检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Teoh, Eugene Patidar, Sumit Ma, Xiao James, Stephen Dyson Robot Learning Lab United Kingdom

Generalising vision-based manipulation policies to novel environments remains a challenging area with limited exploration. Current practices involve collecting data in one location, training imitation learning or reinforcement learning policies with this data, and deploying the policy in the same location. However, this approach lacks scalability as it necessitates data collection in multiple locations for each task. This paper proposes a novel approach where data is collected in a location predominantly featuring green screens. We introduce Greenscreen Augmentation (GreenAug), employing a chroma key algorithm to overlay background textures onto a green screen. Through extensive real-world empirical studies with over 850 training demonstrations and 8.2k evaluation episodes, we demonstrate that GreenAug surpasses no augmentation, standard computer vision augmentation, and prior generative augmentation methods in performance. While no algorithmic novelties are claimed, our paper advocates for a fundamental shift in data collection practices. We propose that real-world demonstrations in future research should utilise green screens, followed by the application of GreenAug. We believe GreenAug unlocks policy generalisation to visually distinct novel locations, addressing the current scene generalisation limitations in robot learning. © 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

ST-RRT*: Asymptotically-Optimal Bidirectional Motion Planning through Space-Time 39

ST-RRT*: Asymptotically-Optimal Bidirectional Motion Plannin...

引用

IEEE International Conference on robotics and Automation (ICRA)

作者： Grothe, Francesco Hartmann, Valentin N. Orthey, Andreas Toussaint, Marc TU Berlin Learning & Intelligent Syst Grp Berlin Germany Univ Stuttgart Machine Learning & Robot Lab Stuttgart Germany

ISBN: (纸本)9781728196817

We present a motion planner for planning through space-time with dynamic obstacles, velocity constraints, and unknown arrival time. Our algorithm, Space-Time RRT* (ST-RRT*), is a probabilistically complete, bidirectional motion planning algorithm, which is asymptotically optimal with respect to the shortest arrival time. We experimentally evaluate ST-RRT* in both abstract (2D disk, 8D disk in cluttered spaces, and on a narrow passage problem), and simulated robotic path planning problems (sequential planning of 8DoF mobile robots, and 7DoF robotic arms). The proposed planner outperforms RRT-Connect and RRT* on both initial solution time, and attained final solution cost. The code for ST-RRT* is available in the Open Motion Planning Library (OMPL).

关键词： Torque robot kinematics Heuristic algorithms Dynamics Probabilistic logic Manipulators Libraries

来源：评论

学校读者我要写书评

暂无评论

Real-time Mapping of Physical Scene Properties with an Autonomous robot Experimenter 6

Real-time Mapping of Physical Scene Properties with an Auton...

引用

6th Conference on robot learning (CoRL)

作者： Haughton, Iain Sucar, Edgar Mouton, Andre Johns, Edward Davison, Andrew J. Dyson Technol Ltd Malmesbury England Imperial Coll Dyson Robot Lab London England Imperial Coll Robot Learning Lab London England

Neural fields can be trained from scratch to represent the shape and appearance of 3D scenes efficiently. It has also been shown that they can densely map correlated properties such as semantics, via sparse interactions from a human labeller. In this work, we show that a robot can densely annotate a scene with arbitrary discrete or continuous physical properties via its own fully-autonomous experimental interactions, as it simultaneously scans and maps it with an RGB-D camera. A variety of scene interactions are possible, including poking with force sensing to determine rigidity, measuring local material type with single-pixel spectroscopy or predicting force distributions by pushing. Sparse experimental interactions are guided by entropy to enable high efficiency, with tabletop scene properties densely mapped from scratch in a few minutes from a few tens of interactions.

关键词： neural field robot experimentation physical properties

来源：评论

学校读者我要写书评

暂无评论

GPyro: uncertainty-aware temperature predictions for additive manufacturing

引用

JOURNAL OF INTELLIGENT MANUFACTURING 2023年第1期34卷 243-259页

作者： Sideris, Iason Crivelli, Francesco Bambach, Markus Swiss Fed Inst Technol Adv Mfg Lab Zurich Ramistr 101 CH-8092 Zurich Switzerland CSEM SA Grp Robot & Machine Learning Grundlistr 1 CH-6055 Alpnach Switzerland

In additive manufacturing, process-induced temperature profiles are directly linked to part properties, and their prediction is crucial for achieving high-quality products. Temperature predictions require an accurate process model, which is usually either a physics-based or a data-driven simulator. Although many high-performance models have been developed, they all suffer from disadvantages such as long execution times, the need for large datasets, and error accumulation in long prediction horizons. These caveats undermine the utility of such modeling approaches and pose problems in their integration within iterative optimization and closed-loop control schemes. In this work, we introduce GPyro, a generative model family specifically designed to address these issues and enable fast probabilistic temperature predictions. GPyro combines physics-informed and parametric regressors with a set of smooth attention mechanisms and learns the evolution of the dynamics inherent to a system by employing Gaussian processes. The model predictions are equipped with confidence intervals quantifying the uncertainty at each timestep. We applied GPyro to Wire-arc additive manufacturing and learned an accurate model from a single experiment on a real welding cell, almost in real-time. Our model can be easily integrated within existing loop-shaping and optimization frameworks.

关键词： Additive manufacturing Machine learning Thermal model Data-driven modelling WAAM Uncertainty quantification

来源：评论

学校读者我要写书评

暂无评论

Generative Image as Action Models

arXiv

引用

arXiv 2024年

作者： Shridhar, Mohit Lo, Yat Long James, Stephen Dyson Robot Learning Lab United Kingdom

Image-generation diffusion models have been fine-tuned to unlock new capabilities such as image-editing and novel view synthesis. Can we similarly unlock image-generation models for visuomotor control? We present GENIMA, a behavior-cloning agent that fine-tunes Stable Diffusion to "draw joint-actions" as targets on RGB images. These images are fed into a controller that maps the visual targets into a sequence of joint-positions. We study GENIMA on 25 RLBench and 9 real-world manipulation tasks. We find that, by lifting actions into image-space, internet pre-trained diffusion models can generate policies that outperform state-of-the-art visuomotor approaches, especially in robustness to scene perturbations and generalizing to novel objects. Our method is also competitive with 3D agents, despite lacking priors such as depth, keypoints, or motion-planners. © 2024, CC BY.

关键词： Clone cells

来源：评论

学校读者我要写书评

暂无评论

Adapting Skills to Novel Grasps: A Self-Supervised Approach

Adapting Skills to Novel Grasps: A Self-Supervised Approach

引用

IEEE/RSJ International Conference on Intelligent robots and Systems (IROS)

作者： Georgios Papagiannis Kamil Dreczkowski Vitalis Vosylius Edward Johns The Robot Learning Lab at Imperial College London

ISBN: (数字)9798350377705

ISBN: (纸本)9798350377712

In this paper, we study the problem of adapting manipulation trajectories involving grasped objects (e.g. tools) defined for a single grasp pose to novel grasp poses. A common approach to address this is to define a new trajectory for each possible grasp explicitly, but this is highly inefficient. Instead, we propose a method to adapt such trajectories directly while only requiring a period of self-supervised data collection, during which a camera observes the robot’s end-effector moving with the object rigidly grasped. Importantly, our method requires no prior knowledge of the grasped object (such as a 3D CAD model), it can work with RGB images, depth images, or both, and it requires no camera calibration. Through a series of real-world experiments involving 1360 evaluations, we find that self-supervised RGB data consistently outperforms alternatives that rely on depth images including several state-of-the-art pose estimation methods. Compared to the best-performing baseline, our method results in an average of 28.5% higher success rate when adapting manipulation trajectories to novel grasps on several everyday tasks. The appendix accompanying the paper and videos of the experiments are available on our webpage at ***/adapting-skills.

关键词： Solid modeling Three-dimensional displays Imitation learning robot vision systems Pose estimation Cameras End effectors Trajectory Intelligent robots Videos

来源：评论

学校读者我要写书评

暂无评论

DINOBot: robot Manipulation via Retrieval and Alignment with Vision Foundation Models

DINOBot: Robot Manipulation via Retrieval and Alignment with...

引用

IEEE International Conference on robotics and Automation (ICRA)

作者： Norman Di Palo Edward Johns The Robot Learning Lab at Imperial College London

ISBN: (数字)9798350384574

ISBN: (纸本)9798350384581

We propose DINOBot, a novel imitation learning framework for robot manipulation, which leverages the image-level and pixel-level capabilities of features extracted from Vision Transformers trained with DINO. When interacting with a novel object, DINOBot first uses these features to retrieve the most visually similar object experienced during human demonstrations, and then uses this object to align its endeffector with the novel object to enable effective interaction. Through a series of real-world experiments on everyday tasks, we show that exploiting both the image-level and pixel-level properties of vision foundation models enables unprecedented learning efficiency and generalisation. Videos and code are available at https://***/dinobot.

关键词： Visualization Computer vision Codes Imitation learning Feature extraction Transformers Task analysis

来源：评论

学校读者我要写书评

暂无评论

A learning-based Controller for Multi-Contact Grasps on Unknown Objects with a Dexterous Hand

A Learning-based Controller for Multi-Contact Grasps on Unkn...

引用

2024 International Conference on Intelligent robots and Systems

作者： Winkelbauer, Dominik Triebel, Rudolph Baeuml, Berthold DLR Inst Robot & Mechatron Wessling Germany Tech Univ Munich Learning AI Dextrous Robots Lab Munich Germany Karlsruhe Inst Technol KIT Karlsruhe Germany

ISBN: (纸本)9798350377712;9798350377705

Existing grasp controllers usually either only support finger-tip grasps or need explicit configuration of the inner forces. We propose a novel grasp controller that supports arbitrary grasp types, including power grasps with multi-contacts, while operating self-contained on before unseen objects. No detailed contact information is needed, but only a rough 3D model, e.g., reconstructed from a single depth image. First, the external wrench being applied to the object is estimated by using the measured torques at the joints. Then, the torques necessary to counteract the estimated wrench while keeping the object at its initial pose are predicted. The torques are commanded via desired joint angles to an underlying joint-level impedance controller. To reach real-time performance, we propose a learning-based approach that is based on a wrench estimator- and a torque predictor neural network. Both networks are trained in a supervised fashion using data generated via the analytical formulation of the controller. In an extensive simulation-based evaluation, we show that our controller is able to keep 83.1% of the tested grasps stable when applying external wrenches with up to 10 N. At the same time, we outperform the two tested baselines by being more efficient and inducing less involuntary object movement. Finally, we show that the controller also works on the real DLR-Hand II, reaching a cycle time of 6ms. Website: ***/grasping

关键词： 3D reconstruction

来源：评论

学校读者我要写书评

暂无评论

Continuous Control with Coarse-to-fine Reinforcement learning

arXiv

引用

arXiv 2024年

作者： Seo, Younggyo Uruç, Jafar James, Stephen Dyson Robot Learning Lab United Kingdom

Despite recent advances in improving the sample-efficiency of reinforcement learning (RL) algorithms, designing an RL algorithm that can be practically deployed in real-world environments remains a challenge. In this paper, we present Coarse-to-fine Reinforcement learning (CRL), a framework that trains RL agents to zoom-into a continuous action space in a coarse-to-fine manner, enabling the use of stable, sample-efficient value-based RL algorithms for fine-grained continuous control tasks. Our key idea is to train agents that output actions by iterating the procedure of (i) discretizing the continuous action space into multiple intervals and (ii) selecting the interval with the highest Q-value to further discretize at the next level. We then introduce a concrete, value-based algorithm within the CRL framework called Coarse-to-fine Q-Network (CQN). Our experiments demonstrate that CQN significantly outperforms RL and behavior cloning baselines on 20 sparsely-rewarded RLBench manipulation tasks with a modest number of environment interactions and expert demonstrations. We also show that CQN robustly learns to solve real-world manipulation tasks within a few minutes of online training. Project website: ***/cqn. Copyright © 2024, The Authors. All rights reserved.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Effective Streaming Low-Tubal-Rank Tensor Approximation via Frequent Directions

引用

IEEE TRANSACTIONS ON NEURAL NETWORKS AND learning SYSTEMS 2024年第1期35卷 1113-1126页

作者： Yi, Qianxin Wang, Chenhao Wang, Kaidong Wang, Yao Xi An Jiao Tong Univ Ctr Intelligent Decis Making & Machine Learning Sch Management Xian 710049 Peoples R China Chinese Acad Sci State Key Lab Robot Shenyang Inst Automat Shenyang 110016 Peoples R China

Low-tubal-rank tensor approximation has been proposed to analyze large-scale and multidimensional data. However, finding such an accurate approximation is challenging in the streaming setting, due to the limited computational resources. To alleviate this issue, this article extends a popular matrix sketching technique, namely, frequent directions (FDs), for constructing an efficient and accurate low-tubal-rank tensor approximation from streaming data based on the tensor singular value decomposition (t-SVD). Specifically, the new algorithm allows the tensor data to be observed slice by slice but only needs to maintain and incrementally update a much smaller sketch, which could capture the principal information of the original tensor. The rigorous theoretical analysis shows that the approximation error of the new algorithm can be arbitrarily small when the sketch size grows linearly. Extensive experimental results on both synthetic and real multidimensional data further reveal the superiority of the proposed algorithm compared with other sketching algorithms for getting low-tubal-rank approximation, in terms of both efficiency and accuracy.

关键词： Tensors Approximation algorithms Matrix decomposition Streaming media Hyperspectral imaging Memory management learning systems Frequent directions (FDs) streaming low-tubal-rank approximation tensor singular value decomposition (t-SVD) tensor sketching

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：