检索结果-内蒙古大学图书馆

Generalization of neural network for manipulator inverse dynamics model learning

APPLIED INTELLIGENCE 2025年第7期55卷 1-20页

作者： Huang, Wenhui Lin, Yunhan Chen, Jie Liu, Mingxin Min, Huasong Wuhan Univ Sci & Technol Inst Robot & Intelligent Syst Heping St Wuhan 430080 Hubei Peoples R China

The inverse dynamics model of manipulators learned from recurrent neural networks demonstrates higher precision than those obtained through analytical modeling methods. Variations in end-effector loads and previously unseen trajectory points can lead to inaccurate torque estimations in dynamic models of manipulators. This paper integrates innovative feature expansion, feature enhancement, and regularization into an end-to-end inverse dynamics model learning framework. The proposed model employs a bidirectional long short-term memory (BiLSTM) network, augmented by a spatial attention mechanism with Convolutional Neural Networks (CNN) and a Max-Pooling method, which enhances the extraction of latent spatial features, and a multi-scale parallel temporal attention mechanism, which captures the dynamic changes of objects in the temporal dimension. A novel motion residual vector is designed to expand features, and a motion residual module is proposed to assist the network in perceiving changes in end-effector loads. To prevent overfitting, novel spatial attention standard deviation regularization are implemented. Experimental results across different trajectories and end-effector loads validate the generalization capability of the proposed method. The proposed method is compared with five methods, experimental results across different trajectories and end-effector loads validate the generalization capability of the proposed method. It surpasses state-of-the-art methods, achieving the highest overall accuracy. In cross-validation experiments, the validation loss remains stable as the training loss decreases, demonstrating the proposed approach's strong generalization performance in dynamics model learning.

关键词： Manipulator dynamics model learning Motion residual Spatiotemporal attention Velocity aware Normalization

来源：评论

学校读者我要写书评

暂无评论

An enhanced state-aware model learning approach for security analysis in lightweight protocol implementations

引用

JOURNAL OF CLOUD COMPUTING-ADVANCES SYSTEMS AND APPLICATIONS 2024年第1期13卷 28-28页

作者： Guo, Jiaxing Zhao, Dongliang Gu, Chunxiang Chen, Xi Zhang, Xieli Ju, Mengcheng Henan Key Lab Network Cryptog Technol Zhengzhou 450000 Peoples R China

Owing to the emergence and rapid advances of new-generation information and digitalization technologies, the concept of model-driven digital twin has received widespread attentions and is developing vigorously. Driven by data and simulators, the digital twin can create the virtual twins of physical objects to perform monitoring, simulation, prediction, optimization, and so on. Hence, the application of digital twin can increase efficiency and security of systems by providing reliable model and decision supports. In this paper, we propose a state-aware model learning method to simulate and analyze the lightweight protocol implementations in edge/cloud environments. We introduce the data flow of program execution and network interaction inputs/outputs (I/O) into the extended finite state machine (EFSM) to expand the modeling scope and insight. We aim to calibrate the states and construct an accurate state-machine model using a digital twin based layered approach to reasonably reflect the correlation of a device's external behavior and internal data. This, in turn, improves our ability to verify the logic and evaluate the security for protocol implementations. This method firstly involves instrumenting the target device to monitor variable activity during its execution. We then employ learning algorithms to produce multiple rounds of message queries. Both the I/O data corresponding to these query sequences and the state calibration information derived from filtered memory variables are obtained through the mapper and execution monitor, respectively. These two aspects of information are combined to dynamically and incrementally construct the protocol's state machine. We apply this method to develop SALearn and evaluate the effectiveness of SALearn on two lightweight protocol implementations. Our experimental results indicate that SALearn outperforms existing protocol model learning tools, achieving higher learning efficiency and uncovering more interesting states and se

关键词： model learning State aware Digital twin Lightweight protocol implementations

来源：评论

学校读者我要写书评

暂无评论

Velocity-aware spatial-temporal attention LSTM model for inverse dynamic model learning of manipulators

引用

FRONTIERS IN NEUROROBOTICS 2024年 18卷 1353879页

作者： Huang, Wenhui Lin, Yunhan Liu, Mingxin Min, Huasong Wuhan Univ Sci & Technol Inst Robot & Intelligent Syst Wuhan Hubei Peoples R China

Introduction An accurate inverse dynamics model of manipulators can be effectively learned using neural networks. However, further research is required to investigate the impact of spatiotemporal variations in manipulator motion sequences on network learning. In this work, the Velocity Aware Spatial-Temporal Attention Residual LSTM neural network (VA-STA-ResLSTM) is proposed to learn a more accurate inverse dynamics model, which uses a velocity-aware spatial-temporal attention mechanism to extract dynamic spatiotemporal features selectively from the motion sequence of the serial *** The multi-layer perception (MLP) attention mechanism is adopted to capture the correlation between joint position and velocity in the motion sequence, and the state correlation between hidden units in the LSTM network to reduce the weight of invalid features. A velocity-aware state fusion approach of LSTM network hidden units' states is proposed, which utilizes variation in joint velocity to adapt to the temporal characteristics of the manipulator dynamic motion, improving the generalization and accuracy of the neural *** Comparative experiments have been conducted on two open datasets and a self-built dataset. Specifically, the proposed method achieved an average accuracy improvement of 61.88% and 43.93% on the two different open datasets and 71.13% on the self-built dataset compared to the LSTM network. These results demonstrate a significant advancement in accuracy for the proposed *** Compared with the state-of-the-art inverse dynamics model learning methods of manipulators, the modeling accuracy of the proposed method in this paper is higher by an average of 10%. Finally, by visualizing attention weights to explain the training procedure, it was found that dynamic modeling only relies on partial features, which is meaningful for future optimization of inverse dynamic model learning methods.

关键词： manipulators dynamics model learning long short-term memory network (LSTM) spatial-temporal attention velocity aware

来源：评论

学校读者我要写书评

暂无评论

A model learning framework for inferring the dynamics of transmission rate depending on exogenous variables for epidemic forecasts

引用

COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING 2025年 437卷

作者： Ziarelli, Giovanni Pagani, Stefano Parolini, Nicola Regazzoni, Francesco Verani, Marco Politecn Milan Dept Math MOX Piazza Leonardo Da Vinci 32 I-20133 Milan Italy

In this work, we aim to formalize a novel scientific machine learning framework to reconstruct the hidden dynamics of the transmission rate, whose inaccurate extrapolation can significantly impair the quality of the epidemic forecasts, by incorporating the influence of exogenous variables (such as environmental conditions and strain-specific characteristics). We propose a hybrid model that blends a data-driven layer with a physics-based one. The data-driven layer is based on a neural ordinary differential equation that learns the dynamics of the transmission rate, conditioned on the meteorological data and wave-specific latent parameters. The physics- based layer, instead, consists of a standard SEIR compartmental model, wherein the transmission rate represents an input. The learning strategy follows an end-to-end approach: the loss function quantifies the mismatch between the actual numbers of infections and its numerical prediction obtained from the SEIR model incorporating as an input the transmission rate predicted by the neural ordinary differential equation. We apply this original approach to both a synthetic test case and a realistic test case based on meteorological data (temperature and humidity) and influenza data from Italy between 2010 and 2020. In both scenarios, we achieve low generalization error on the test set and observe strong alignment between the reconstructed model and established findings on the influence of meteorological factors on epidemic spread. Finally, we implement a data assimilation strategy to adapt the neural equation to the specific characteristics of an epidemic wave under investigation, and we conduct sensitivity tests on the network's hyperparameters.

关键词： Scientific machine learning model learning Data assimilation Epidemiology Forecasts Neural differential equations Hidden dynamics

来源：评论

学校读者我要写书评

暂无评论

Enhancing model learning in reinforcement learning through Q-function-guided trajectory alignment

引用

APPLIED INTELLIGENCE 2025年第7期55卷 1-22页

作者： Du, Xin Zhong, Shan Gong, Shengrong Si, Yali Qi, Zhenyu Changshu Inst Technol Sch Comp Sci & Engn Suzhou 215506 Jiangsu Peoples R China Suzhou Univ Sci & Technol Sch Elect & Informat Engn Suzhou 215506 Jiangsu Peoples R China Univ Arizona Tucson AZ 85711 USA

model-based reinforcement learning (MBRL) methods hold great promise for achieving excellent sample efficiency by fitting a dynamics model to previously observed data and leveraging it for RL or planning. However, the resulting trajectories may diverge from actual-world trajectories due to the accumulation of errors in multi-step model sampling, particularly for longer horizons. This undermines the performance of MBRL and significantly affects sample efficiency. Therefore, we present a trajectory alignment capable of aligning simulated trajectories with their real counterparts from any initial random state and with adaptive length, enabling the preparation of paired real-simulated samples to minimize compounding errors. Additionally, we design a Q-function function to estimate Q values for the paired real-simulated samples. The simulated samples whose Q-value difference from the real ones surpasses a given threshold will be discarded, thus preventing the model from over-fitting to erroneous samples. Experimental results demonstrate that both trajectory alignment and Q-function guided sample filtration contribute to improving policy and sample efficiency. Our method surpasses previous state-of-the-art model-based approaches in both sample efficiency and asymptotic performance across a series of challenging control tasks. The code is open source and available at https://***/duxin0618/***.

关键词： Reinforcement learning Dynamics model model-based RL algorithms model learning

来源：评论

学校读者我要写书评

暂无评论

model learning With Backlash Compensation for a Tendon-Driven Surgical Robot

引用

IEEE ROBOTICS AND AUTOMATION LETTERS 2022年第3期7卷 7958-7965页

作者： Cursi, Francesco Bai, Weibang Yeatman, Eric M. Kormushev, Petar Imperial Coll London Hamlyn Ctr London SW7 2BX England Imperial Coll London Robot Intelligence Lab London SW7 2BX England

Robots for minimally invasive surgery are becoming more and more complex, due to miniaturization and flexibility requirements. The vast majority of surgical robots are tendon-driven and this, along with the complex design, causes high nonlinearities in the system which are difficult to model analytically. In this work we analyse how incorporating a backlash model and compensation can improve model learning and control. We combine a backlash compensation technique and a Feedforward Artificial Neural Network (ANN) with differential relationships to learn the kinematics at position and velocity level of highly articulated tendon-driven robots. Experimental results show that the proposed backlash compensation is effective in reducing nonlinearities in the system, that compensating for backlash improves model learning and control, and that our proposed ANN outperforms traditional ANN in terms of path tracking accuracy.

关键词： model learning backlash compensation tendon-driven robots minimally invasive surgery

来源：评论

学校读者我要写书评

暂无评论

Global model learning for Large Deformation Control of Elastic Deformable Linear Objects: An Efficient and Adaptive Approach

引用

IEEE TRANSACTIONS ON ROBOTICS 2023年第1期39卷 417-436页

作者： Yu, Mingrui Lv, Kangchen Zhong, Hanzhong Song, Shiji Li, Xiang Tsinghua Univ Dept Automat Beijing 100084 Peoples R China CUHK Shenzhen Res Inst Shenzhen 518057 Peoples R China

Robotic manipulation of deformable linear objects (DLOs) has broad application prospects in many fields. However, a key issue is to obtain the exact deformation models (i.e., how robot motion affects DLO deformation), which are hard to theoretically calculate and vary among different DLOs. Thus, shape control of DLOs is challenging, especially for large deformation control which requires global and more accurate models. In this paper, we propose a coupled offline and online data-driven method for efficiently learning a global deformation model, allowing for both accurate modeling through offline learning and further updating for new DLOs via online adaptation. Specifically, the model approximated by a neural network is first trained offline on random data, then seamlessly migrated to the online phase, and further updated online during actual manipulation. Several strategies are introduced to improve the model's efficiency and generalization ability. We propose a convex-optimization-based controller, and analyze the system's stability using the Lyapunov method. Detailed simulations and real-world experiments demonstrate that our method can efficiently and precisely estimate the deformation model, and achieve large deformation control of untrained DLOs in 2D and 3D dual-arm manipulation tasks better than the existing methods. It accomplishes all 24 tasks with different desired shapes on different DLOs in the real world, using only simulation data for the offline learning.

关键词： Strain Deformable models Shape Jacobian matrices Data models Adaptation models Task analysis Deformable linear objects (DLOs) model learning robotic manipulation shape control

来源：评论

学校读者我要写书评

暂无评论

Active model learning of Git Version Control System

Active Model Learning of Git Version Control System

引用

17th IEEE International Conference on Software Testing, Verification, and Validation (ICST)

作者： Muskardin, Edi Burgstaller, Tamim Tappler, Martin Aichernig, Bernhard K. Graz Univ Technol Inst Software Technol DES Lab Silicon Austria Labs Graz Austria Graz Univ Technol Inst Software Technol Graz Austria

ISBN: (纸本)9798350344806;9798350344790

Git is a distributed version control system that enables developers to seamlessly collaborate on a project. It tracks changes made to the source code and implements various features upon which the modern software development process is built. Reliance on Git in the software development process requires comprehensive testing to ensure its functionalities operate reliably and consistently across projects. We show how active automata learning can enable model-based testing of Git. Active automata learning automatically derives models of the inputoutput behavior of Git and learned models provide a foundation for model-based testing. In addition, learned models could be used to ensure compatibility between Git versions.

关键词： active automata learning Git version control model learning

来源：评论

学校读者我要写书评

暂无评论

Unsupervised model learning of Pneumatic Musculoskeletal Robots with Inadequate Sensors

Unsupervised Model Learning of Pneumatic Musculoskeletal Rob...

引用

IEEE International Instrumentation and Measurement Technology Conference (I2MTC)

作者： Xu, Haoran Fan, Jianyin Ma, Hongxu Wang, Qiang Harbin Inst Technol Dept Control Sci & Engn Harbin Peoples R China

ISBN: (纸本)9798350380903;9798350380910

This paper provides an unsupervised model learning method for pneumatic musculoskeletal robots without torque sensors. In particular, we focus on learning the inverse dynamics model and employ multilayer perceptrons (MLPs) to capture the nonlinear and time-varying behaviors of pneumatic artificial muscles. To overcome the inadequacy of measurement devices that hinder the process of collecting the necessary input-output data for supervised training, a novel way to update MLPs is derived by coupling the process of parameter tuning with a model-based controller design scheme and solving for the condition that guarantees concurrent performance improvements of the controller and the learning model. The derived method applies to MLPs with any number of hidden layers and the word unsupervised refers to the fact that it is not built upon back-propagating a cost function. The effectiveness of our method is verified on a self-made musculoskeletal robot joint with one degree-of-freedom and the experiment results show that the learned model is accurate enough for both model-based trajectory tracking control and future state prediction.

关键词： model learning inverse dynamics pneumatic actuator

来源：评论

学校读者我要写书评

暂无评论

model learning and Knowledge Sharing for Cooperative Multiagent Systems in Stochastic Environment

引用

IEEE TRANSACTIONS ON CYBERNETICS 2021年第12期51卷 5717-5727页

作者： Jiang, Wei-Cheng Narayanan, Vignesh Li, Jr-Shin Tunghai Univ Dept Elect Engn Taichung 40704 Taiwan Washington Univ Dept Elect & Syst Engn St Louis MO 63130 USA

An imposing task for a reinforcement learning agent in an uncertain environment is to expeditiously learn a policy or a sequence of actions, with which it can achieve the desired goal. In this article, we present an incremental model learning scheme to reconstruct the model of a stochastic environment. In the proposed learning scheme, we introduce a clustering algorithm to assimilate the model information and estimate the probability for each state transition. In addition, utilizing the reconstructed model, we present an experience replay strategy to create virtual interactive experiences by incorporating a balance between exploration and exploitation, which greatly accelerates learning and enables planning. Furthermore, we extend the proposed learning scheme for a multiagent framework to decrease the effort required for exploration and to reduce the learning time in a large environment. In this multiagent framework, we introduce a knowledge-sharing algorithm to share the reconstructed model information among the different agents, as needed, and develop a computationally efficient knowledge fusing mechanism to fuse the knowledge acquired using the agents' own experience with the knowledge received from its teammates. Finally, the simulation results with comparative analysis are provided to demonstrate the efficacy of the proposed methods in the complex learning tasks.

关键词： Stochastic processes Task analysis Computational modeling Clustering algorithms Numerical models Multi-agent systems Fuses Knowledge sharing model learning multiagent system reinforcement learning (RL) sample efficiency

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：