检索结果-内蒙古大学图书馆

Kinematics Analysis and Singularity Avoidance of a Parallel Mechanism with Kinematic Redundancy

Chinese Journal of Mechanical engineering 2022年第5期35卷 244-261页

作者： Chaoyu Shen Haibo Qu Sheng Guo Xiao Li Robotics Research Center School of MechanicalElectronic and Control EngineeringBeijing Jiaotong UniversityBeijing 100044China China Waterborne Transport Research Institute Beijing 100088China Key Laboratory of Vehicle Advanced Manufacturing Measuring and Control TechnologyMinistry of EducationBeijing Jiaotong UniversityBeijing 100044China

The kinematic redundancy is considered as a way to improve the performance of the parallel *** this paper,the kinematics performance of a three degree-of-freedom parallel mechanism with kinematic redundancy(3-DOF PM-KR)and the influence of redundant parts on the PM-KR are ***,the kinematics model of the PM-KR is *** inverse solutions,the Jacobian matrix,and the workspace of the PM-KR are ***,the influence of redundancy on the PM-KR is *** there exists kinematic redundancy,the PM-KR possesses fault-tolerant *** locking one actuating joint or two actuating joints simultaneously,the fault-tolerant workspace is *** the position of the redundant part is changed,the workspace and singularity will be *** results show that kinematic redundancy can be used to avoid ***,the simulations are performed to prove the theoretical analysis.

关键词： Parallel mechanism Kinematic redundancy Singularity Simulation analysis

来源：评论

学校读者我要写书评

暂无评论

CaKDP: Category-Aware Knowledge Distillation and Pruning Framework for Lightweight 3D Object Detection

CaKDP: Category-Aware Knowledge Distillation and Pruning Fra...

引用

Conference on Computer Vision and Pattern Recognition (CVPR)

作者： Haonan Zhang Longjun Liu Yuqi Huang Zhao Yang Xinyu Lei Bihan Wen National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center for Visual Information and Applications and Institute of Artificial Intelligence and Robotics Xi'an Jiaotong University Nanyang Technological University

ISBN: (数字)9798350353006

ISBN: (纸本)9798350353013

Knowledge distillation (KD) possesses immense potential to accelerate the deep neural networks (DNNs) for LiDAR-based 3D detection. However, in most of prevailing approaches, the suboptimal teacher models and insufficient student architecture investigations limit the performance gains. To address these issues, we propose a simple yet effective Category-aware Knowledge Distillation and Pruning (CaKDP) framework for compressing 3D detectors. Firstly, CaKDP transfers the knowledge of two-stage detector to one-stage student one, mitigating the impact of inadequate teacher models. To bridge the gap between the heterogeneous detectors, we investigate their differences, and then introduce the student-motivated category-aware KD to align the category prediction between distillation pairs. Secondly, we propose a category-aware pruning scheme to obtain the customizable architecture of compact student model. The method calculates the category prediction gap before and after removing each filter to evaluate the importance of filters, and retains the important filters. Finally, to further improve the student performance, a modified IOU-aware refinement module with negligible computations is leveraged to remove the redundant false positive predictions. Experiments demonstrate that CaKDP achieves the compact detector with high performance. For example, on WOD, CaKDP accelerates centerPoint by half while boosting L2 mAPH by 1.61%. The code is available at https://***/zhnxjtu/CaKDP.

关键词： Knowledge engineering Computer vision Three-dimensional displays Computational modeling Detectors Computer architecture Object detection

来源：评论

学校读者我要写书评

暂无评论

Imagine, Initialize, and Explore: An Effective Exploration Method in Multi-Agent Reinforcement Learning

arXiv

引用

arXiv 2024年

作者： Liu, Zeyang Wan, Lipeng Yang, Xinrui Chen, Zhuoran Chen, Xingyu Lan, Xuguang National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center for Visual Information and Application Institute of Artificial Intelligence and Robotics Xi’an Jiaotong University Xi’an710049 China

Effective exploration is crucial to discovering optimal strategies for multi-agent reinforcement learning (MARL) in complex coordination tasks. Existing methods mainly utilize intrinsic rewards to enable committed exploration or use role-based learning for decomposing joint action spaces instead of directly conducting a collective search in the entire action-observation space. However, they often face challenges obtaining specific joint action sequences to reach successful states in long-horizon tasks. To address this limitation, we propose Imagine, Initialize, and Explore (IIE), a novel method that offers a promising solution for efficient multi-agent exploration in complex scenarios. IIE employs a transformer model to imagine how the agents reach a critical state that can influence each other’s transition functions. Then, we initialize the environment at this state using a simulator before the exploration phase. We formulate the imagination as a sequence modeling problem, where the states, observations, prompts, actions, and rewards are predicted autoregressively. The prompt consists of timestep-to-go, return-to-go, influence value, and one-shot demonstration, specifying the desired state and trajectory as well as guiding the action generation. By initializing agents at the critical states, IIE significantly increases the likelihood of discovering potentially important under-explored regions. Despite its simplicity, empirical results demonstrate that our method outperforms multi-agent exploration baselines on the StarCraft Multi-Agent Challenge (SMAC) and SMACv2 environments. Particularly, IIE shows improved performance in the sparse-reward SMAC tasks and produces more effective curricula over the initialized states than other generative methods, such as CVAE-GAN and diffusion models. © 2024, CC BY.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Robust Noisy Label Learning via Two-Stream Sample Distillation

arXiv

引用

arXiv 2024年

作者： Bai, Sihan Zhou, Sanping Qin, Zheng Wang, Le Zheng, Nanning The National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center for Visual Information and Applications Institute of Artificial Intelligence and Robotics Xi’an Jiaotong University Shaanxi 710049 China

Noisy label learning aims to learn robust networks under the supervision of noisy labels, which plays a critical role in deep learning. Existing work either conducts s ample selection or label correction to deal with noisy labels during the model training process. In this paper, we design a simple yet effective sample selection framework, termed Two-Stream Sample Distillation (TSSD), for noisy label learning, which can extract more high-quality samples with clean labels to improve the robustness of network training. Firstly, a novel Parallel Sample Division (PSD) module is designed to generate a certain training set with sufficient reliable positive and negative samples by jointly considering the sample structure in feature space and the human prior in loss space. Secondly, a novel Meta Sample Purification (MSP) module is further designed to mine adequate semi-hard samples from the remaining uncertain training set by learning a strong meta classifier with extra golden data. As a result, more and more high-quality samples will be distilled from the noisy training set to train networks robustly in every iteration. Extensive experiments on four benchmark datasets, including CIFAR-10, CIFAR-100, Tiny-ImageNet, and Clothing-1M, show that our method has achieved state-of-the-art results over its competitors. Copyright © 2024, The Authors. All rights reserved.

关键词： Image classification

来源：评论

学校读者我要写书评

暂无评论

Maximizing Resource Efficiency in Cloud Data centers through Knowledge-Based Flower Pollination Algorithm (KB-FPA)

引用

Computers, Materials & Continua 2024年第6期79卷 3757-3782页

作者： Nidhika Chauhan Navneet Kaur Kamaljit Singh Saini Sahil Verma Kavita Ruba Abu Khurma Pedro A.Castillo University Institute of Computing Department Chandigarh UniversityPunjab140413India Department of Computer Science and Engineering Chandigarh UniversityPunjab140413India Universidade Federal do Piauí TeresinaPiauí64049-550Brazil MEU Research Unit Faculty of Information TechnologyMiddle East UniversityAmman11831Jordan Applied Science Research Center Applied Science Private UniversityAmman11931Jordan Department of Computer Engineering Automatics and RoboticsUniversity of GranadaGranada18071Spain

Cloud computing is a dynamic and rapidly evolving field,where the demand for resources fluctuates *** paper delves into the imperative need for adaptability in the allocation of resources to applications and services within cloud computing *** motivation stems from the pressing issue of accommodating fluctuating levels of user demand *** adhering to the proposed resource allocation method,we aim to achieve a substantial reduction in energy *** reduction hinges on the precise and efficient allocation of resources to the tasks that require those most,aligning with the broader goal of sustainable and eco-friendly cloud computing *** enhance the resource allocation process,we introduce a novel knowledge-based optimization *** this study,we rigorously evaluate its efficacy by comparing it to existing algorithms,including the Flower Pollination Algorithm(FPA),Spark Lion Whale Optimization(SLWO),and Firefly *** findings reveal that our proposed algorithm,Knowledge Based Flower Pollination Algorithm(KB-FPA),consistently outperforms these conventional methods in both resource allocation efficiency and energy consumption *** paper underscores the profound significance of resource allocation in the realm of cloud *** addressing the critical issue of adaptability and energy efficiency,it lays the groundwork for a more sustainable future in cloud computing *** contribution to the field lies in the introduction of a new resource allocation strategy,offering the potential for significantly improved efficiency and sustainability within cloud computing infrastructures.

关键词： Cloud computing resource allocation energy consumption optimization algorithm flower pollination algorithm

来源：评论

学校读者我要写书评

暂无评论

TPR: topology-preserving reservoirs for generalized zero-shot learning 24

TPR: topology-preserving reservoirs for generalized zero-sho...

引用

Proceedings of the 38th International Conference on Neural Information Processing Systems

作者： Hui Chen Yanbin Liu Yongqiang Ma Nanning Zheng Xin Yu National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center of Visual Information and Applications and Institute of Artificial Intelligence and Robotics Xi'an Jiaotong University Auckland University of Technology The University of Queensland

ISBN: (纸本)9798331314385

Pre-trained vision-language models (VLMs) such as CLIP have shown excellent performance for zero-shot classification. Based on CLIP, recent methods design various learnable prompts to evaluate the zero-shot generalization capability on a base-to-novel setting. This setting assumes test samples are already divided into either base or novel classes, limiting its application to realistic scenarios. In this paper, we focus on a more challenging and practical setting: generalized zero-shot learning (GZSL), i.e., testing with no information about the base/novel division. To address this challenging zero-shot problem, we introduce two unique designs that enable us to classify an image without the need of knowing whether it comes from seen or unseen classes. Firstly, most existing methods only adopt a single latent space to align visual and linguistic features, which has a limited ability to represent complex visual-linguistic patterns, especially for fine-grained tasks. Instead, we propose a dual-space feature alignment module that effectively augments the latent space with a novel attribute space induced by a well-devised attribute reservoir. In particular, the attribute reservoir consists of a static vocabulary and learnable tokens complementing each other for flexible control over feature granularity. Secondly, finetuning CLIP models (e.g., prompt learning) on seen base classes usually sacrifices the model's original generalization capability on unseen novel classes. To mitigate this issue, we present a new topology-preserving objective that can enforce feature topology structures of the combined base and novel classes to resemble the topology of CLIP. In this manner, our model will inherit the generalization ability of CLIP through maintaining the pairwise class angles in the attribute space. Extensive experiments on twelve object recognition datasets demonstrate that our model, termed Topology-Preserving Reservoir (TPR), outperforms strong baselines including both pro

关键词：

来源：评论

学校读者我要写书评

暂无评论

Grounded Answers for Multi-agent Decision-making Problem through Generative World Model

arXiv

引用

arXiv 2024年

作者： Liu, Zeyang Yang, Xinrui Sun, Shiguang Qian, Long Wan, Lipeng Chen, Xingyu Lan, Xuguang National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center for Visual Information and Application Institute of Artificial Intelligence and Robotics Xi'an Jiaotong University Xi’an710049 China

Recent progress in generative models has stimulated significant innovations in many fields, such as image generation and chatbots. Despite their success, these models often produce sketchy and misleading solutions for complex multi-agent decision-making problems because they miss the trial-and-error experience and reasoning as humans. To address this limitation, we explore a paradigm that integrates a language-guided simulator into the multi-agent reinforcement learning pipeline to enhance the generated answer. The simulator is a world model that separately learns dynamics and reward, where the dynamics model comprises an image tokenizer as well as a causal transformer to generate interaction transitions autoregressively, and the reward model is a bidirectional transformer learned by maximizing the likelihood of trajectories in the expert demonstrations under language guidance. Given an image of the current state and the task description, we use the world model to train the joint policy and produce the image sequence as the answer by running the converged policy on the dynamics model. The empirical results demonstrate that this framework can improve the answers for multi-agent decision-making problems by showing superior performance on the training and unseen tasks of the StarCraft Multi-Agent Challenge benchmark. In particular, it can generate consistent interaction sequences and explainable reward functions at interaction states, opening the path for training generative models of the future. © 2024, CC BY.

关键词： Chatbots

来源：评论

学校读者我要写书评

暂无评论

POMDP-Based Real-Time Path Planning for Manipulation of Multiple Microparticles via Optoelectronic Tweezers

引用

Cyborg and Bionic Systems 2022年第1期2022卷 303-313页

作者： Jiaxin Liu Huaping Wang Menghua Liu Ran Zhao Yanfeng Zhao Tao Sun Qing Shi Intelligent Robotics Institute School of Mechatronical Engineering Bejing Institute of TechnologyBeijing 10081China Key Laboratory of Biomimetic Robots and Systems(Beijing Institute of Technology) Ministry of EducationBeijing 100081China School of Medical Technology Beijing Institute of TechnologyBeijing 100081China Bejing Advanced Innovation Center for Itelligent Robots and Systems Beijing Institute of TechnologyBeijing 100081China

With high throughput and high flexibility,optoelectronic tweerers(OETs)hold huge potential for massively parallel ***,the trajectory of the virtual electrode has been planned in advance in most synchronous manipulations for multiple targets based on an optically induced dielectrophoresis(ODEP)mechanism,which is insuficient to ensure the stability and eficiency in an environment with potential collision *** this paper,a synchronously discretized manipulation method based on a centralized and decoupled path planner is proposed for transporting microparticles of diferent types with an OET *** approach based on the K uhn-Munkres algorithm is utilized to achieve the goal assignment between target microparticles and goal *** the assistance of a visual feedback module,a path planning approach based on the POMDP algorithm dynamically determines the motion strategies of the particle movement to avoid potential *** geometrical parameters of the virtual electrodes are optimized for different types of particles with the goal of maximum transport *** experiments of micropatterning with different morphologies and transporting multiple micropartides(e.g,polystyrene microspheres and 3T3 cells)to goal positions are *** results demonstrate that the proposed manipulation method based on optoelectronic tweezers is effective for multicell transport and promises to be used in biomedical manipulation tasks with high flexibility and efficiency.

关键词： manipulation positions assignment

来源：评论

学校读者我要写书评

暂无评论

MDRPC: Music-Driven Robot Primitives Choreography 36

MDRPC: Music-Driven Robot Primitives Choreography

引用

36th IEEE International Conference on Tools with Artificial Intelligence, ICTAI 2024

作者： Guan, Haiyang Wei, Xiaoyi Long, Weifan Yang, Dingkang Zhai, Peng Zhang, Lihua Fudan University Academy for Engineering and Technology Shanghai200433 China Ji Hua Laboratory Foshan528251 China Engineering Research Center of AI and Robotics Ministry of Education Shanghai200433 China Institute of Meta-Medical Fudan University Shanghai200433 China Jilin Provincial Key Laboratory of Intelligence Science and Engineering Changchun130013 China

ISBN: (纸本)9798331527235

Dance has been an important art form and means of communication since the dawn of human civilization. Equipping humanoid robots with the ability to perform smooth dance movements to music is a key research priority in artificial intelligence, robotics and human-computer interaction. However, existing kinematics-based dance generation methods often violate real-world physical laws as they do not consider physical constraints, leading to unrealistic movements. Additionally, due to the diversity and dynamic variability of input music, most existing physics-based methods, which rely on task-specific reward functions, face significant challenges in effectively handling music-driven dance generation tasks. To address these issues, we introduce MDRPC, the first physics-based, music-driven dance generation method for humanoid robots. Inspired by human choreographic principles, MDRPC is defined as a two-phase framework. The initial phase utilizes adversarial imitation learning to acquire a rich set of reusable dance primitives from a music-dance dataset. In the subsequent phase, these dance primitives are orchestrated under the guidance of musical theory and choreographic rules to generate complex humanoid dance sequences. Specifically, we propose beat alignment and dance diversity reward functions to synchronize motion rhythms with music beats and enhance the diversity of dance movements. We implement MDRPC on a simulated humanoid robot, and the results confirm that our method effectively controls the humanoid, enabling it to perform dance movements harmoniously synchronized with the music. © 2024 IEEE.

关键词： Anthropomorphic robots

来源：评论

学校读者我要写书评

暂无评论

OCViTnet: Pulse Wave Learning for Cuffless Blood Pressure Estimation of Medical Service Robots

OCViTnet: Pulse Wave Learning for Cuffless Blood Pressure Es...

引用

IEEE International Conference on robotics and Biomimetics

作者： Yuhui Cen Jingjing Luo Hongbo Wang Li Chen Xing Zhu Jingchun Luo Human Phenome Institute Fudan University Shanghai China The Institute of AI and Robotics Academy for Engineering and Technology Fudan University Ministry of Education Engineering Research Center of AI & Robotics Shanghai P.R.China

Blood pressure (BP) estimation through cuffless approaches is gaining prominence in precision healthcare, and the emergence of robotic medical services adds impetus to this trend with enhanced diagnostic capabilities. In our endeavor to enable medical service robots for cuffless BP self-estimation, we propose OCViTnet, a deep-learning network model grounded in pulse wave analysis (PWA). The model introduces a novel waveform processing architecture: an omni-scale convolution subnet captures single-period waveform characteristics, the Vision Transformer subnet extracts multi-period waveform features, and a regressor fuses phenotypic information with pulse wave features and performs BP regression. Experimental results perform exceptionally well in non-contact BP estimation from forehead imaging photoplethysmography. Additionally, this paper pioneers the utilization of the radial artery pulse wave for BP estimation. Leveraging the British Hypertension Society standard, our approach yields notable results, assessing the pulse diagnosis robot’s BP estimation performance at grade C/B (systolic/diastolic BP). This effort advances the potential of cuffless BP estimation in pulse diagnosis robotics and underscores the practical viability of our proposed intelligent methodology based on PWA.

关键词：

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：