Science Wuhan University State Key Laboratory for Management and Control of Complex Systems at the Institute of Automation Chinese Academy of Sciences School of Mechanical and Automotive Engineering Qingdao University of Technology

ISBN: (数字)9798350377705

ISBN: (纸本)9798350377712

Safety-critical scenarios are infrequent in natural driving environments but hold significant importance for the training and testing of autonomous driving systems. The prevailing approach involves generating safety-critical scenarios automatically in simulation by introducing adversarial adjustments to natural environments. These adjustments are often tailored to specific tested systems, thereby disregarding their transferability across different systems. In this paper, we propose AdvDiffuser, an adversarial framework for generating safety-critical driving scenarios through guided diffusion. By incorporating a diffusion model to capture plausible collective behaviors of background vehicles and a lightweight guide model to effectively handle adversarial scenarios, AdvDiffuser facilitates transferability. Experimental results on the nuScenes dataset demonstrate that AdvDiffuser, trained on offline driving logs, can be applied to various tested systems with minimal warm-up episode data and outperform other existing methods in terms of realism, diversity, and adversarial performance.

关键词： Training Diffusion models Intelligent robots Autonomous vehicles Testing

来源：评论

学校读者我要写书评

暂无评论

Framework and key Technologies for High-speed Railway Comprehensive Transport Hub

Framework and Key Technologies for High-speed Railway Compre...

引用

Digital Twins and Parallel Intelligence (DTPI), IEEE International Conference on

作者： Gang Xiong Wei Zhang Xiaoming Liu Zhiming Yuan Fenghua Zhu Zundong Zhang State Key Laboratory for Multimodal Artificial Intelligence Systems Chinese Academy of Sciences Beijing China School of Electrical and Control Engineering North China University of Technology Beijing China Communication Signal Research Institute China Academy of Railway Science Beijing China State Key Laboratory for Management and Control of Complex Systems Chinese Academy of Sciences Beijing China

The surge in urban population density in China has underscored the vital role of high-speed rail and subway systems in urban transit. Comprehensive transportation hubs, notably high-speed rail passenger terminals, have emerged as linchpins for seamless multimodal transit both within and beyond urban centers. China's railway sector vigorously champions the innovative concept of "station-city integration" and has articulated a forward-looking "16-word policy" for modern railway stations: "Seamless integration, environmental friendliness, economic viability, and intelligent accessibility". The judicious allocation of resources and the augmentation of coordination and control capabilities within comprehensive transportation hubs are imperative for the harmonious evolution of urban transit. The implementation of intelligent control systems for transportation hubs has emerged as a pivotal area of research. This study delves into the framework and critical technologies for high-speed railway comprehensive transport hubs, encompassing various components and their intricate interplay, including transport modalities, infrastructure, travelers, environmental factors, and information systems, thus forming a prototypical cyber-physical-social system. Addressing the prevalent and pressing imperative for the advancement of "station-city integration" high-speed railway comprehensive transport hubs, this research endeavor aims to surmount bottleneck challenges and refine the oversight of passenger flows at high-speed rail terminals. The findings hold noteworthy theoretical import and portend promising practical applications.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Turning Maneuverability Analysis of a Bionic Gliding Robotic Dolphin

Turning Maneuverability Analysis of a Bionic Gliding Robotic...

引用

IEEE International Conference on Robotics and Biomimetics (IEEE ROBIO)

作者： Zhang, Yang Wu, Zhengxing Wang, Jian Tan, Min Chinese Acad Sci State Key Lab Management & Control Complex Syst Inst Automat Beijing 100190 Peoples R China Univ Chinese Acad Sci Beijing 100049 Peoples R China

ISBN: (纸本)9781665405355

This paper focuses on the turning maneuverability of a bionic gliding robotic dolphin. Inspired by natural dolphin's turning styles, a novel gliding robotic dolphin with a special yaw joint is developed to purse high turning maneuverability. Thus, the robot can realize several turning patterns based on its flippers, yaw joint and dolphin-like dorsal-abdominal undulating mechanism. Considering both dolphin-like mode and gliding mode, a dynamic model is constructed to further analyze the turning performance through adopting the Newton-Euler method. Thereafter, five turning patterns are selected to explore how the special mechanism as well as some key parameters affect the turning performance of the robotic dolphin. The simulation results reveal that every pattern can successfully and effectively realize turning motions, furthermore, appropriate turning pattern as well as control parameters can contribute to high maneuverability, i.e., small turning radius and high turning velocity, which provides valuable guidance for future control and application of the bionic gliding robotic dolphin.

关键词： Gliding robotic dolphin bionic robot motion analysis turning patterns

来源：评论

学校读者我要写书评

暂无评论

Autonomous Crowdsensing: Operating and Organizing Crowdsensing for Sensing Automation

arXiv

引用

arXiv 2024年

作者： Wu, Wansen Yang, Weiyi Li, Juanjuan Zhao, Yong Zhu, Zhengqiu Chen, Bin Qiu, Sihang Peng, Yong Wang, Fei-Yue College of Systems Engineering National University of Defense Technology Hunan Province Changsha410073 China Hunan Institute of Advanced Technology Hunan Province Changsha410073 China State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of Automation Chinese Academy of Sciences Beijing100190 China State Key Laboratory for Management and Control of Complex Systems Institute of Automation Chinese Academy of Sciences Beijing100190 China

—The precise characterization and modeling of Cyber-Physical-Social systems (CPSS) requires more comprehensive and accurate data, which imposes heightened demands on intelligent sensing capabilities. To address this issue, Crowdsensing Intelligence (CSI) has been proposed to collect data from CPSS by harnessing the collective intelligence of a diverse workforce. Our first and second Distributed/Decentralized Hybrid Workshop on Crowdsensing Intelligence (DHW-CSI) have focused on principles and high-level processes of organizing and operating CSI, as well as the participants, methods, and stages involved in CSI. This letter reports the outcomes of the latest DHW-CSI, focusing on Autonomous Crowdsensing (ACS) enabled by a range of technologies such as decentralized autonomous organizations and operations, large language models, and human-oriented operating systems. Specifically, we explain what ACS is and explore its distinctive features in comparison to traditional crowdsensing. Moreover, we present the "6A-goal" of ACS and propose potential avenues for future research. Copyright © 2024, The Authors. All rights reserved.

关键词： Cyber Physical System

来源：评论

学校读者我要写书评

暂无评论

Design of Novel End-effectors for Robot-assisted Swab Sampling to Combat Respiratory Infectious Diseases 43

Design of Novel End-effectors for Robot-assisted Swab Sampli...

引用

43rd Annual International Conference of the IEEE-Engineering-in-Medicine-and-Biology-Society (IEEE EMBC)

作者： Tang, Ruijie Zheng, Jia Wang, Shuangyi Chinese Acad Sci Inst Automat State Key Lab Management & Control Complex Syst Beijing 100190 Peoples R China Beihang Univ Sch Gen Engn Beijing Peoples R China

ISBN: (纸本)9781728111797

The COVID-19 outbreak has caused the mortality worldwide and the use of swab sampling is a common way of screening and diagnosis. To combat respiratory infectious diseases and assist sampling, robots have been utilized and shown promising potentials. Nonetheless, a safe, patient-friendly, and low-cost swabbing system would be crucial for the practical implementation of robots in hospitals or inspection stations. In this study, we proposed two recyclable and cost-efficient end-effector designs that can be equipped at the distal end of a robot to passively regulate or actively sense the force exerted onto patients. One way is to introduce passive compliant mechanisms with soft material to increase the flexibility of the swabbing system, while the other way is utilizing a force-sensing gripper with embedded optoelectronic sensors to actively sense the force or torque. The proposed designs were modelled computationally and tested experimentally. It is identified that the passive compliant mechanisms can increase the flexibility of the swabbing system when subjected to the lateral force and mitigate the vertical force resulted from buckling. The lateral force range that the force-sensing gripper can detect is 0-0.35 N and the vertical force range causing buckling effect that can be sensed by gripper is 1.5-2.5 N.

关键词： Manufacturing processes Torque Infectious diseases Force Robot sensing systems End effectors Sensor systems

来源：评论

学校读者我要写书评

暂无评论

Learning Robot Grasping from a Random Pile with Deep Q-Learning 14th

Learning Robot Grasping from a Random Pile with Deep Q-Learn...

引用

14th International Conference on Intelligent Robotics and Applications (ICIRA)

作者： Chen, Bin Su, Jianhua Wang, Lili Gu, Qipeng Univ Chinese Acad Sci Beijing 100190 Peoples R China Chinese Acad Sci Inst Automat State Key Lab Management & Control Complex Syst Beijing 100190 Peoples R China

ISBN: (纸本)9783030890988;9783030890971

Grasping from a random pile is a great challenging application for robots. Most deep reinforcement learning-based methods focus on grasping of a single object. This paper proposes a novel structure for robot grasping from a pile with deep Q-learning, where each robot action is determined by the result of its current step and the next n steps. In the learning structure, a convolution neural network is employed to extract the target position, and a full connection network is applied to calculate the Q value of the grasping action. The former network is a pre-trained network and the latter one is a critical network structure. Moreover, we deal with the "reality gap" from the deep Q-learning policy learned in simulated environments to the real-world by large-scale simulation data and small-scale real data.

关键词： Grasping Deep Q-learning Simulation to real

来源：评论

学校读者我要写书评

暂无评论

Lifelong CycleGAN for continual multi-task image restoration

引用

PATTERN RECOGNITION LETTERS 2022年 153卷 183-189页

作者： Li, Yuping Nie, Xiangli Diao, Wenhui Zheng, Suiwu Chinese Acad Sci Inst Automat State Key Lab Management & Control Complex Syst Beijing 100190 Peoples R China Univ Chinese Acad Sci Beijing 100049 Peoples R China Chinese Acad Sci Aerosp Informat Res Inst Beijing 100190 Peoples R China Chinese Acad Sci Aerosp Informat Res Inst Key Lab Network Informat Syst Technol NIST Beijing 100094 Peoples R China Huizhou Zhongke Adv Mfg Res Ctr Co Ltd Huizhou 516000 Peoples R China

Recent years have witnessed the great success of deep learning in the applications of image restoration. However, there are still some challenging problems. First, most deep learning methods rely heavily on paired training images which are difficult to capture in the real world. Second, most existing methods are generally designed for a specific task of image restoration and they suffer from catastrophic forgetting when learn multiple tasks continually. Third, for most multi-task image restoration methods, they learn an individual network for each task and require to know the type of distortion for both training and test, which costs a lot of computational time and memory. To address the above issues, we propose a new lifelong learning framework based on CycleGAN for continual multi-task image restoration, called Lifelong CycleGAN (LCGAN), which can enhance low-light images, deblur and denoise simultaneously. The model utilizes knowledge distillation and memory replay to transfer knowledge and replay information that learned previously to alleviate forgetting. In addition, to regularize the unpaired training, we introduce the local discriminator and feature consistency constraint to preserve the color, edge and texture of input images. The proposed method can continually learn the three tasks using one network model and doesn't need to prejudge the type of distortion, which has low time and memory requirements. Experimental results demonstrate that LCGAN can achieve better visual and numerical results across the three tasks. (c) 2021 Elsevier B.V. All rights reserved.

关键词： Image restoration Lifelong continual learning CycleGAN Knowledge distillation

来源：评论

学校读者我要写书评

暂无评论

The Planar Force Perception on Robotic Bipolar Forceps 6

The Planar Force Perception on Robotic Bipolar Forceps

引用

6th IEEE International Conference on Advanced Robotics and Mechatronics (ICARM)

作者： Zhang, Xiuheng Zhang, Heng Li, Zhen Bian, Guibin Shenyang Ligong Univ Shenyang 110159 Liaoning Peoples R China Chinese Acad Sci Inst Automat State Key Lab Management & Control Complex Syst Beijing 100190 Peoples R China

ISBN: (纸本)9780738133645

In neurosurgery, bipolar forceps is used to coagulate and dissect brain tissues. As in robot-assisted neurosurgical scenario, force perception on bipolar forceps can quantify intraoperative contacting forces between tips of forceps and brain tissues. The quantified tool-tissue contacting force information can help quantify experts experience and ensure safety of robot-assisted surgery. It is challengeable to facilitate the forceps with forces sensing ability because of the limited space for installing sensors. The craniocerebral tissues, interstitial fluids and currents passing though polar of forceps also bring restrictions on force sensing methods. In this article, we propose a planar force perception method to collect force information on bipolar forceps tips in real-time by installing two Fiber Bragg Grating Sensors (FBGs) perpendicularly on each polar of bipolar forceps. FBGs are free from electromagnetic interference. The slim configuration and silicon materials of FBGs make it a suitable sensing method for little disturbance on working state of bipolar forceps. In experiments, axial forces and planar forces were monitored in real-time. It is possible to sense forces at a minimum of 0.01 N. The calibrating revolution on x, y axis is 0.01, 0.03 N, separately.

关键词： bipolar forceps medical device force perception neurosurgery

来源：评论

学校读者我要写书评

暂无评论

Dynamic adaptive multi-objective optimization algorithm based on type detection

引用

INFORMATION SCIENCES 2024年 654卷

作者： Cai, Xingjuan Wu, Linjie Zhao, Tianhao Wu, Di Zhang, Wensheng Chen, Jinjun Taiyuan Univ Sci & Technol Shanxi Key Lab Big Data Anal & Parallel Comp Taiyuan 030024 Peoples R China Nanjing Univ State Key Lab Novel Software Technol Nanjing Peoples R China Beijing Univ Technol Fac Informat Technol Beijing 100124 Peoples R China Chinese Acad Sci Inst Automat State Key Lab Management & Control Complex Syst Beijing Peoples R China Swinburne Univ Technol Dept Comp Sci & Software Engn Melbourne 3000 Australia

Dynamic multi-objective optimization problems (DMOPs) are multiobjective problems that are influenced by dynamically changing environmental parameters. Most current algorithms for solving DMOPs only respond to dynamic changes in the decision space or objective space and also ignore the impact of the type of DMOPs on the algorithm. The changes in the Paretooptimal solution (POS) and Pareto-optimal front (POF) may affect the type of change in DMOPs. Therefore, this paper proposed an adaptive dynamic multi-objective evolutionary algorithm for type detection (TDA-DMOEA). First, the dynamic detection operator is designed to identify the types of dynamic problems. The Wilcoxon signed-rank test and Hyper Volume (HV) are used to detect the difference of POS and POF in two adjacent environments respectively. Then, different response strategies are designed to cope with different types of changes in DMOP. In particular, a multi-angle-based transfer learning method (MA-TL) with a closed kernel function is derived when faced with simultaneous changes in POS and POF. Finally, a comprehensive study of the commonly used benchmark set of DMOPs is presented, and the proposed algorithm achieves better performance in optimizing DMOPs.

关键词： Adaptive response strategy Type detection Dynamic multi-objective optimization Transfer learning

来源：评论

学校读者我要写书评

暂无评论

Pedestrian Trajectory Prediction by Modeling the Interactions using Social LSTM Extensions

Pedestrian Trajectory Prediction by Modeling the Interaction...

引用

Chinese Automation Congress (CAC)

作者： Zeze Si Hongxia Zhao Haina Tang Fenghua Zhu Yisheng Lv Hao Lu State Key Laboratory for Management and Control of Complex Systems Institute of Autamation Chinese Academy of Sciences Beijing China School of Artificial Intelligence University of Chinese Academy of Sciences Beijing State Key Laboratory for Management and Control of Complex Systems Institute of Automation Chinese Academic of Sciences Beijing China Qingdao Academy of Intelligent Industries Qingdao China

In many driving situations, human mobility is an important topic in trajectory prediction. Considering the pedestrian trajectory as a sequence generative task, a prediction algorithm based on Social Long Short-Term Memory (Social LSTM) is implemented. In order to simulate the social interaction between pedestrians, Social Pooling (S-Pooling) is used to aggregate the hidden state of pedestrians, while the attention mechanism is utilized to aggregate information differently according to the importance of surrounding pedestrians. Furthermore, Convolutional Neural Networks (CNN) is introduced into Social LSTM model to consider both the interaction between people and the characteristic of scene scale in the prediction process. Experiments are carried out against baseline methods, and the results demonstrated that combining Social LSTM with attention mechanism or CNN can improve the performance of pedestrian trajectory prediction.

关键词： Automation Aggregates Predictive models Prediction algorithms Trajectory Convolutional neural networks Task analysis

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：