检索结果-内蒙古大学图书馆

10th International Conference on Learning Representations, ICLR 2022

作者： Li, Jiayi Lu, Tao Cao, Xiaoge Cai, Yinghao Wang, Shuo School of Artificial Intelligence University of Chinese Academy of Sciences China State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences China Center for Excellence in Brain Science and Intelligence Technology Chinese Academy of Sciences China

Meta-Imitation Learning is a promising technique for the robot to learn a new task from observing one or a few human demonstrations. However, it usually requires a significant number of demonstrations both from humans and robots during the meta-training phase, which is a laborious and hard work for data collection, especially in recording the actions and specifying the correspondence between human and robot. In this work, we present an approach of meta-imitation learning by watching video demonstrations from humans. In comparison to prior works, our approach is able to translate human videos into practical robot demonstrations and train the meta-policy with adaptive loss based on the quality of the translated data. Our approach relies only on human videos and does not require robot demonstration, which facilitates data collection and is more in line with human imitation behavior. Experiments reveal that our method achieves the comparable performance to the baseline on fast learning a set of vision-based tasks through watching a single video demonstration. © 2022 ICLR 2022 - 10th International Conference on Learning Representationss. All rights reserved.

关键词： Robots

来源：评论

学校读者我要写书评

暂无评论

Parallel-Data-Based Social Evolution Modeling

引用

Tsinghua Science and Technology 2021年第6期26卷 878-885页

作者： Weishan Zhang Zhaoxiang Hou Xiao Wang Zhidong Xu Xin Liu Fei-Yue Wang China University of Petroleum Qingdao 266580China Qingdao Academy of Intelligent Industry Qingdao 266111China State Key Laboratory of Management and Control for Complex Systems Institute of AutomationChinese Academy of SciencesBeijing 100049China Institute of National Security National Defense UniversityBeijing 100081China

Abnormal or drastic changes in the natural environment may lead to unexpected events,such as tsunamis and earthquakes,which are becoming a major threat to national ***,no effective assessment approach can deduce a situation and determine the optimal response strategy when a natural disaster *** this study,we propose a social evolution modeling approach and construct a deduction model for self-playing,self-learning,and self-upgrading on the basis of the idea of parallel data and reinforcement *** proposed approach can evaluate the impact of an event,deduce the situation,and provide optimal strategies for *** the breakage of a submarine cable caused by earthquake as an example,we find that the proposed modeling approach can obtain a higher reward compared with other existing methods.

关键词： paral el data reinforcement learning decision-making

来源：评论

学校读者我要写书评

暂无评论

A DAOs-Based Publishing system for Advancing Open Access

A DAOs-Based Publishing System for Advancing Open Access

引用

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

作者： Siji Ma Juanjuan Li Sangtian Guan Qinghua Ni Fei Lin Tengchao Zhang Jun Huang Fei-Yue Wang Faculty of Innovation Engineering Macau University of Science and Technology Macau China State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of Automation Chinese Academy of Sciences Beijing China State Key Laboratory for Management and Control of Complex Systems Institute of Automation Chinese Academy of Sciences Beijing China

ISBN: (数字)9798350349252

ISBN: (纸本)9798350349269

The development of the scientific publishing system has remarkably enhanced global accessibility to research findings and substantially increased the visibility and dissemination of academic publications. However, significant challenges still exist in effectively safeguarding the intellectual property rights of contributors, such as the unauthorized usage of materials, the complexity of enforcing intellectual property rights across various legal jurisdictions, and high instances of plagiarism and content misuse. Additionally, financial barriers related to open access may restrict the participation of economically disadvantaged researchers, potentially biasing scientific records towards more affluent research initiatives. To address these issues, a novel decentralized framework is formulated to ensure truly open access. This framework leverages blockchain for immutable record-keeping and clear attribution of authorship, to prevent unauthorized usage and plagiarism. Besides, it also utilizes a copyright-sharing model based on decentralized autonomous organizations and operations (DAOs), where smart contracts automatically enforce copyright and access policies to ensure fair compensation for authors and researchers. Furthermore, the copyright sharing model based on non-fungible tokens (NFT) and gradual ownership optimization (GOO) mechanism is proposed to ensure fair and accurate recognition and compensation for scholarly contributions.

关键词： Open Access Law Plagiarism Smart contracts Intellectual property Research initiatives Scientific publishing Nonfungible tokens Remuneration Optimization

来源：评论

学校读者我要写书评

暂无评论

Mechanical design paradigm based on ACP method in parallel manufacturing 1

Mechanical design paradigm based on ACP method in parallel m...

引用

1st IEEE International Conference on Digital Twins and Parallel Intelligence, DTPI 2021

作者： Li, Shimeng Wang, Yutong Wang, Xiao Wang, Feiyue Institute of Automation Chinese Academy of Sciences The State Key Laboratory for Management and Control of Complex Systems Beijing China Qingdao Academy of Intelligent Industries Qingdao China

ISBN: (纸本)9781665433372

Parallel Manufacturing is a new manufacturing paradigm in industry, deeply integrating informalization, automation, and artificial intelligence. In this paper we propose a new mechanical design paradigm in Parallel Manufacturing based on ACP method. The key is to regard the design procedure based on artificial design and emulation method as two independent procedures, which can be modeled as a parallel system. The design procedure based on ACP method does not include a real system, which is an inventive extension of the traditional parallel system. This method can be implemented with social information by introducing the definition of SDV, SDM, and Intelligent Design Manager, making it highly adaptive for social manufacturing and Parallel Manufacturing. © 2021 IEEE.

关键词： Manufacture

来源：评论

学校读者我要写书评

暂无评论

Speed control of a Biomimetic Robotic Fish Based on Linear Active Disturbance Rejection control

Speed Control of a Biomimetic Robotic Fish Based on Linear A...

引用

2021 China Automation Congress, CAC 2021

作者： Huang, Yupei Wu, Zhengxing Du, Sheng The State Key Laboratory of Management and Control for Complex Systems Insititute of Automation Chinese Academy of Sciences Beijing China University of Chinese Academy of Sciences Beijing China

ISBN: (纸本)9781665426473

This paper proposes a speed control method for a biomimetic robotic fish based on linear active disturbance rejection control. Inspired by a bluefin tuna in nature, a robotic fish with a two-joint propulsive mechanism and a crescent-shaped caudal fin is first designed. Thereafter, for a steady forward swimming speed, a speed control framework based on the linear form of active disturbance rejection control method is proposed, where the oscillation frequency of the caudal fin is adopted to regulate the swimming speed. Meanwhile, a swimming speed estimation approach based on the oscillation period is provided to effectively reduce the phenomenon of swimming speed fluctuation. Finally, both simulation and experiment results illustrate that the proposed method is effective for the speed control of robotic fish. © 2021 IEEE

关键词： Speed control

来源：评论

学校读者我要写书评

暂无评论

Backstepping-based parallel control for cascaded nonlinear systems 1

Backstepping-based parallel control for cascaded nonlinear s...

引用

1st IEEE International Conference on Digital Twins and Parallel Intelligence, DTPI 2021

作者： Lu, Jingwei Wei, Qinglai Zhou, Tianmin Han, Liyuan Wang, Fei-Yue School of Artificial Intelligence University of Chinese Academy of Sciences Beijing China The State Key Laboratory for Management and Control of Complex Systems Institute of Automation Beijing China

ISBN: (纸本)9781665433372

In this study, a novel nonlinear parallel control method is proposed for cascaded nonlinear systems using the backstepping technique. Unlike the existing state feedback control methods, the control input is taken into the feedback system. First, an augmented system is constructed to facilitate the constructing the Lyapunov function. Then, the backstepping technique can be applied to obtain the nonlinear parallel control law, and the stability analysis is shown using the Lyapunov theory. Finally, a simulation is conducted to demonstrate the effectiveness of the proposed parallel control method. © 2021 IEEE.

关键词： Backstepping

来源：评论

学校读者我要写书评

暂无评论

Parallel control-Based Event-Triggered Optimal control for Constrained Discrete-Time Nonlinear systems

Parallel Control-Based Event-Triggered Optimal Control for C...

引用

2021 China Automation Congress, CAC 2021

作者： Lu, Jingwei Bai, Tianxiang Wei, Qinglai Zhou, Tianmin Wang, Fei-Yue School of Artificial Intelligence University of Chinese Academy of Sciences Beijing China The State Key Laboratory for Management and Control of Complex Systems Institute of Automation Beijing China

ISBN: (纸本)9781665426473

This study proposes a new event-triggered optimal control (ETOC) method for discrete-time (DT) constrained nonlinear systems. First, a new triggering condition is proposed. We show the asymptotic stability of the closed-loop system using the proposed triggering condition and analyze the degeneration degree of the real performance index. Second, to perform the proposed ETOC method effectively, parallel control (PC) combined with adaptive dynamic programming (ADP) is applied. Finally, the validity of the ETOC method is validated by a simulation. © 2021 IEEE

关键词： Asymptotic stability

来源：评论

学校读者我要写书评

暂无评论

Embed Trajectory Imitation in Reinforcement Learning: A Hybrid Method for Autonomous Vehicle Planning

Embed Trajectory Imitation in Reinforcement Learning: A Hybr...

引用

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

作者： Yuxiao Wang Xingyuan Dai Kara Wang Hub Ali Fenghua Zhu State Key Laboratory of Management and Control for Complex System Institute of Automation Chinese Academy of Sciences Beijing China School of Artificial Intelligence University of Chinese Academy of Sciences Beijing China School of Artificial Intelligence Anhui University Hefei China

Learning-based autonomous vehicle trajectory planning methods have shown excellent performance in a variety of complex traffic scenarios. However, the existing imitation learning (IL) and reinforcement learning (RL) algorithms still have their limitations, such as poor safety and generalizability for IL, and low data efficiency for RL. To leverage their respective advantages and mitigate the limitations, this paper proposes a novel hybrid RL algorithm for autonomous vehicle planning, where IL is embedded in it to guide its exploration with expert knowledge. Different from existing approaches, we use multi-step trajectory prediction instead of behavior cloning as the IL method integrated with online RL. Through such design, we make a further step in the research about how expert demonstration can be helpful to RL. Moreover, we conduct parallel training and testing of the algorithm based on real-world driving data. Experimental results demonstrate that our proposed approach outperforms standalone IL and RL methods, and performs better than RL methods enhanced by behavior cloning.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Open-/Closed-loop Active Learning for Data-driven Predictive control

arXiv

引用

arXiv 2024年

作者： Feng, Shilun Shi, Dawei Shi, Yang Zheng, Kaikai The State Key Laboratory of Intelligent Control and Decision of Complex Systems MIIT Key Laboratory of Servo Motion System Drive and Control School of Automation Beijing Institute of Technology Beijing100081 China The Department of Mechanical Engineering Faculty of Engineering University of Victoria VictoriaBCV8N 3P6 Canada

An important question in data-driven control is how to obtain an informative dataset. In this work, we consider the problem of effective data acquisition of an unknown linear system with bounded disturbance for both open-loop and closed-loop stages. The learning objective is to minimize the volume of the set of admissible systems. First, a performance measure based on historical data and the input sequence is introduced to characterize the upper bound of the volume of the set of admissible systems. On the basis of this performance measure, an open-loop active learning strategy is proposed to minimize the volume by actively designing inputs during the open-loop stage. For the closed-loop stage, an closed-loop active learning strategy is designed to select and learn from informative closed-loop data. The efficiency of the proposed closed-loop active learning strategy is proved by showing that the unselected data cannot benefit the learning performance. Furthermore, an adaptive predictive controller is designed in accordance with the proposed data acquisition approach. The recursive feasibility and the stability of the controller are proved by analyzing the effect of the closed-loop active learning strategy. Finally, numerical examples and comparisons illustrate the effectiveness of the proposed data acquisition strategy. Copyright © 2024, The Authors. All rights reserved.

关键词： Active learning

来源：评论

学校读者我要写书评

暂无评论

A Novel Heterogeneous Actor-critic Algorithm with Recent Emphasizing Replay Memory

引用

International Journal of Automation and computing 2021年第4期18卷 619-631页

作者： Bao Xi Rui Wang Ying-Hao Cai Tao Lu Shuo Wang State Key Laboratory of Management and Control for Complex Systems Institute of AutomationChinese Academy of SciencesBeijing 100190China University of Chinese Academy of Sciences Beijing 100049China Center for Excellence in Brain Science and Intelligence Technology Chinese Academy of SciencesShanghai 200031China

Reinforcement learning(RL) algorithms have been demonstrated to solve a variety of continuous control tasks. However,the training efficiency and performance of such methods limit further applications. In this paper, we propose an off-policy heterogeneous actor-critic(HAC) algorithm, which contains soft Q-function and ordinary Q-function. The soft Q-function encourages the exploration of a Gaussian policy, and the ordinary Q-function optimizes the mean of the Gaussian policy to improve the training efficiency. Experience replay memory is another vital component of off-policy RL methods. We propose a new sampling technique that emphasizes recently experienced transitions to boost the policy training. Besides, we integrate HAC with hindsight experience replay(HER) to deal with sparse reward tasks, which are common in the robotic manipulation domain. Finally, we evaluate our methods on a series of continuous control benchmark tasks and robotic manipulation tasks. The experimental results show that our method outperforms prior state-of-the-art methods in terms of training efficiency and performance, which validates the effectiveness of our method.

关键词： Reinforcement learning(RL) actor-critic experience replay training efficiency manipulation skill learning

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：