检索结果-内蒙古大学图书馆

Proceedings of the 37th International Conference on Neural Information Processing systems

作者： Shuang Li Ke Li Wei Li Control and Simulation Center Harbin Institute of Technology China and National Key Laboratory of Modeling and Simulation for Complex Systems China Department of Computer Science University of Exeter Exeter UK

Bayesian Optimization (BO) is a powerful method for tackling expensive blackbox optimization problems. As a sequential model-based optimization strategy, BO iteratively explores promising solutions until a predetermined budget, either iterations or time, is exhausted. The decision on when to terminate BO significantly influences both the quality of solutions and its computational efficiency. In this paper, we propose a simple, yet theoretically grounded, two-step method for automatically terminating BO. Our core concept is to proactively identify if the search is within a convex region by examining previously observed samples. BO is halted once the local regret within this convex region falls below a predetermined threshold. To enhance numerical stability, we propose an approximation method for calculating the termination indicator by solving a bilevel optimization problem. We conduct extensive empirical studies on diverse benchmark problems, including synthetic functions, reinforcement learning, and hyperparameter optimization. Experimental results demonstrate that our proposed method saves up to ≈ 80% computational budget yet is with an order of magnitude smaller performance degradation, comparing against the other peer methods. In addition, our proposed termination method is robust in terms of the setting of its termination criterion.

关键词：

来源：评论

学校读者我要写书评

暂无评论

META-IMITATION LEARNING BY WATCHING VIDEO DEMONSTRATIONS 10

META-IMITATION LEARNING BY WATCHING VIDEO DEMONSTRATIONS

引用

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

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

Bounded Extremum Seeking for Single-Variable Static Map with Large Measurement Delay via Time-Delay Approach to Averaging

Bounded Extremum Seeking for Single-Variable Static Map with...

引用

IEEE Conference on Decision and control

作者： Xuefei Yang Emilia Fridman Bowen Zhao Center for Control Theory and Guidance Technology Harbin Institute of Technology Harbin China National Key Laboratory of Modeling and Simulation for Complex Systems School of Electrical Engineering Tel-Aviv University Israel

In this paper, we present a time-delay approach to gradient-based bounded extremum seeking (ES) with large measurement constant delay, for an unknown single-input static quadratic map. We assume that the extremum point and the Hessian $H$ belong to known intervals, whereas the sign of $H$ is known. We apply a time-delay approach to the bounded ES system and arrive at the neutral type system with a nominal linear delayed system. We present the latter system as a retarded one and employ variation of constants formula for practical stability analysis. Explicit conditions in terms of simple scalar inequalities depending on tuning parameters and delay are established to guarantee the practical stability of the bounded ES control systems. Given any delay and neighborhood of the extremum point and through the solution of the constructed inequalities, we find lower bounds on the dither period that ensures the practical stability.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Robust Time-Delay Approach to Continuous-Time Extremum Seeking for Multi-Variable Static Map

A Robust Time-Delay Approach to Continuous-Time Extremum See...

引用

IEEE Conference on Decision and control

作者： Xuefei Yang Emilia Fridman Center for Control Theory and Guidance Technology Harbin Institute of Technology Harbin China National Key Laboratory of Modeling and Simulation for Complex Systems School of Electrical Engineering Tel-Aviv University Israel

In this article, we introduce a time-delay approach to gradient-based extremum seeking (ES) in the continuous domain for n-dimensional (nD) static quadratic maps. As in the recently introduced (for 2D maps in the continuous domain), we transform the system to the time-delay one (neutral type system). This system is $\mathbf{O}(\varepsilon)$ . ‘perturbation of the averaged linear ODE system. We further explicitly present the neutral system as the linear ODE, where $\mathrm{O}(\varepsilon)$ -terms are considered as disturbances with distributed delays of the length of the small parameter $\varepsilon$ . Quantitative (for uncertain map) and qualitative (for unknown map) practical stability analyses are provided by employing a variation of constants formula that greatly simplifies the results compared to the previously used Lyapunov-Krasovskii (L-K) method. The new approach also simplifies the conditions and improves the results. Examples from the literature illustrate the efficiency of the new approach, allowing essentially large uncertainty of the Hessian matrix with bounds on $\varepsilon$ that are not too small.

关键词：

来源：评论

学校读者我要写书评

暂无评论

modeling and analysis of an underwater biomimetic vehicle-manipulator system

引用

Science China(Information Sciences) 2022年第3期65卷 255-257页

作者： Xuejian BAI Yu WANG Shuo WANG Rui WANG Min TAN Wei WANG School of Artificial Intelligence University of Chinese Academy of Sciences State Key Laboratory of Management and Control for Complex Systems Institute of AutomationChinese Academy of Sciences CAS Center for Excellence in Brain Science and Intelligence Technology School of Control Science and Engineering Shandong University

In recent years, underwater vehicle-manipulator systems(UVMSs) have captured researchers' attention. Inspired by fish, bionic mechanisms were introduced into a *** underwater biomimetic vehicle-manipulator system(UBVMS) is actuated by two symmetrically arranged biomimetic undulatory fin propulsors [1]. Based on the UBVMS, some topics have been researched, such as freefloating autonomous operation [2, 3], three-dimensional helical path following [4], and underwater video processing [5].

关键词： underwater biomimetic vehicle underwater robots underwater vehicle-manipulator system biomimetic robots hydrodynamic modeling

来源：评论

学校读者我要写书评

暂无评论

Real-Time Marker Localization Learning for GelStereo Tactile Sensing

Real-Time Marker Localization Learning for GelStereo Tactile...

引用

2022 IEEE International Conference on Robotics and Biomimetics, ROBIO 2022

作者： Liu, Shujuan Cui, Shaowei Zhang, Chaofan Cai, Yinghao Wang, Shuo University of Science and Technology Beijing China Institute of Automation Chinese Academy of Sciences State Key Laboratory of Management and Control for Complex Systems Beijing100190 China University of Chinese Academy of Sciences School of Artificial Intelligence Beijing100049 China Chinese Academy of Sciences Institute of Automation Beijing100190 China The State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences Beijing100190 China The Center for Excellence in Brain Science and Intelligence Technology Chinese Academy of Sciences Shanghai200031 China

ISBN: (纸本)9781665481090

Visuotactile sensing technology is becoming more popular in tactile sensing, but the effectiveness of the existing marker detection/localization methods remains to be further explored. Instead of contour-based blob detection, this paper presents a learning-based marker localization network for GelStereo visuotactile sensing called Marknet. Specifically, the Marknet presents a grid regression architecture to incorporate the distribution of the GelStereo markers. Furthermore, a marker rationality evaluator (MRE) is modelled to screen suitable prediction results. The experimental results show that the Marknet combined with MRE achieves 93.90% precision for irregular markers in contact areas, which outperforms the traditional contour-based blob detection method by a large margin of 42.32%. Meanwhile, the proposed learning-based marker localization method can achieve better realtime performance beyond the blob detection interface provided by the OpenCV library through GPU acceleration, which we believe will lead to considerable perceptual sensitivity gains in various robotic manipulation tasks. © 2022 IEEE.

关键词： Chemical detection

来源：评论

学校读者我要写书评

暂无评论

Hydrodynamic modeling and Parameter Identification of a Bionic Underwater Vehicle:RobDact

引用

Cyborg and Bionic systems 2022年第1期2022卷 164-176页

作者： Qiyuan Cao Rui Wang Tiandong Zhang Yu Wang Shuo Wang State Key Laboratory of Management and Control for Complex Systems Institute of A utomationChinese Academy of SciencesBeijingChina School of Artific ial Itelligence University of Chinese Academy of Sciences BeijingChina Center for Excellence in Brain Science and Itelligence Technology Chinese Academy of SciencesShanghaiChina

In this paper,the hydrodynamic modeling and parameter identifcation of the RobDact,a bionic underwater vehide inspired byDactylopteridae,are carried out based on computational fluid dynamics(CFD)and force measurement ***,thePaper briely describes the RobDact,then establishes the kinematis model and rigid body dynamics model of the RobDactaccording to the hydrodynamic force and moment *** CFD simulations,the hydrodynamic force of theRobDact at diferent speeds is obtained,and then,the hydrodynamic model parameters are ***,themeasurement platform is developed to obtain the relationship between the thrust generated by the RobDact and the inputfluctuation parameters,Finally,combining the rigid body dynamics model and the fin'thrust mapping model,thehydrodynamic model of the RobDact at diferent motion states is constructed.

关键词： dynamics(CFD) equations hydrodynamic

来源：评论

学校读者我要写书评

暂无评论

Learning to Manipulate Tools Using Deep Reinforcement Learning and Anchor Information

Learning to Manipulate Tools Using Deep Reinforcement Learni...

引用

2022 IEEE International Conference on Robotics and Biomimetics, ROBIO 2022

作者： Wei, Junhang Cui, Shaowei Hao, Peng Wang, Shuo The School of Future Technology University of Chinese Academy of Sciences Beijing100049 China The State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences Beijing100190 China The University of Chinese Academy of Sciences Beijing100049 China The Center for Excellence in Brain Science and Intelligence Technology Chinese Academy of Sciences Shanghai200031 China

ISBN: (纸本)9781665481090

Endowing robots with tool manipulation skills helps them accomplish challenging tasks. While robots manipulate tools to achieve goals, the alignment of tools and targets is a noise-sensitive and contact-rich task. However, it is difficult to access the accurate pose of the tool and the target. When there is unknown noise in the observations, reinforcement learning can't be sure to perform well. In this paper, we define the easier-to-obtain accurate task-related information as anchor information and introduce a tool manipulation method based on reinforcement learning and anchor information, which can perform well when the observations include unknown noise. To evaluate the method, we build a simulated environment ToolGym, which includes four different kinds of tools and different noise sampling functions for each tool. Finally, we compare our method with baseline methods to show the effectiveness of the proposed method. © 2022 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Visuotactile Feedback Parallel Gripper for Robotic Adaptive Grasping

Visuotactile Feedback Parallel Gripper for Robotic Adaptive ...

引用

2022 IEEE International Conference on Robotics and Biomimetics, ROBIO 2022

作者： Zhang, Boyue Cui, Shaowei Zhang, Chaofan Hu, Jingyi Wang, Shuo School of Artificial Intelligence University of Chinese Academy of Sciences Beijing100049 China Institute of Automation Chinese Academy of Sciences Beijing100190 China Institute of Automation Chinese Academy of Sciences State Key Laboratory of Management and Control for Complex Systems Beijing100190 China Center for Excellence in Brain Science and Intelligence Technology Chinese Academy of Sciences Shanghai200031 China

ISBN: (纸本)9781665481090

In robot grasp and dexterous manipulation tasks, tactile sensing is important for the control adjustment of the manipulator. In this paper, we present a novel low-cost parallel gripper with high-resolution tactile sensing, named the GelStereo Gripper. Furthermore, an adaptive grasp strategy is proposed to endow the gripper with tactile-feedback grasp stability-maintaining ability. We install the gripper on our robot platform and conduct various grasp experiments by utilizing proposed control methods. Experimental results verify the reliability of the GelStereo gripper and also prove the effectiveness of the proposed strategy for experimental objects with different features. © 2022 IEEE.

关键词： Grippers

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：