检索结果-内蒙古大学图书馆

Hierarchical human-motion prediction and logic-geometric programming for minimal interference human-robot tasks

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Le, An T. Kratzer, Philipp Hagenmayer, Simon Toussaint, Marc Mainprice, Jim Machine Learning and Robotics Lab University of Stuttgart Germany Max Planck Institute for Intelligent Systems IS-MPI Tübingen/Stuttgart Germany Technische Universität Berlin TUB Germany

In this paper, we tackle the problem of human-robot coordination in sequences of manipulation tasks. Our approach integrates hierarchical human motion prediction with Task and Motion Planning (TAMP). We first devise a hierarchical motion prediction approach by combining Inverse Reinforcement learning and short-term motion prediction using a Recurrent Neural Network. In a second step, we propose a dynamic version of the TAMP algorithm Logic-Geometric Programming (LGP) [1]. Our version of Dynamic LGP, replans periodically to handle the mismatch between the human motion prediction and the actual human behavior. We assess the efficacy of the approach by training the prediction algorithms and testing the framework on the publicly available MoGaze dataset [2]. © 2021, CC BY-NC-ND.

关键词： Computer circuits

Agent self-assessment: Determining policy quality without execution

学校读者我要写书评

暂无评论

Agent self-assessment: Determining policy quality without ex...

IEEE Symposium on Adaptive Dynamic Programming and Reinforcement learning

作者： Hans, Alexander Duell, Siegmund Udluft, Steffen Neuroinformatics and Cognitive Robotics Lab Ilmenau University of Technology Ilmenau Germany Machine Learning Group Berlin Institute of Technology Berlin Germany Intelligent Systems and Control Siemens AG Corporate Technology Munich Munich Germany

ISBN: (纸本)9781424498888

With the development of data-efficient reinforcement learning (RL) methods, a promising data-driven solution for optimal control of complex technical systems has become available. For the application of RL to a technical system, it is usually required to evaluate a policy before actually applying it to ensure it operates the system safely and within required performance bounds. In benchmark applications one can use the system dynamics directly to measure the policy quality. In real applications, however, this might be too expensive or even impossible. Being unable to evaluate the policy without using the actual system hinders the application of RL to autonomous controllers. As a first step toward agent self-assessment, we deal with discrete MDPs in this paper. We propose to use the value function along with its uncertainty to assess a policy's quality and show that, when dealing with an MDP estimated from observations, the value function itself can be misleading. We address this problem by determining the value function's uncertainty through uncertainty propagation and evaluate the approach using a number of benchmark applications. © 2011 IEEE.

关键词： Reinforcement learning

Probabilistic Recurrent State-Space Models

学校读者我要写书评

暂无评论

arXiv 2018年

作者： Doerr, Andreas Daniel, Christian Schiegg, Martin Nguyen-Tuong, Duy Schaal, Stefan Toussaint, Marc Trimpe, Sebastian Bosch Center for Artificial Intelligence Renningen Germany Max Planck Institute for Intelligent Systems Stuttgart/Tübingen Germany Computational Learning and Motor Control Lab University of Southern California United States Machine Learning and Robotics Lab University of Stuttgart Germany

State-space models (SSMs) are a highly expressive model class for learning patterns in time series data and for system identification. Deterministic versions of SSMs (e.g. LSTMs) proved extremely successful in modeling complex time series data. Fully probabilistic SSMs, however, are often found hard to train, even for smaller problems. To overcome this limitation, we propose a novel model formulation and a scalable training algorithm based on doubly stochastic variational inference and Gaussian processes. In contrast to existing work, the proposed variational approximation allows one to fully capture the latent state temporal correlations. These correlations are the key to robust training. The effectiveness of the proposed PR-SSM is evaluated on a set of realworld benchmark datasets in comparison to stateof-the-art probabilistic model learning methods. Scalability and robustness are demonstrated on a high dimensional problem. Copyright © 2018, The Authors. All rights reserved.

关键词： State space methods

Vectorized conditional neural fields: a framework for solving time-dependent parametric partial differential equations 24

学校读者我要写书评

暂无评论

Vectorized conditional neural fields: a framework for solvin...

Proceedings of the 41st International Conference on machine learning

作者： Jan Hagnberger Marimuthu Kalimuthu Daniel Musekamp Mathias Niepert Machine Learning and Simulation Lab Institute for Artificial Intelligence University of Stuttgart Stuttgart Germany Machine Learning and Simulation Lab Institute for Artificial Intelligence University of Stuttgart Stuttgart Germany and Stuttgart Center for Simulation Science (SimTech) and International Max Planck Research School for Intelligent Systems (IMPRS-IS) Machine Learning and Simulation Lab Institute for Artificial Intelligence University of Stuttgart Stuttgart Germany and International Max Planck Research School for Intelligent Systems (IMPRS-IS)

Transformer models are increasingly used for solving Partial Differential Equations (PDEs). Several adaptations have been proposed, all of which suffer from the typical problems of Transformers, such as quadratic memory and time complexity. Furthermore, all prevalent architectures for PDE solving lack at least one of several desirable properties of an ideal surrogate model, such as (i) generalization to PDE parameters not seen during training, (ii) spatial and temporal zero-shot super-resolution, (iii) continuous temporal extrapolation, (iv) support for 1D, 2D, and 3D PDEs, and (v) efficient inference for longer temporal rollouts. To address these limitations, we propose Vectorized Conditional Neural Fields (VCNeFs), which represent the solution of time-dependent PDEs as neural fields. Contrary to prior methods, however, VCNeFs compute, for a set of multiple spatio-temporal query points, their solutions in parallel and model their dependencies through attention mechanisms. Moreover, VCNeF can condition the neural field on both the initial conditions and the parameters of the PDEs. An extensive set of experiments demonstrates that VCNeFs are competitive with and often outperform existing ML-based surrogate models.

关键词：

A system for traded control teleoperation of manipulation tasks using intent prediction from hand gestures

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Oh, Yoojin Schäfer, Tim Rüther, Benedikt Toussaint, Marc Mainprice, Jim Machine Learning and Robotics Lab IPVS University of Stuttgart Germany Max Planck Institute for Intelligent Systems MPI-IS Tübingen/Stuttgart Germany Technische Universität Berlin TUB Germany

This paper presents a teleoperation system that includes robot perception and intent prediction from hand gestures. The perception module identifies the objects present in the robot workspace and the intent prediction module which object the user likely wants to grasp. This architecture allows the approach to rely on traded control instead of direct control: we use hand gestures to specify the goal objects for a sequential manipulation task, the robot then autonomously generates a grasping or a retrieving motion using trajectory optimization. The perception module relies on the model-based tracker to precisely track the 6D pose of the objects and makes use of a state of the art learning-based object detection and segmentation method, to initialize the tracker by automatically detecting objects in the scene. Goal objects are identified from user hand gestures using a trained a multi-layer perceptron classifier. After presenting all the components of the system and their empirical evaluation, we present experimental results comparing our pipeline to a direct traded control approach (i.e., one that does not use prediction) which shows that using intent prediction allows to bring down the overall task execution time. © 2021, CC BY-NC-ND.

关键词： Forecasting

GraspME - Grasp manifold estimator

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Hager, Janik Bauer, Ruben Toussaint, Marc Mainprice, Jim Machine Learning and Robotics Lab IPVS University of Stuttgart Germany Max Planck Institute for Intelligent Systems IS-MPI Tübingen/Stuttgart Germany Technische Universität Berlin TUB Germany

In this paper, we introduce a Grasp Manifold Estimator (GraspME) to detect grasp affordances for objects directly in 2D camera images. To perform manipulation tasks autonomously it is crucial for robots to have such graspability models of the surrounding objects. Grasp manifolds have the advantage of providing continuously infinitely many grasps, which is not the case when using other grasp representations such as predefined grasp points. For instance, this property can be leveraged in motion optimization to define goal sets as implicit surface constraints in the robot configuration space. In this work, we restrict ourselves to the case of estimating possible end-effector positions directly from 2D camera images. To this extend, we define grasp manifolds via a set of keypoints and locate them in images using a Mask R-CNN [1] backbone. Using learned features allows to generalize to different view angle, with potentially noisy images, and objects that were not part of the training set. We rely on simulation data only and perform experiments on simple and complex objects, including unseen ones. Our framework achieves an inference speed of 11.5 fps on a GPU, an average precision for keypoint estimation of 94.5% and a mean pixel distance of only 1.29. This shows that we can estimate the objects very well via bounding boxes and segmentation masks as well as approximate the correct grasp manifold's keypoint coordinates. © 2021, CC BY-NC-ND.

关键词： Cameras

Temperature-Aware Memory Mapping and Active Cooling of Neural Processing Units

学校读者我要写书评

暂无评论

Temperature-Aware Memory Mapping and Active Cooling of Neura...

International Symposium on Low Power Electronics and Design (ISLPED)

作者： Vahidreza Moghaddas Hammam Kattan Tim Bücher Mikail Yayla Jian-Jia Chen Hussam Amrouch TU Dortmund University University of Stuttgart Lamarr Institute for Machine Learning and Artificial Intelligences AI Processor Design Technical University of Munich Munich Institute of Robotics and Machine Intelligence

Neural processing units (NPUs) have become indispensable for meeting the high computational demands of deep neural networks (DNNs). They provide a very efficient solution, thanks to having a huge MAC array that enables massive parallelism. Nevertheless, such an architecture exhibits excessive on-chip power densities leading to a localized hot-spot that seriously heats its surroundings. This work demonstrates how the on-chip temperatures induced by the MAC array create a spatial thermal gradient through the on-chip SRAM memory. This makes the memory regions sensitive to different error probabilities (Perror), leading to significant accuracy drops when DNNs are being executed. To surmount this challenge, we employ on-chip superlattice thermoelectric (TEC) cooling devices that effectively reduce the memory temperature. Although scaling the memory voltage makes SRAM cells more sensitive to errors, it significantly decreases the leakage power, which compensates for the power consumed by the incorporated TEC devices. Furthermore, operating the SRAM at a lower voltage and temperature substantially increases its lifetime because voltage and temperature are key stimuli of transistor aging. By running multi-physics simulations using commercial finite-element tools and SPICE simulations for the 14nm FinFET technology, we accurately derive the relation between the Perror in different memory regions and the corresponding cooling cost. We then propose a three-stage temperature-aware layer-wise memory mapping that exploits different degrees of the sensitivity of NN layers to errors towards maximizing the DNN accuracy while minimizing the cooling cost. Experimental results reveal that our method notably improves the DNN accuracy compared to existing temperature-oblivious memory mapping.

关键词：

Safe Multi-Agent Reinforcement learning for Behavior-Based Cooperative Navigation

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Dawood, Murad Pan, Sicong Dengler, Nils Zhou, Siqi Schoellig, Angela P. Bennewitz, Maren The Humanoid Robots Lab University of Bonn Germany The Lamarr Institute for Machine Learning and Artificial Intelligence and the Center for Robotics Bonn Germany The Learning Systems and Robotics lab The Technical University of Munich Germany

In this paper, we address the problem of behavior-based cooperative navigation of mobile robots using safe multi-agent reinforcement learning (MARL). Our work is the first to focus on cooperative navigation without individual reference targets for the robots, using a single target for the formation's centroid. This eliminates the complexities involved in having several path planners to control a team of robots. To ensure safety, our MARL framework uses model predictive control (MPC) to prevent actions that could lead to collisions during training and execution. We demonstrate the effectiveness of our method in simulation and on real robots, achieving safe behavior-based cooperative navigation without using individual reference targets, with zero collisions, and faster target reaching compared to baselines. Finally, we study the impact of MPC safety filters on the learning process, revealing that we achieve faster convergence during training and we show that our approach can be safely deployed on real robots, even during early stages of the training. © 2023, CC0.

关键词： Adversarial machine learning

An efficient minimum vocabulary construction algorithm for language modeling 12

学校读者我要写书评

暂无评论

25th International Conference on Industrial Engineering and Other Applications of Applied Intelligent Systems, IEA/AIE 2012

作者： Lin, Sina Qin, Zengchang Huang, Zehua Wan, Tao Intelligent Computing and Machine Learning Lab. School of ASEE Beihang University Beijing China School of Advanced Engineering Beihang University China Robotics Institute Carnegie Mellon University Pittsburgh United States School of Medicine Boston University Boston United States

ISBN: (纸本)9783642310867

In learning a new word by a dictionary, we first need to know a set of "basic words" which are frequently appeared in word definitions. It often happens that you cannot understand the word you looked up because there are still some words you do not understand in its definitions or explanations provided by the dictionary. You can keep looking up these new words recursively till they all can be well explained by some basic words you already knew. How to automatically find a minimum set of such basic words to define (or recursively define) the entire vocabulary in a given dictionary is what are going to discuss in this paper. We propose an efficient algorithm to construct the Minimum Vocabulary (MV) using the word frequency information. The minimum vocabulary can be used for language modeling and experimental results demonstrate the effectiveness of using the minimum vocabulary as features in text classification. © 2012 Springer-Verlag.

关键词： Classification (of information)