检索结果-内蒙古大学图书馆

LLM-controller: Dynamic robot control adaptation using large language models

robotICS AND AUTONOMOUS SYSTEMS 2025年 186卷

作者： Zahedifar, Rasoul Baghshah, Mahdieh Soleymani Taheri, Alireza Sharif Univ Technol Mech Engn Dept Social & Cognit Robot Lab Tehran Iran Sharif Univ Technol Comp Engn Dept Machine Learning Lab Tehran Iran

In this study, a dynamic adaptation of a robot controller is investigated using large language models (LLMs). We propose our controller called the LLM-Controller, where, in response to changes in the system dynamics or reference signals, the LLM adapts the controller to the new context. Various scenarios reflecting real-world conditions, including unknown disturbances, unmodeled dynamics, and changing reference signals, were analyzed. Using the proposed LLM-Controller, one can adapt to new conditions automatically without manual tuning. Additionally, the controller's performance is investigated using different prompting techniques, such as zero-shot and few-shot chain-of-thought (COT), which facilitate step-by-step reasoning and improve adaptation to new contexts. The proposed scheme is applied to two case studies involving robot manipulators. First, it is tested on a 2-link robot manipulator, followed by a 3-link manipulator to enhance its generalizability. The algorithm's adaptability and effectiveness are further evaluated across a range of tasks and conditions, demonstrating its versatility in various scenarios. The results demonstrate that the LLM-Controller achieved a 100 % success rate in adapting the controller to new conditions for the 2-link manipulator, with a significant improvement in trial efficiency;while for the 3-link system, the controller maintained a 90 % success rate, showing greater adaptability to changes in reference signals or dynamic conditions in under 20 s. These outcomes could be further enhanced by employing a COT approach, potentially leading to higher success rates, fewer trials, and optimized costs. In contrast, the classic nonlinear adaptive controller struggled to adjust to the new conditions, while the LLM-Controller automatically adapts, guiding the system to new stable states. This research provides valuable insights into how LLMs can enhance decision-making, improving stability and performance in dynamic and uncertain environme

关键词： Large language models (LLMs) LLM-controller robot controller Dynamic adaptation Chain-of-thought (COT) Lyapunov theory Nonlinear adaptive controller

来源：评论

学校读者我要写书评

暂无评论

Dynamic Electromagnetic Navigation

引用

IEEE robotICS AND AUTOMATION LETTERS 2025年第6期10卷 6095-6102页

作者： Zughaibi, Jasan Nelson, Bradley J. Muehlebach, Michael Swiss Fed Inst Technol Multiscale Robot Lab CH-8092 Zurich Switzerland Max Planck Inst Intelligent Syst Learning & Dynam Syst Grp D-72076 Tubingen Germany

Magnetic navigation offers wireless control over magnetic objects, which has important medical applications, such as targeted drug delivery and minimally invasive surgery. Magnetic navigation systems are categorized into systems using permanent magnets and systems based on electromagnets. Electromagnetic Navigation Systems (eMNSs) are believed to have a superior actuation bandwidth, facilitating trajectory tracking and disturbance rejection. This greatly expands the range of potential medical applications and includes even dynamic environments as encountered in cardiovascular interventions. To showcase the dynamic capabilities of eMNSs, we successfully stabilize a (non-magnetic) inverted pendulum on the tip of a magnetically driven arm. Our approach employs a model-based framework that leverages Lagrangian mechanics to capture the interaction between the mechanical dynamics and the magnetic field. Using system identification, we estimate unknown parameters, the actuation bandwidth, and characterize the system's nonlinearity. To explore the limits of electromagnetic navigation and evaluate its scalability, we characterize the electrical system dynamics and perform reference measurements on a clinical-scale eMNS, affirming that the proposed dynamic control methodologies effectively translate to larger coil configurations. A state-feedback controller stabilizes the inherently unstable pendulum, and an iterative learning control scheme enables accurate tracking of non-equilibrium trajectories. Furthermore, to understand structural limitations of our control strategy, we analyze the influence of magnetic field gradients on the motion of the system. To our knowledge, this is the first demonstration to stabilize a 3D inverted pendulum through electromagnetic navigation.

关键词： Navigation Magnetic fields Heuristic algorithms Bandwidth Actuators Permanent magnets Electromagnetics Dynamics Catheters Vectors Inverted pendulum electromagnetic navigation iterative learning control

来源：评论

学校读者我要写书评

暂无评论

Unified Guidance and Jerk-Level Dynamic Inversion for Accurate Position Control of Hybrid UAVs

引用

IEEE TRANSACTIONS ON robotICS 2025年 41卷 708-728页

作者： Rohr, David Andersson, Olov Lawrance, Nicholas Stastny, Thomas Siegwart, Roland Swiss Fed Inst Technol Autonomous Syst Lab CH-8092 Zurich Switzerland KTH Royal Inst Technol Div Robot Percept & Learning S-10044 Stockholm Sweden CSIRO Data61 Robot Percept & Auton Grp Pullenvale Qld 4069 Australia

By combining rotary- and fixed-wing flight, hybrid uncrewed aerial vehicles (H-UAVs) can uniquely address missions combining long-range aerial transport and precise ground-relative tasks, such as the placement or retrieval of payloads. However, to leverage their full maneuverability, first, the fundamentally different operating modes of rotary- and fixed-wing vehicles need to be unified and second, the system be controlled precisely despite complex aerodynamic effects. This work presents a general and lightweight, cascaded control formulation for such versatile and accurate operation of H-UAVs. First, a novel guidance law unifies ground- and air-relative position control modes typical for the individual flight regimes. Second, we formulate a jerk-level feedback-linearization to accurately track the guidance outputs despite model errors and disturbances. In extensive real flight tests with a tiltwing H-UAV, we demonstrate the versatile allocation of (hybrid) flight states and the overall accuracy enabled by the control system. Position errors remain below 0.5 m (one quarter of the wingspan) in the full flight envelope, including accelerated maneuvers up to 10 ms(2) and gusting wind reaching 12 m/s. Finally, the control system demonstrates exploiting hybrid flight for transport-related missions with a precise, in-flight pickup of a payload.

关键词： Payloads Kinematics Position control Control systems Aerodynamics Accuracy Computational modeling Vehicle dynamics Velocity control Trajectory tracking Aerial systems aircraft guidance convertible and hybrid uncrewed aerial vehicles (H-UAV) dynamic inversion mechanics and control path following trajectory tracking

来源：评论

学校读者我要写书评

暂无评论

Isolated Kalman filtering: theory and decoupled estimator design

引用

AUTONOMOUS robotS 2025年第1期49卷 1-25页

作者： Jung, Roland Luft, Lukas Weiss, Stephan Univ Klagenfurt Control Networked Syst Grp Univ Str 65 A-9020 Klagenfurt Carinthia Austria Univ Freiburg Robot Learning Lab Georges Kohler Allee 80 D-79110 Freiburg Germany

In this paper, we propose a state decoupling strategy for Kalman filtering problems, when the dynamics of individual estimates are decoupled and their outputs are sparsely coupled. The algorithm is termed Isolated Kalman Filtering (IsoKF) and exploits the sparsity in the output coupling by applying approximations that mitigate the need for non-involved estimates. We prove that the approximations made during the isolated coupling of estimates are based on an implicit maximum determinant completion of the incomplete a priori covariance matrix. The steady state behavior is studied on eleven different observation graphs and a buffering scheme to support delayed (i.e. out-of-order) measurements is proposed. We discussed handling of delayed measurements in both, an optimal or a suboptimal way. The credibility of the isolated estimates are evaluated on a linear and nonlinear toy example in Monte Carlo simulations. The presented paradigm is made available online to the community within a generic C++ estimation framework supporting both, modular sensor fusion and collaborative state estimation.

关键词： Sensor fusion Kalman filter decoupling Distributed filtering

来源：评论

学校读者我要写书评

暂无评论

Sensorimotor learning With Stability Guarantees via Autonomous Neural Dynamic Policies

引用

IEEE robotICS AND AUTOMATION LETTERS 2025年第2期10卷 1760-1767页

作者： Totsila, Dionis Chatzilygeroudis, Konstantinos Modugno, Valerio Hadjivelichkov, Denis Kanoulas, Dimitrios Univ Lorraine Inria CNRS Loria F-54000 Nancy France Univ Patras Dept Math Computat Intelligence Lab CILab Patras 26504 Greece Univ Patras Dept Elect & Comp Engn Lab Automat & Robot LAR Patras 26504 Greece UCL Dept Comp Sci Robot Percept & Learning Lab RPL Lab London WC1E 6BT England Archimedes Athena RC Maroussi 15125 Greece

State-of-the-art sensorimotor learning algorithms, either in the context of reinforcement learning or imitation learning, offer policies that can often produce unstable behaviors, damaging the robot and/or the environment. Moreover, it is very difficult to interpret the optimized controller and analyze its behavior and/or performance. Traditional robot learning, on the contrary, relies on dynamical system-based policies that can be analyzed for stability/safety. Such policies, however, are neither flexible nor generic and usually work only with proprioceptive sensor states. In this work, we bridge the gap between generic neural network policies and dynamical system-based policies, and we introduce Autonomous Neural Dynamic Policies (ANDPs) that: (a) are based on autonomous dynamical systems, (b) always produce asymptotically stable behaviors, and (c) are more flexible than traditional stable dynamical system-based policies. ANDPs are fully differentiable, flexible generic-policies that accept any observation input, while ensuring asymptotic stability. Through several experiments, we explore the flexibility and capacity of ANDPs in several imitation learning tasks including experiments with image observations. The results show that ANDPs combine the benefits of both neural network-based and dynamical system-based methods.

关键词： robots Artificial neural networks Dynamical systems Stability analysis Asymptotic stability Aerospace electronics robot learning Imitation learning Trajectory Heuristic algorithms learning from demonstration machine learning for robot control sensorimotor learning

来源：评论

学校读者我要写书评

暂无评论

Rapid learning with phase-change memory-based in-memory computing through learning-to-learn

引用

NATURE COMMUNICATIONS 2025年第1期16卷 1-16页

作者： Ortner, Thomas Petschenig, Horst Vasilopoulos, Athanasios Renner, Roland Brglez, Spela Limbacher, Thomas Pinero, Enrique Linares-Barranco, Alejandro Pantazi, Angeliki Legenstein, Robert IBM Res Europe Zurich Ruschlikon Switzerland Graz Univ Technol Inst Machine Learning & Neural Computat Graz Austria Univ Seville SCORE Lab EPS Robot & Technol Comp ETSII Seville Spain

There is a growing demand for low-power, autonomously learning artificial intelligence (AI) systems that can be applied at the edge and rapidly adapt to the specific situation at deployment site. However, current AI models struggle in such scenarios, often requiring extensive fine-tuning, computational resources, and data. In contrast, humans can effortlessly adjust to new tasks by transferring knowledge from related ones. The concept of learning-to-learn (L2L) mimics this process and enables AI models to rapidly adapt with only little computational effort and data. In-memory computing neuromorphic hardware (NMHW) is inspired by the brain's operating principles and mimics its physical co-location of memory and compute. In this work, we pair L2L with in-memory computing NMHW based on phase-change memory devices to build efficient AI models that can rapidly adapt to new tasks. We demonstrate the versatility of our approach in two scenarios: a convolutional neural network performing image classification and a biologically-inspired spiking neural network generating motor commands for a real robotic arm. Both models rapidly learn with few parameter updates. Deployed on the NMHW, they perform on-par with their software equivalents. Moreover, meta-training of these models can be performed in software with high-precision, alleviating the need for accurate hardware models.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Safety Filtering While Training: Improving the Performance and Sample Efficiency of Reinforcement learning Agents

引用

IEEE robotICS AND AUTOMATION LETTERS 2025年第1期10卷 788-795页

作者： Bejarano, Federico Pizarro Brunke, Lukas Schoellig, Angela P. Univ Toronto Learning Syst & Robot Lab Toronto ON M5S 1A1 Canada Univ Toronto Robot Inst Toronto ON M5G 0C6 Canada Vector Inst Artificial Intelligence Toronto ON M5G 0C6 Canada Tech Univ Munich Munich Inst Robot & Machine Intelligence MIRMI D-80992 Munich Germany

Reinforcement learning (RL) controllers are flexible and performant but rarely guarantee safety. Safety filters impart hard safety guarantees to RL controllers while maintaining flexibility. However, safety filters can cause undesired behaviours due to the separation between the controller and the safety filter, often degrading performance and robustness. In this letter, we analyze several modifications to incorporating the safety filter in training RL controllers rather than solely applying it during evaluation. The modifications allow the RL controller to learn to account for the safety filter. This letter presents a comprehensive analysis of training RL with safety filters, featuring simulated and real-world experiments with a Crazyflie 2.0 drone. We examine how various training modifications and hyperparameters impact performance, sample efficiency, safety, and chattering. Our findings serve as a guide for practitioners and researchers focused on safety filters and safe RL.

关键词： Machine learning for robot control reinforcement learning robot safety safety filtering Machine learning for robot control reinforcement learning robot safety safety filtering

来源：评论

学校读者我要写书评

暂无评论

Optical Tactile Sensing for Aerial Multicontact Interaction: Design, Integration, and Evaluation

引用

IEEE TRANSACTIONS ON robotICS 2025年 41卷 364-377页

作者： Aucone, Emanuele Sferrazza, Carmelo Gregor, Manuel D'Andrea, Raffaello Mintchev, Stefano Swiss Fed Inst Technol Environm Robot Lab Dept Environm Syst Sci CH-8092 Zurich Switzerland Swiss Fed Inst Forest Snow & Landscape Res WSL CH-8903 Birmensdorf Switzerland Univ Calif Berkeley Robot Learning Lab Berkeley CA 94704 USA Swiss Fed Inst Technol Inst Dynam Syst & Control Dept Mech & Proc Engn CH-8092 Zurich Switzerland

Distributed tactile sensing for multiforce detection is crucial for various aerial robot interaction tasks. However, current contact sensing solutions on drones only exploit single end-effector sensors and cannot provide distributed multicontact sensing. Designed to be easily mounted at the bottom of a drone, we propose an optical tactile sensor that features a large and curved soft-sensing surface, a hollow structure and a new illumination system. Even when spaced only 2 cm apart, multiple contacts can be detected simultaneously using our software pipeline, which provides real-world quantities of 3-D contact locations (mm) and 3-D force vectors (N), with an accuracy of 1.5 mm and 0.17 N, respectively. We demonstrate the sensor's applicability and reliability onboard and in real time with two demos related to, first, the estimation of the compliance of different perches and subsequent realignment and landing on the stiffer one, and second, the mapping of sparse obstacles. The implementation of our distributed tactile sensor represents a significant step toward attaining the full potential of drones as versatile robots capable of interacting with and navigating within complex environments.

关键词： Sensors robot sensing systems Drones Tactile sensors robots Optical sensors Force Optical feedback Sensor arrays Navigation Aerial robots: applications physical interactions by touching system-level integration of tactile sensors in robots tactile sensors design

来源：评论

学校读者我要写书评

暂无评论

TryOn-Adapter: Efficient Fine-Grained Clothing Identity Adaptation for High-Fidelity Virtual Try-On

引用

INTERNATIONAL JOURNAL OF COMPUTER VISION 2025年第6期133卷 3781-3802页

作者： Xing, Jiazheng Xu, Chao Qian, Yijie Liu, Yang Dai, Guang Sun, Baigui Liu, Yong Wang, Jingdong Zhejiang Univ Coll Control Sci & Engn Lab Adv Percept Robot & Intelligent Learning Hangzhou 310027 Zhejiang Peoples R China Alibaba Grp Hangzhou Peoples R China State Grid Shaanxi Elect Power Co SGIT AI Lab Xian Peoples R China Baidu Inc Beijing Peoples R China

Virtual try-on focuses on adjusting the given clothes to fit a specific person seamlessly while avoiding any distortion of the patterns and textures of the garment. However, the clothing identity uncontrollability and training inefficiency of existing diffusion-based methods, which struggle to maintain the identity even with full parameter training, are significant limitations that hinder the widespread applications. In this work, we propose an effective and efficient framework, termed TryOn-Adapter. Specifically, we first decouple clothing identity into fine-grained factors: style for color and category information, texture for high-frequency details, and structure for smooth spatial adaptive transformation. Our approach utilizes a pre-trained exemplar-based diffusion model as the fundamental network, whose parameters are frozen except for the attention layers. We then customize three lightweight modules (Style Preserving, Texture Highlighting, and Structure Adapting) incorporated with fine-tuning techniques to enable precise and efficient identity control. Meanwhile, we introduce the training-free T-RePaint strategy to further enhance clothing identity preservation while maintaining the realistic try-on effect during the inference. Our experiments demonstrate that our approach achieves state-of-the-art performance on two widely-used benchmarks. Additionally, compared with recent full-tuning diffusion-based methods, we only use about half of their tunable parameters during training. The code will be made publicly available at https://***/jiazheng-xing/TryOn-Adapter.

关键词： Virtual Try-On Large-scale generative models Diffusion model Identity preservation

来源：评论

学校读者我要写书评

暂无评论

One-Shot Dual-Arm Imitation learning

arXiv

引用

arXiv 2025年

作者： Wang, Yilong Johns, Edward The Robot Learning Lab Imperial College London United Kingdom

We introduce One-Shot Dual-Arm Imitation learning (ODIL), which enables dual-arm robots to learn precise and coordinated everyday tasks from just a single demonstration of the task. ODIL uses a new three-stage visual servoing (3-VS) method for precise alignment between the end-effector and target object, after which replay of the demonstration trajectory is sufficient to perform the task. This is achieved without requiring prior task or object knowledge, or additional data collection and training following the single demonstration. Furthermore, we propose a new dual-arm coordination paradigm for learning dual-arm tasks from a single demonstration. ODIL was tested on a real-world dual-arm robot, demonstrating state-of-the-art performance across six precise and coordinated tasks in both 4-DoF and 6-DoF settings, and showing robustness in the presence of distractor objects and partial occlusions. Videos are available at: https://***-learning. uk/one-shot-dual-arm. © 2025, CC BY.

关键词： Adversarial machine 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：