检索结果-内蒙古大学图书馆

arXiv 2021年

作者： Freymuth, Niklas Becker, Philipp Neumann, Gerhard Autonomous Learning Robots Karlsruhe Institute of Technology Karlsruhe Germany

Inverse Reinforcement learning infers a reward function from expert demonstrations, aiming to encode the behavior and intentions of the expert. Current approaches usually do this with generative and uni-modal models, meaning that they encode a single behavior. In the common setting, where there are various solutions to a problem and the experts show versatile behavior this severely limits the generalization capabilities of these methods. We propose a novel method for Inverse Reinforcement learning that overcomes these problems by formulating the recovered reward as a sum of iteratively trained discriminators. We show on simulated tasks that our approach is able to recover general, high-quality reward functions and produces policies of the same quality as behavioral cloning approaches designed for versatile behavior. © 2021, CC BY.

关键词： Encoding (symbols)

来源：评论

学校读者我要写书评

暂无评论

End-to-End learning of Hybrid Inverse Dynamics Models for Precise and Compliant Impedance Control

arXiv

引用

arXiv 2022年

作者： Reuss, Moritz van Duijkeren, Niels Krug, Robert Becker, Philipp Shaj, Vaisakh Neumann, Gerhard Bosch Corporate Research Renningen Germany LCAS University Of Lincoln United Kingdom Intuitive Robots Lab Karlsruhe Institute of Technology Germany Autonomous Learning Robots Karlsruhe Institute of Technology Germany

It is well-known that inverse dynamics models can improve tracking performance in robot control. These models need to precisely capture the robot dynamics, which consist of well-understood components, e.g., rigid body dynamics, and effects that remain challenging to capture, e.g., stick-slip friction and mechanical flexibilities. Such effects exhibit hysteresis and partial observability, rendering them, particularly challenging to model. Hence, hybrid models, which combine a physical prior with data-driven approaches are especially well-suited in this setting. We present a novel hybrid model formulation that enables us to identify fully physically consistent inertial parameters of a rigid body dynamics model which is paired with a recurrent neural network architecture, allowing us to capture unmodeled partially observable effects using the network memory. We compare our approach against state-of-the-art inverse dynamics models on a 7 degree of freedom manipulator. Using data sets obtained through an optimal experiment design approach, we study the accuracy of offline torque prediction and generalization capabilities of joint learning methods. In control experiments on the real system, we evaluate the model as a feed-forward term for impedance control and show the feedback gains can be drastically reduced to achieve a given tracking accuracy. © 2022, CC BY-NC-ND.

关键词： Inverse problems

来源：评论

学校读者我要写书评

暂无评论

EXPECTED INFORMATION MAXIMIZATION USING THE I-PROJECTION FOR MIXTURE DENSITY ESTIMATION 8

EXPECTED INFORMATION MAXIMIZATION USING THE I-PROJECTION FOR...

引用

8th International Conference on learning Representations, ICLR 2020

作者： Becker, Philipp Arenz, Oleg Neumann, Gerhard Autonomous Learning Robots KIT Bosch Center for Artificial Intelligence Intelligent Autonomous Systems TU Darmstadt Germany University of Tübingen Germany

Modelling highly multi-modal data is a challenging problem in machine learning. Most algorithms are based on maximizing the likelihood, which corresponds to the M(oment)-projection of the data distribution to the model distribution. The M-projection forces the model to average over modes it cannot represent. In contrast, the I(nformation)-projection ignores such modes in the data and concentrates on the modes the model can represent. Such behavior is appealing whenever we deal with highly multi-modal data where modelling single modes correctly is more important than covering all the modes. Despite this advantage, the I-projection is rarely used in practice due to the lack of algorithms that can efficiently optimize it based on data. In this work, we present a new algorithm called Expected Information Maximization (EIM) for computing the I-projection solely based on samples for general latent variable models, where we focus on Gaussian mixtures models and Gaussian mixtures of experts. Our approach applies a variational upper bound to the I-projection objective which decomposes the original objective into single objectives for each mixture component as well as for the coefficients, allowing an efficient optimization. Similar to GANs, our approach employs discriminators but uses a more stable optimization procedure, using a tight upper bound. We show that our algorithm is much more effective in computing the I-projection than recent GAN approaches and we illustrate the effectiveness of our approach for modelling multi-modal behavior on two pedestrian and traffic prediction datasets. © 2020 8th International Conference on learning Representations, ICLR 2020. All rights reserved.

关键词： Modal analysis

来源：评论

学校读者我要写书评

暂无评论

Deep Adversarial Reinforcement learning for Object Disentangling

Deep Adversarial Reinforcement Learning for Object Disentang...

引用

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

作者： Laux, Melvin Arenz, Oleg Peters, Jan Pajarinen, Joni Tech Univ Darmstadt Intelligent Autonomous Syst Darmstadt Germany MPI Intelligent Syst Tubingen Germany Tampere Univ Learning Intelligent Autonomous Robots Tampere Finland

ISBN: (纸本)9781728162126

Deep learning in combination with improved training techniques and high computational power has led to recent advances in the field of reinforcement learning (RL) and to successful robotic RL applications such as in-hand manipulation. However, most robotic RL relies on a well known initial state distribution. In real-world tasks, this information is however often not available. For example, when disentangling waste objects the actual position of the robot w.r.t. the objects may not match the positions the RL policy was trained for. To solve this problem, we present a novel adversarial reinforcement learning (ARL) framework. The ARL framework utilizes an adversary, which is trained to steer the original agent, the protagonist, to challenging states. We train the protagonist and the adversary jointly to allow them to adapt to the changing policy of their opponent. We show that our method can generalize from training to test scenarios by training an end-to-end system for robot control to solve a challenging object disentangling task. Experiments with a KUKA LBR+ 7-DOF robot arm show that our approach outperforms the baseline method in disentangling when starting from different initial states than provided during training.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

HIDDEN PARAMETER RECURRENT STATE SPACE MODELS FOR CHANGING DYNAMICS SCENARIOS

arXiv

引用

arXiv 2022年

作者： Shaj, Vaisakh Büchler, Dieter Sonker, Rohit Becker, Philipp Neumann, Gerhard Autonomous Learning Robots KIT Germany LCAS University Of Lincoln United Kingdom Max Planck Institute for Intelligent Systems Tübingen Germany Indian Institute Of Technology Kanpur India

Recurrent State-space models (RSSMs) are highly expressive models for learning patterns in time series data and system identification. However, these models assume that the dynamics are fixed and unchanging, which is rarely the case in real-world scenarios. Many control applications often exhibit tasks with similar but not identical dynamics which can be modeled as a latent variable. We introduce the Hidden Parameter Recurrent State Space Models (HiP-RSSMs), a framework that parametrizes a family of related dynamical systems with a low-dimensional set of latent factors. We present a simple and effective way of learning and performing inference over this Gaussian graphical model that avoids approximations like variational inference. We show that HiP-RSSMs outperforms RSSMs and competing multi-task models on several challenging robotic benchmarks both on real-world systems and simulations. © 2022, CC BY-SA.

关键词： Dynamics

来源：评论

学校读者我要写书评

暂无评论

A study on dense and sparse (Visual) rewards in robot policy learning

arXiv

引用

arXiv 2021年

作者： Mohtasib, Abdalkarim Neumann, Gerhard Cuayáhuitl, Heriberto Lincoln Centre for Autonomous Systems University of Lincoln Lincoln United Kingdom Autonomous Learning Robots Karlsruhe Institute of Technology Karlsruhe Germany

Deep Reinforcement learning (DRL) is a promising approach for teaching robots new behaviour. However, one of its main limitations is the need for carefully hand-coded reward signals by an expert. We argue that it is crucial to automate the reward learning process so that new skills can be taught to robots by their users. To address such automation, we consider task success classifiers using visual observations to estimate the rewards in terms of task success. In this work, we study the performance of multiple state-of-the-art deep reinforcement learning algorithms under different types of reward: Dense, Sparse, Visual Dense, and Visual Sparse rewards. Our experiments in various simulation tasks (Pendulum, Reacher, Pusher, and Fetch Reach) show that while DRL agents can learn successful behaviours using visual rewards when the goal targets are distinguishable, their performance may decrease if the task goal is not clearly visible. Our results also show that visual dense rewards are more successful than visual sparse rewards and that there is no single best algorithm for all tasks. © 2021, CC BY-NC-SA.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

learning Riemannian Manifolds for Geodesic Motion Skills 17th

Learning Riemannian Manifolds for Geodesic Motion Skills

引用

Conference on Robotics - Science and Systems

作者： Beik-Mohammadi, Hadi Hauberg, Soren Arvanitidis, Georgios Neumann, Gerhard Rozo, Leonel Bosch Ctr Artificial Intelligence BCAI Renningen Germany Karlsruhe Inst Technol KIT Autonomous Learning Robots Lab Karlsruhe Germany Tech Univ Denmark DTU Sect Cognit Syst Lyngby Denmark Max Planck Inst Intelligent Syst MPI IS Tubingen Germany

ISBN: (纸本)9780992374778

*For robots to work alongside humans and perform in unstructured environments, they must learn new motion skills and adapt them to unseen situations on the fly. This demands learning models that capture relevant motion patterns, while offering enough flexibility to adapt the encoded skills to new requirements, such as dynamic obstacle avoidance. We introduce a Riemannian manifold perspective on this problem, and propose to learn a Riemannian manifold from human demonstrations on which geodesics are natural motion skills. We realize this with a variational autoencoder (VAE) over the space of position and orientations of the robot end-effector. Geodesic motion skills let a robot plan movements from and to arbitrary points on the data manifold. They also provide a straightforward method to avoid obstacles by redefining the ambient metric in an online fashion. Moreover, geodesics naturally exploit the manifold resulting from multiple-mode tasks to design motions that were not explicitly demonstrated previously. We test our learning framework using a 7-DoF robotic manipulator, where the robot satisfactorily learns and reproduces realistic skills featuring elaborated motion patterns, avoids previously-unseen obstacles, and generates novel movements in multiple-mode settings.

关键词： Manipulators

来源：评论

学校读者我要写书评

暂无评论

Machine learning With Computer Networks: Techniques, Datasets, and Models

引用

IEEE ACCESS 2024年 12卷 54673-54720页

作者： Afifi, Haitham Pochaba, Sabrina Boltres, Andreas Laniewski, Dominic Haberer, Janek Paeleke, Leonard Poorzare, Reza Stolpmann, Daniel Wehner, Nikolas Redder, Adrian Samikwa, Eric Seufert, Michael Accenture D-61476 Kronberg Germany Salzburg Res Forsch Gesell mbH A-5020 Salzburg Austria Karlsruhe Inst Technol KIT Autonomous Learning Robots Lab D-76131 Karlsruhe Germany Osnabruck Univ Inst Comp Sci D-49076 Osnabruck Germany Univ Kiel Distributed Syst Grp D-24118 Kiel Germany Univ Potsdam Digital Engn Fac D-14482 Potsdam Germany Hasso Plattner Inst Digital Hlth & Machine Learning D-14482 Potsdam Germany Hsch Karlsruhe Tech Inst Appl Res Wirtschaft Ctr Appl Res Data Centr Software Syst DSS Res Grp D-76133 Karlsruhe Germany Hamburg Univ Technol Inst Commun Networks D-21073 Hamburg Germany Univ Wurzburg Chair Commun Networks D-97074 Wurzburg Germany Univ Paderborn D-33098 Paderborn Germany Univ Bern Inst Comp Sci CH-3012 Bern Switzerland Univ Augsburg Chair Networked Embedded Syst & Commun Syst D-86159 Augsburg Germany

Machine learning has found many applications in network contexts. These include solving optimisation problems and managing network operations. Conversely, networks are essential for facilitating machine learning training and inference, whether performed centrally or in a distributed fashion. To conduct rigorous research in this area, researchers must have a comprehensive understanding of fundamental techniques, specific frameworks, and access to relevant datasets. Additionally, access to training data can serve as a benchmark or a springboard for further investigation. All these techniques are summarized in this article;serving as a primer paper and hopefully providing an efficient start for anybody doing research regarding machine learning for networks or using networks for machine learning.

关键词： Computer networking datasets machine learning metrics tools

来源：评论

学校读者我要写书评

暂无评论

autonomous Search for Underground Mine Rescue Using Aerial robots

Autonomous Search for Underground Mine Rescue Using Aerial R...

引用

IEEE Aerospace Conference

作者： Tung Dang Mascarich, Frank Khattak, Shehryar Huan Nguyen Hai Nguyen Hirsh, Satchel Reinhart, Russell Papachristos, Christos Alexis, Kostas Univ Nevada Autonomous Robots Lab Reno NV 89557 USA Northeastern Univ Lab Learning & Planning Robot Helping Hands Lab Boston MA 02115 USA

ISBN: (数字)9781728127347

ISBN: (纸本)9781728127347

In this paper we present a comprehensive solution for autonomous underground mine rescue using aerial robots. In particular, a new class of Micro Aerial Vehicles are equipped with the ability to localize and map in subterranean settings, explore unknown mine environments on their own, and perform detection and localization of objects of interest for the purposes of mine rescue (i.e., "human survivors" and associated objects such as "backpacks", "smartphones" or "tools"). For the purposes of GPS-denied localization and mapping in the visuallydegraded underground environments (e.g., a smoke-filled mine during an accident) the solution relies on the fusion of LiDAR data with thermal vision frames and inertial cues. autonomous exploration is enabled through a graph-based search algorithm and an online volumetric representation of the environment. Object search is then enabled through a deep learning-based classifier, while the associated location is queried using the online reconstructed map. The complete software framework runs onboard the aerial robots utilizing the integrated embedded processing resources. The overall system is extensively evaluated in real-life deployments in underground mines.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Human-machine symbiosis: A multivariate perspective for physically coupled human-machine systems

引用

INTERNATIONAL JOURNAL OF HUMAN-COMPUTER STUDIES 2023年 170卷

作者： Inga, Jairo Ruess, Miriam Robens, Jan Heinrich Nelius, Thomas Rothfuss, Simon Kille, Sean Dahlinger, Philipp Lindenmann, Andreas Thomaschke, Roland Neumann, Gerhard Matthiesen, Sven Hohmann, Soren Kiesel, Andrea Karlsruhe Inst Technol KIT Inst Control Syst IRS Kaiserstr 12 D-76131 Karlsruhe Germany Univ Freiburg Dept Psychol Cognit Act & Sustainabil Unit Engelbergerstr 41 D-79085 Freiburg Germany Karlsruhe Inst Technol KIT Inst Prod Engn IPEK Kaiserstr 12 D-76131 Karlsruhe Germany Karlsruhe Inst Technol KIT Inst Anthropomat & Robot IAR Autonomous Learning Robots Kaiserstr 12 D-76131 Karlsruhe Germany

The notion of symbiosis has been increasingly mentioned in research on physically coupled human-machine systems. Yet, a uniform specification on which aspects constitute human-machine symbiosis is missing. By combining the expertise of different disciplines, we elaborate on a multivariate perspective of symbiosis as the highest form of interaction in physically coupled human-machine systems, characterized by a oneness of the human and the machine. Four dimensions are considered: Task, interaction, performance, and experience. First, human and machine accomplish a common objective by completing tasks conceptualized on a decomposition, a decision and an action level (task dimension). Second, each partner possesses an internal representation of the oneness they form, including the partner's inner states (e.g. experiences) and their joint influence on the envi-ronment. This representation constitutes the "symbiotic understanding" between both partners, being the basis of a joint and highly coordinated action (interaction dimension). Third, the symbiotic interaction leads to synergetic effects regarding the complementary strengths of the partners, resulting in a higher overall performance (per-formance dimension). Fourth, symbiotic systems specifically change the user's experiences, like flow, accep-tance, sense of agency, and embodiment (experience dimension). Our multivariate perspective allows a clear description of symbiotic human-machine systems and helps to bridge barriers between different disciplines.

关键词： Human-machine symbiosis Human-machine cooperation Human-machine interaction Shared control Sense of agency

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：