检索结果-内蒙古大学图书馆

Reinforcement learning of non-additive joint steganographic embedding costs with attention mechanism

science China(information sciences) 2023年第3期66卷 273-286页

作者： Weixuan TANG Bin LI Weixiang LI Yuangen WANG Jiwu HUANG Institute of Artificial Intelligence and Blockchain Guangzhou University Guangdong Key Laboratory of Intelligent Information Processing Shenzhen Key Laboratory of Media Security Shenzhen University Shenzhen Institute of Artificial Intelligence and Robotics for Society School of Computer Science and Cyber Engineering Guangzhou University

Image steganography is the art and science of secure communication by concealing information within digital images. In recent years, the techniques of steganographic cost learning have developed rapidly. Although the existing methods can learn satisfactory additive costs, the interplay of different pixels' embedding impacts has not been considered, so the potential of learning may not be fully exploited. To overcome this limitation, in this paper, a reinforcement learning paradigm called Jo Po L(joint policy learning) is proposed to extend the idea of additive cost learning to a non-additive situation. Jo Po L aims to capture the interactions within pixel blocks by defining embedding policies and evaluating contributions of embedding impacts on a block level rather than a pixel level. Then, a policy network is utilized to learn optimal joint embedding policies for pixel blocks through interactions with the environment. Afterwards,these policies can be converted into joint embedding costs for practical message embedding. The structure of the policy network is designed with an effective attention mechanism and incorporated with the domain knowledge derived from traditional non-additive steganographic methods. The environment is responsible for assigning rewards according to the impacts of the sampled joint embedding actions, which are evaluated by the gradient information of a neural network-based steganalyzer. Experimental results show that the proposed non-additive method Jo Po L significantly outperforms the existing additive methods against both feature-based and CNN-based steganalzyers over different payloads.

关键词： information hiding non-additive steganography steganalysis cost learning image processing

来源：评论

学校读者我要写书评

暂无评论

Safe Q-Learning for Data-Driven Nonlinear Optimal Control With Asymmetric State Constraints

引用

IEEE/CAA Journal of Automatica Sinica 2024年第12期11卷 2408-2422页

作者： Mingming Zhao Ding Wang Shijie Song Junfei Qiao the School of Information Science and Technology the Beijing Key Laboratory of Computational Intelligence and Intelligent Systemthe Beijing Laboratory of Smart Environmental Protectionand the Beijing Institute of Artificial IntelligenceBeijing University of Technology.Beijing 100124China IEEE the School of Mechanical and Electrical Engineering University of Electronic Science and Technology of ChinaChengdu 611731China

This article develops a novel data-driven safe Q-learning method to design the safe optimal controller which can guarantee constrained states of nonlinear systems always stay in the safe region while providing an optimal ***,we design an augmented utility function consisting of an adjustable positive definite control obstacle function and a quadratic form of the next state to ensure the safety and ***,by exploiting a pre-designed admissible policy for initialization,an off-policy stabilizing value iteration Q-learning(SVIQL)algorithm is presented to seek the safe optimal policy by using offline data within the safe region rather than the mathematical ***,the monotonicity,safety,and optimality of the SVIQL algorithm are theoretically *** obtain the initial admissible policy for SVIQL,an offline VIQL algorithm with zero initialization is constructed and a new admissibility criterion is established for immature iterative ***,the critic and action networks with precise approximation ability are established to promote the operation of VIQL and SVIQL ***,three simulation experiments are conducted to demonstrate the virtue and superiority of the developed safe Q-learning method.

关键词： Adaptive critic control adaptive dynamic programming(ADP) control barrier functions(CBF) stabilizing value iteration Q-learning(SVIQL) state constraints

来源：评论

学校读者我要写书评

暂无评论

Design and Evaluation of a Bidirectional Ankle Exoskeleton system 38

Design and Evaluation of a Bidirectional Ankle Exoskeleton S...

引用

38th Youth Academic Annual Conference of Chinese Association of Automation, YAC 2023

作者： Peng, Yiying Chen, Jianyu Wang, Liping Han, Jianda Zhang, Juanjuan Nankai University Tianjin Key Laboratory of Intelligent Robotics Institute of Robotics and Automatic Information System College of Artificial Intelligence School of Materials Science and Engineering Smart Sensing Interdisciplinary Science Center Tianjin300350 China

ISBN: (纸本)9798350303636

Patients with foot-drop are unable to dorsiflexion during the swing phase, resulting in toe dragging and increasing the risk of falling. Ankle-foot-orthoses (AFOs) can reduce foot-drop of patients but limit their ankle plantarflexion. Assistive exoskeletons have a significant improvement on walking economy and the assistance can enhance human mobility. Here, we developed a bidirectional ankle exoskeleton to provide passive dorsiflexion assistance during the swing phase and active plantarflexion assistance during the stance phase. The performance of our exoskeleton was tested based on experiments of one healthy subject. We changed the front passive assistive materials and evaluated the kinematics and biological efforts of the subject with different assistive conditions. Results showed that the maximum ankle dorsiflexion angles were increased during the swing phase and ankle joint torques were reduced under assistance. The root mean square (RMS) of the soleus and tibialis anterior muscle activities were decreased by 21.8% and 17.9% under the assistance of elastic ropes condition. Our design expected to provide assistance for individuals with impaired dorsiflexion and plantarflexor muscle strength, which can enhance their locomotion ability. © 2023 IEEE.

关键词： bidirectional ankle exoskeleton foot-drop passive assistance wearable robotics

来源：评论

学校读者我要写书评

暂无评论

Recent Progress in Reinforcement Learning and Adaptive Dynamic Programming for Advanced Control Applications

引用

IEEE/CAA Journal of Automatica Sinica 2024年第1期11卷 18-36页

作者： Ding Wang Ning Gao Derong Liu Jinna Li Frank L.Lewis IEEE the Faculty of Information Technology Beijing Key Laboratory of Computational Intelligence and Intelligent SystemBeijing Laboratory of Smart Environmental Protectionand Beijing Institute of Artificial IntelligenceBeijing University of TechnologyBeijing 100124China the School of System Design and Intelligent Manufacturing Southern University of Science and TechnologyShenzhen 518055China the Department of Electrical and Computer Engineering University of Illinois at ChicagoChicago IL 60607 USA the School of Information and Control Engineering Liaoning Petrochemical UniversityFushun 113001China the UTA Research Institute the University of Texas at ArlingtonArlington TX 76118 USA

Reinforcement learning(RL) has roots in dynamic programming and it is called adaptive/approximate dynamic programming(ADP) within the control community. This paper reviews recent developments in ADP along with RL and its applications to various advanced control fields. First, the background of the development of ADP is described, emphasizing the significance of regulation and tracking control problems. Some effective offline and online algorithms for ADP/adaptive critic control are displayed, where the main results towards discrete-time systems and continuous-time systems are surveyed, ***, the research progress on adaptive critic control based on the event-triggered framework and under uncertain environment is discussed, respectively, where event-based design, robust stabilization, and game design are reviewed. Moreover, the extensions of ADP for addressing control problems under complex environment attract enormous attention. The ADP architecture is revisited under the perspective of data-driven and RL frameworks,showing how they promote ADP formulation ***, several typical control applications with respect to RL and ADP are summarized, particularly in the fields of wastewater treatment processes and power systems, followed by some general prospects for future research. Overall, the comprehensive survey on ADP and RL for advanced control applications has d emonstrated its remarkable potential within the artificial intelligence era. In addition, it also plays a vital role in promoting environmental protection and industrial intelligence.

关键词： Adaptive dynamic programming(ADP) advanced control complex environment data-driven control event-triggered design intelligent control neural networks nonlinear systems optimal control reinforcement learning(RL)

来源：评论

学校读者我要写书评

暂无评论

FaRo-PPF: Fast and Robust Point Pair Feature for 6D Pose Estimation in Industrial Stamping 8

FaRo-PPF: Fast and Robust Point Pair Feature for 6D Pose Est...

引用

8th IEEE International Conference on Advanced robotics and Mechatronics, ICARM 2023

作者： He, Cheng Zhang, Xuebo Zhao, Zhenjie Institute of Robotics and Automatic Information System College of Artificial Intelligence Nankai University Tianjin Key Laboratory of Intelligent Robotics Tianjin300350 China School of Computer Science Nanjing University of Information Science and Technology Nanjing210044 China

ISBN: (纸本)9798350300178

Estimating 6D poses of targets efficiently is critical for industrial stamping tasks, in which the Point Pair Feature (PPF) method has been widely used. Based on PPF, this paper proposes Fast and Robust PPF, i.e. FaRo-PPF, which improves PPF in the following three key aspects: adaptive down-sampling based on surface features, point pair matching based on voting ball, and normal-based pose verification. The three designs alleviate existing problems of the local matching stage in PPF, and make FaRo-PPF a stronger method of 6D pose estimation in industrial stamping. To demonstrate the effectiveness of FaRo-PPF, we compare it with PPF on five publically available datasets. Experiment results showed that FaRo-PPF is able to significantly improve accuracy by about 15% and reduce the execution time by about 40% across all test data. We further conduct a grasping and assembly experiment on a physical robot arm, and similar improvement can be observed. FaRo-PPF achieves a higher success rate of assembly and reduces the execution time by about 50%. © 2023 IEEE.

关键词： Stamping

来源：评论

学校读者我要写书评

暂无评论

Squeezing More Past Knowledge for Online Class-Incremental Continual Learning

引用

IEEE/CAA Journal of Automatica Sinica 2023年第3期10卷 722-736页

作者： Da Yu Mingyi Zhang Mantian Li Fusheng Zha Junge Zhang Lining Sun Kaiqi Huang the State Key Laboratory of Robotics and System Harbin Institute of Technology(HIT)Harbin 150080China Center for Research on Intelligent System and Engineering Institute of AutomationChinese Academy of Sciences(CASIA)Beijing 100190 School of Artificial Intelligence University of Chinese Academy of SciencesBeijing 100049China Center for Research on Intelligent System and Engineering Institute of AutomationChinese Academy of Sciences(CASIA)Beijing 100190the School of Artificial IntelligenceUniversity of Chinese Academy of SciencesBeijing 100049 the CAS Center for Excellence in Brain Science and Intelligence Technology Shanghai 200031China IEEE

Continual learning(CL)studies the problem of learning to accumulate knowledge over time from a stream of data.A crucial challenge is that neural networks suffer from performance degradation on previously seen data,known as catastrophic forgetting,due to allowing parameter *** this work,we consider a more practical online class-incremental CL setting,where the model learns new samples in an online manner and may continuously experience new ***,prior knowledge is unavailable during training and *** works usually explore sample usages from a single dimension,which ignores a lot of valuable supervisory *** better tackle the setting,we propose a novel replay-based CL method,which leverages multi-level representations produced by the intermediate process of training samples for replay and strengthens supervision to consolidate previous ***,besides the previous raw samples,we store the corresponding logits and features in the ***,to imitate the prediction of the past model,we construct extra constraints by leveraging multi-level information stored in the *** the same number of samples for replay,our method can use more past knowledge to prevent *** conduct extensive evaluations on several popular CL datasets,and experiments show that our method consistently outperforms state-of-the-art methods with various sizes of episodic *** further provide a detailed analysis of these results and demonstrate that our method is more viable in practical scenarios.

关键词： Catastrophic forgetting class-incremental learning continual learning(CL) experience replay

来源：评论

学校读者我要写书评

暂无评论

A Non-Calibration Nonlinear Errors Correction Method for Phase-Shifting Patterns

A Non-Calibration Nonlinear Errors Correction Method for Pha...

引用

Chinese Automation Congress (CAC)

作者： Yujie Zheng Bin Chen Shiqian Wu Institute of Robotics and Intelligent System School of Information Science and Engineering Wuhan University of Science and Technology Wuhan China

ISBN: (数字)9798350368604

ISBN: (纸本)9798350368611

To reduce phase errors caused by nonlinear responses in the measurement system, we propose a non-calibration correction method for nonlinear errors in phase-shifting patterns. First, the captured grating fringes are transformed using a Fourier transform to convert the images from the gray scale to the frequency domain. The first harmonic frequency is then determined using an appropriate method. Second, the appropriate filter cutoff frequency is calculated by combining the phase-shift algorithm's sensitivity to different harmonics with frequency response curves. Low-pass filtering is then performed on the phase-shift patterns to suppress high frequencies caused by nonlinear responses, thus reducing phase errors in the actual measurement process. Finally, the standard plane and other irregularly shaped objects were reconstructed using the proposed method and other techniques. Measurement errors under different conditions were then compared. Experimental results show that the proposed method significantly suppresses gamma nonlinear responses and achieves high measurement accuracy and efficiency.

关键词： Optical filters Measurement errors Phase measurement Nonlinear distortion Measurement uncertainty Low-pass filters Optical distortion Harmonic analysis Distortion measurement Power harmonic filters

来源：评论

学校读者我要写书评

暂无评论

Quadruped robot target following based on a gimbal mechanism 18

Quadruped robot target following based on a gimbal mechanism

引用

18th IEEE Conference on Industrial Electronics and Applications, ICIEA 2023

作者： Chen, Xiuqi Cheng, Lei Xie, Zongquan Wuhan University of Science and Technology School of Information Science and Engineering Wuhan China Wuhan University of Science and Technology Institute of Robotics and Intelligent Systems Wuhan China

ISBN: (纸本)9798350312201

Human-robot collaboration is gradually becoming the way of life and production in the future. In this work, we focus on promoting the application scenarios of robots that can autonomously follow users, and propose a quadruped robot that can achieve autonomous following in complex environments. To address the problem that traditional quadruped robots using fixed visual sensors easily lose track of the target, we propose a method of using a gimbal mechanism equipped with a visual sensor for visual servoing tracking. By combining the chassis direction to plan the target servo strategy and controlling the gimbal rotation, we can achieve visual target following. On this basis, this paper controls the foot trajectory by using the cubic spline interpolation method, so that the robot can walk along various expected trajectories to achieve target following. This paper uses a field-of-view evaluation index to evaluate the tracking performance of this method and proves the effectiveness of the proposed target following method through experiments. © 2023 IEEE.

关键词： Computer vision

来源：评论

学校读者我要写书评

暂无评论

A Directed Jump Point Search with Improved Preprocess for Path Planning 18

A Directed Jump Point Search with Improved Preprocess for Pa...

引用

18th IEEE Conference on Industrial Electronics and Applications, ICIEA 2023

作者： Xie, Zongquan Cheng, Lei Li, Xiang Chen, Xiuqi Wuhan University of Science and Technology School of Information Science and Engineering Wuhan China Institute of Robotics and Intelligent Systems Wuhan University of Science and Technology Wuhan China

ISBN: (纸本)9798350312201

We introduce a new algorithm called directed JPS+(Jump Point Search Plus) that addresses the large memory overheads of JPS+ and improves its performance in cases where the obstacle scale is small. Our approach involves two main modifications. First, we reduce memory usage by recording only jump points in the preprocessing stage instead of eight labels for each nodes in JPS+. And we introduce a technique to connect these precomputed jump points into jump point set. Second, we apply an optimization to JPS+ that speeds up path planning in maps with small obstacle scale and it has no negative impact on other cases. We implement our algorithm in Python and evaluate its performance in terms of memory usage and planning time. Our results demonstrate that it saves memory and speeds up the path planning, particularly in maps with small obstacle scale. © 2023 IEEE.

关键词： Motion planning

来源：评论

学校读者我要写书评

暂无评论

Saliency-SLAM: Saliency Prediction for Visual SLAM

Saliency-SLAM: Saliency Prediction for Visual SLAM

引用

2023 Tianjin University-IET Electrical and information engineering Doctoral Forum, TJU IET 2023

作者： Jin, Sheng Meng, Qinghao Institute of Robotics and Autonomous Systems Tianjin Key Laboratory of Intelligent Unmanned Swarm Technology and System School of Electrical and Information Engineering Tianjin University Tianjin China

ISBN: (纸本)9781839539824

This paper proposes a novel approach for visual simultaneous localization and mapping (vSLAM), called Saliency-SLAM. This method effectively emulates the visual attention mechanism by integrating a saliency prediction model. In contrast to conventional vSLAM methods, which treat all feature points as equally significant, salient feature points are given greater importance in Saliency-SLAM. Therefore, we construct a novel saliency dataset specifically for vSLAM task to train a saliency prediction model by integrating multi-level visual information. Then, we propose to use the salient value of predicted result as the weight of significance for the feature points during the optimization process. Exhaustive experiments have been conducted on the KITTI dataset, demonstrating that our proposed method outperforms existing approaches in terms of localization accuracy. © The Institution of engineering & Technology 2023.

关键词： Forecasting

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：