检索结果-内蒙古大学图书馆

learning Active Force–Torque Based Policy for Sub-mm Localization of Unseen Holes

IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS 2024年第4期20卷 6726-6738页

作者： Xie, Liang Zhang, Haodong Zhao, Yinghao Zhang, Yu Xiong, Rong Ren, Yi Wang, Yue Zhejiang Univ State Key Lab Ind Control Technol Hangzhou 310058 Zhejiang Peoples R China Zhejiang Univ Inst Cyber Syst & Control Hangzhou 310058 Zhejiang Peoples R China Huawei Technol Adv Mfg Lab Shenzhen 518129 Peoples R China

Hole localization is crucial in the peg-in-hole process. Our goal is to enable robots to operate effectively in contact-rich environments with tight tolerances, and adapt to new tasks involving unseen peg-hole pairs. Most existing "black-box" methods train a policy that performs the task directly from perceptual inputs, which requires extensive real-world interactions for task adaptation. Departing from this direct mapping paradigm, our work propose to formulate the task as a force matching and localization problem, where the objective is to establish correspondences between current and template force-torque observation maps for localization purpose. The formulation enables the design of a decoupled map-locator-policy framework, offering improved success rates, efficiency, and augmented generalization capabilities, surpassing current state-of-the-art methods. Experiments demonstrate the effectiveness of the proposed method, achieving a 90% success rate across 12 unseen 3-D models and a variety of unseen tight workpieces. Within a mere 5-min adaption process, the performance can be further improved by more than 95%.

关键词： Location awareness Task analysis Transformers Robots Force Data models Adaptation models deep learning in robotics and automation peg-in-hole assembly robotic manipulation

来源：评论

学校读者我要写书评

暂无评论

Evetac: An Event-Based Optical Tactile Sensor for Robotic Manipulation

引用

IEEE TRANSACTIONS ON robotics 2024年 40卷 3812-3832页

作者： Funk, Niklas Helmut, Erik Chalvatzaki, Georgia Calandra, Roberto Peters, Jan Tech Univ Darmstadt Comp Sci Dept D-64289 Darmstadt Germany Hessian Ctr Artificial Intelligence D-64293 Darmstadt Germany Tech Univ Dresden LASR Lab D-01069 Dresden Germany Ctr Tactile Internet Human In The Loop D-01062 Dresden Germany German Res Ctr AI DFKI SAIROL D-64293 Darmstadt Germany Ctr Cognit Sci D-64283 Darmstadt Germany

Optical tactile sensors have recently become popular. They provide high spatial resolution, but struggle to offer fine temporal resolutions. To overcome this shortcoming, we study the idea of replacing the RGB camera with an event-based camera and introduce a new event-based optical tactile sensor called Evetac. Along with hardware design, we develop touch processing algorithms to process its measurements online at 1000 Hz. We devise an efficient algorithm to track the elastomer's deformation through the imprinted markers despite the sensor's sparse output. Benchmarking experiments demonstrate Evetac's capabilities of sensing vibrations up to 498 Hz, reconstructing shear forces, and significantly reducing data rates compared to RGB optical tactile sensors. Moreover, Evetac's output and the marker tracking provide meaningful features for learning data-driven slip detection and prediction models. The learned models form the basis for a robust and adaptive closed-loop grasp controller capable of handling a wide range of objects. We believe that fast and efficient event-based tactile sensors like Evetac will be essential for bringing human-like manipulation capabilities to robotics.

关键词： Robots Tactile sensors Optical sensors Cameras Robot vision systems Optical feedback Spatial resolution deep learning in robotics and automation event-based sensing force and tactile sensing perception for grasping and manipulation

来源：评论

学校读者我要写书评

暂无评论

High-Fidelity Integrated Aerial Platform Simulation for Control, Perception, and learning

引用

IEEE TRANSACTIONS ON automation SCIENCE AND ENGINEERING 2025年 22卷 13662-13683页

作者： Du, Jianrui Wang, Kaidi Fan, Yingjun Lai, Ganghua Yu, Yushu Beijing Inst Technol Sch Mechatron Engn Beijing 100081 Peoples R China

This paper presents a simulator framework tailored Integrated Aerial Platforms (IAPs) using multiple quadrotors. Our framework prioritizes photo and contact fidelity, achieved through a modular design that balances rendering and dynamics computation. Key features include: i) support for diverse IAP configurations;ii) a customizable physics engine for realistic motion and contact simulation for aerial manipulation;and iii) Unreal Engine 5 for lifelike rendering, with sensor designs for visual-inertial SLAM positioning simulation. We showcase our framework's versatility through a range of scenarios, including trajectory tracking for both fully and under-actuated IAPs, peg-in-hole and direct wrench control tasks under external wrench influence, tightly-coupled SLAM positioning with physical constraints, and air docking task training and testing using offline-to-online reinforcement learning. Furthermore, we validate our simulator framework's fidelity by comparing results with real flight data for trajectory tracking and direct wrench control tasks. Our simulator framework promises to be valuable for developing and testing integrated aerial platform systems for aerial manipulation.

关键词： Robots Engines Physics Drones Simultaneous localization and mapping Training Aircraft Rendering (computer graphics) Propellers Vehicle dynamics Simulation and animation aerial systems: perception and autonomy SLAM deep learning in robotics and automation

来源：评论

学校读者我要写书评

暂无评论

learning a Low-Dimensional Representation of a Safe Region for Safe Reinforcement learning on Dynamical Systems

引用

IEEE TRANSACTIONS ON NEURAL NETWORKS AND learning SYSTEMS 2023年第5期34卷 2513-2527页

作者： Zhou, Zhehua Oguz, Ozgur S. Leibold, Marion Buss, Martin Tech Univ Munich Chair Automat Control Engn D-80290 Munich Germany Univ Stuttgart Max Planck Inst Intelligent Syst D-70569 Stuttgart Germany

For the safe application of reinforcement learning algorithms to high-dimensional nonlinear dynamical systems, a simplified system model is used to formulate a safe reinforcement learning (SRL) framework. Based on the simplified system model, a low-dimensional representation of the safe region is identified and used to provide safety estimates for learning algorithms. However, finding a satisfying simplified system model for complex dynamical systems usually requires a considerable amount of effort. To overcome this limitation, we propose a general data-driven approach that is able to efficiently learn a low-dimensional representation of the safe region. By employing an online adaptation method, the low-dimensional representation is updated using the feedback data to obtain more accurate safety estimates. The performance of the proposed approach for identifying the low-dimensional representation of the safe region is illustrated using the example of a quadcopter. The results demonstrate a more reliable and representative low-dimensional representation of the safe region compared with previous works, which extends the applicability of the SRL framework.

关键词： Safety Computational modeling Heuristic algorithms Reinforcement learning Control systems Probabilistic logic Nonlinear dynamical systems Data-driven model order reduction deep learning in robotics and automation learning and adaptive systems safe reinforcement learning (SRL)

来源：评论

学校读者我要写书评

暂无评论

The Foreseeable Future: Self-Supervised learning to Predict Dynamic Scenes for Indoor Navigation

引用

IEEE TRANSACTIONS ON robotics 2023年第6期39卷 4581-4599页

作者： Thomas, Hugues Zhang, Jian Barfoot, Timothy D. Univ Toronto Inst Aerosp Studies UTIAS Toronto ON M5S 1A1 Canada Apple Inc Cupertino CA 95014 USA

We present a method for generating, predicting, and using spatiotemporal occupancy grid maps (SOGM), which embed future semantic information of real dynamic scenes. We present an autolabeling process that creates SOGMs from noisy real navigation data. We use a 3-D-2-D feedforward architecture, trained to predict the future time steps of SOGMs, given 3-D Lidar frames as input. Our pipeline is entirely self-supervised, thus enabling lifelong learning for real robots. The network is composed of a 3-D back-end that extracts rich features and enables the semantic segmentation of the lidar frames, and a 2-D front-end that predicts the future information embedded in the SOGM representation, potentially capturing the complexities and uncertainties of real-world multiagent interactions. We also design a navigation system that uses these predicted SOGMs within planning, after they have been transformed into spatiotemporal risk maps. We verify our navigation system's abilities in simulation, validate it on a real robot, study SOGM predictions on real data in various circumstances, and provide a novel indoor 3-D lidar dataset, collected during our experiments, which includes our automated annotations.

关键词： deep learning in robotics and automation indoor navigation learning and adaptive systems reactive and sensor-based planning

来源：评论

学校读者我要写书评

暂无评论

Online Model Predictive Control of Robot Manipulator With Structured deep Koopman Model

引用

IEEE robotics AND automation LETTERS 2023年第5期8卷 3102-3109页

作者： Zhang, Jinxin Wang, Hongze Chinese Acad Sci Inst Automation State Key Lab Management & Control Complex Syst Beijing 100190 Peoples R China Univ Chinese Acad Sci Sch Artificial Intelligence Beijing 100049 Peoples R China

The model predictive control (MPC) can provide the benefit of optimality (sub-optimality, exactly speaking) and explicitly treat hard constraints in both states and inputs, which makes it an attractive approach in the fields of robotics. However, the performance of this approach heavily depends on the system model and it is computationally intensive, which hinders its application in the real-time control of robotic systems with fast dynamics, such as the robotic manipulators. Data-driven modeling approaches based on the Koopman operator have the potential to remove the barriers to adopting the MPC in robotics, through learning a globally linear model. In this letter, we propose a novel Koopman model-the structured deep Koopman model, which can improve the accuracy of the learned linear model and reduce the number of states in the lifted space, through exploiting the deep Lipschitz neural network and making the lifted dynamics structured. We also prove the rationality of the presented method and provide a new perspective on Koopman operator-based models, which brings the local and global linearization methods under the same umbrella. The effectiveness of the presented method has been verified by simulations and a real-world robotic experiment.

关键词： deep learning in robotics and automation model learning for control Koopman operator model predictive control Lipschitz neural network

来源：评论

学校读者我要写书评

暂无评论

learning Modular Robot Control Policies

引用

IEEE TRANSACTIONS ON robotics 2023年第5期39卷 4095-4113页

作者： Whitman, Julian Travers, Matthew Choset, Howie Carnegie Mellon Univ Robot Inst Pittsburgh PA 15213 USA

Modular robots can be rearranged into a new design, perhaps each day, to handle a wide variety of tasks by forming a customized robot for each new task. However, reconfiguring just the mechanism is not sufficient: each design also requires its own unique control policy. One could craft a policy from scratch for each new design, but such an approach is not scalable, especially given the large number of designs that can be generated from even a small set of modules. Instead, we create a modular policy framework where the policy structure is conditioned on the hardware arrangement, and use just one training process to create a policy that controls a wide variety of designs. Our approach leverages the fact that the kinematics of a modular robot can be represented as a design graph, with nodes as modules and edges as connections between them. Given a robot, its design graph is used to create a policy graph with the same structure, where each node contains a deep neural network, and modules of the same type share knowledge via shared parameters (e.g., all legs on a hexapod share the same network parameters). We developed a model-based reinforcement learning algorithm, interleaving model learning and trajectory optimization to train the policy. We show the modular policy generalizes to a large number of designs that were not seen during training without any additional learning. Finally, we demonstrate the policy controlling a variety of designs to locomote with both simulated and real robots.

关键词： Cellular and modular robots deep learning in robotics and automation legged robots model learning for control

来源：评论

学校读者我要写书评

暂无评论

SOZIL: Self-Optimal Zero-Shot Imitation learning

引用

IEEE TRANSACTIONS ON COGNITIVE AND DEVELOPMENTAL SYSTEMS 2023年第4期15卷 2077-2088页

作者： Hao, Peng Lu, Tao Cui, Shaowei Wei, Junhang Cai, Yinghao Wang, Shuo Chinese Acad Sci Inst Automat State Key Lab Management & Control Complex Syst Beijing 100190 Peoples R China Univ Chinese Acad Sci Sch Artificial Intelligence Beijing 100049 Peoples R China Chinese Acad Sci Ctr Excellence Brain Sci & Intelligence Technol Shanghai 200031 Peoples R China

Zero-shot imitation learning has demonstrated its superiority to learn complex robotic tasks with less human participation. Recent studies show convincing performance under the condition that the robot follows the demonstration strictly by the learned inverse model. However, these methods are difficult to achieve satisfactory performance in imitation when the demonstration is suboptimal, and the learning of the learned inverse models is vulnerable to label ambiguity issues. In this article, we propose self-optimal zero-shot imitation learning (SOZIL) to tackle these problems. The contribution of SOZIL is twofold. First, goal consistency loss (GCL) is designed to learn the multistep goal-conditioned policy from exploration data. By directly using the goal state as supervision, GCL solves the label ambiguity problem caused by trajectory and action diversity. Second, estimation-based keyframe extraction (EKE) is developed to optimize demonstrations. We formulate the keyframe extraction process as a path optimization problem under suboptimal control. By predicting the performance of the learned policy in executing transitions of any two states, EKE creates a directed graph containing all candidate paths and extracts keyframes by solving the graph's shortest path problem. Furthermore, the proposed method is evaluated with various simulated and real-world robotic manipulating experiments, such as cable harness assembly, rope manipulation, and block moving. Experimental results show that SOZIL achieves a superior success rate and manipulation efficiency than baselines.

关键词： deep learning in robotics and automation imitation learning keyframe demonstration learning from observation

来源：评论

学校读者我要写书评

暂无评论

Enabling Visual Action Planning for Object Manipulation Through Latent Space Roadmap

引用

IEEE TRANSACTIONS ON robotics 2023年第1期39卷 57-75页

作者： Lippi, Martina Poklukar, Petra Welle, Michael C. Varava, Anastasia Yin, Hang Marino, Alessandro Kragic, Danica KTH Royal Inst Technol S-11428 Stockholm Sweden KTH Royal Inst Technol EECS RPL S-11428 Stockholm Sweden KTH Royal Inst Technol Div Robot Percept & Learning S-11428 Stockholm Sweden KTH Royal Inst Technol Sch Elect Engn & Comp Sci S-11428 Stockholm Sweden Roma Tre Univ I-00154 Rome Italy Univ Cassino & Southern Lazio I-03043 Cassino Italy

In this article, we present a framework for visual action planning of complex manipulation tasks with high-dimensional state spaces, focusing on manipulation of deformable objects. We propose a latent space roadmap (LSR) for task planning, which is a graph-based structure globally capturing the system dynamics in a low-dimensional latent space. Our framework consists of the following three parts. First, a mapping module (MM) that maps observations is given in the form of images into a structured latent space extracting the respective states as well as generates observations from the latent states. Second, the LSR, which builds and connects clusters containing similar states in order to find the latent plans between start and goal states, extracted by MM. Third, the action proposal module that complements the latent plan found by the LSR with the corresponding actions. We present a thorough investigation of our framework on simulated box stacking and rope/box manipulation tasks, and a folding task executed on a real robot.

关键词： Task analysis Planning Visualization Robots Trajectory Stacking Heuristic algorithms deep learning in robotics and automation Latent Space Planning Manipulation Planning Visual learning

来源：评论

学校读者我要写书评

暂无评论

Grasping Living Objects With Adversarial Behaviors Using Inverse Reinforcement learning

引用

IEEE TRANSACTIONS ON robotics 2023年第2期39卷 1151-1163页

作者： Hu, Zhe Zheng, Yu Pan, Jia City Univ Hong Kong Dept Biomed Engn Hong Kong Peoples R China Tencent Robot X Shenzhen 518057 Peoples R China Univ Hong Kong Dept Comp Sci Hong Kong Peoples R China

Living objects are difficult to grasp since they can actively elude capture by adopting adversarial behaviors that are extremely hard to model or predict. In this case, an inappropriately strong contact force may hurt the struggling living objects and a grasping algorithm that can minimize the contact force whenever possible is required. To solve this challenging task, in this article, we present a reinforcement-learning (RL)-based algorithm with two stages: the pregrasp stage and the in-hand stage. In the pregrasp stage, the robot focuses on the living object's adversarial behavior and approaches it in a reliable manner. In particular, we use inverse RL to encode the living object's adversarial behavior into a reward function. The negative value of the learned reward function is then used to train a high-quality grasping policy that can compete with the living object's adversarial behavior with the RL framework. In the in-hand stage, we use RL to train a grasp policy such that the dexterous hand can grab the living object with the minimal force. A set of dense rewards are also specifically designed to encourage the robot to grasp and hold the living object persistently. To further improve the grasp performance, we explicitly take into account the structure of the dexterous robot hand by treating the hand as a graph and adopting graph convolutional network to formulate the grasping policy. We conduct a set of experiments to demonstrate the performance of our proposed method, in which the robot can grasp living objects with the success rate of 90% and 95% in the pregrasp and in-hand stages, respectively. The contact force applied by the robotic hand to the living object is dramatically reduced in comparison with the baseline grasping policy.

关键词： deep learning in robotics and automation dexterous manipulation grasping reinforcement learning (RL)

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：