检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Brunke, Lukas Zhou, Siqi Schoellig, Angela P. Learning Systems and Robotics Lab The Munich Institute of Robotics and Machine Intelligence Technical University of Munich Munich80333 Germany University of Toronto Institute for Aerospace Studies North YorkONM3H 5T6 Canada University of Toronto Robotics Institute TorontoONM5S 1A4 Canada Vector Institute for Artificial Intelligence TorontoONM5G 0C6 Canada

Control barrier function (CBF)-based safety filters are used to certify and modify potentially unsafe control inputs to a system such as those provided by a reinforcement learning agent or a non-expert user. In this context, safety is defined as the satisfaction of state constraints. Originally designed for continuous-time systems, CBF safety filters typically assume that the system’s relative degree is well-defined and is constant across the domain;however, this assumption is restrictive and rarely verified—even linear system dynamics with a quadratic CBF candidate may not satisfy this assumption. In real-world applications, continuous-time CBF safety filters are implemented in discrete time, exacerbating issues related to violating the condition on the relative degree. These violations can lead to the safety filter being unconstrained (any control input may be certified) for a finite time interval and result in chattering issues and constraint violations. We propose an alternative formulation to address these challenges. Specifically, we present a theoretically sound method that employs multiple CBFs to generate bounded control inputs at each state within the safe set, thereby preventing incorrect certification of arbitrary control inputs. Using this approach, we derive conditions on the maximum sampling time to ensure safety in discrete-time implementations. We demonstrate the effectiveness of our proposed method through simulations and real-world quadrotor experiments, successfully preventing chattering and constraint violations. Finally, we discuss the implications of violating the relative degree condition on CBF synthesis and learning-based CBF methods. Copyright © 2024, The Authors. All rights reserved.

关键词： Continuous time systems

来源：评论

学校读者我要写书评

暂无评论

Semantically Safe Robot Manipulation: From Semantic Scene Understanding to Motion Safeguards

arXiv

引用

arXiv 2024年

作者： Brunke, Lukas Zhang, Yanni Römer, Ralf Naimer, Jack Staykov, Nikola Zhou, Siqi Schoellig, Angela P. Learning Systems and Robotics Lab The Munich Institute of Robotics and Machine Intelligence Technical University of Munich Munich80333 Germany University of Toronto Institute for Aerospace Studies North YorkONM3H 5T6 Canada University of Toronto Robotics Institute TorontoONM5S 1A4 Canada Vector Institute for Artificial Intelligence TorontoONM5G 0C6 Canada

Ensuring safe interactions in human-centric environments requires robots to understand and adhere to constraints recognized by humans as "common sense" (e.g., "moving a cup of water above a laptop is unsafe as the water may spill" or "rotating a cup of water is unsafe as it can lead to pouring its content"). Recent advances in computer vision and machine learning have enabled robots to acquire a semantic understanding of and reason about their operating environments. While extensive literature on safe robot decision-making exists, semantic understanding is rarely integrated into these formulations. In this work, we propose a semantic safety filter framework to certify robot inputs with respect to semantically defined constraints (e.g., unsafe spatial relationships, behaviors, and poses) and geometrically defined constraints (e.g., environment-collision and self-collision constraints). In our proposed approach, given perception inputs, we build a semantic map of the 3D environment and leverage the contextual reasoning capabilities of large language models to infer semantically unsafe conditions. These semantically unsafe conditions are then mapped to safe actions through a control barrier certification formulation. We demonstrate the proposed semantic safety filter in teleoperated manipulation tasks and with learned diffusion policies applied in a real-world kitchen environment that further showcases its effectiveness in addressing practical semantic safety constraints. Together, these experiments highlight our approach’s capability to integrate semantics into safety certification, enabling safe robot operation beyond traditional collision avoidance. Copyright © 2024, The Authors. All rights reserved.

关键词： Robot learning

来源：评论

学校读者我要写书评

暂无评论

Multi-Alpha Soft Actor-Critic: Overcoming Stochastic Biases in Maximum Entropy Reinforcement learning

Multi-Alpha Soft Actor-Critic: Overcoming Stochastic Biases ...

引用

IEEE International Conference on robotics and Automation (ICRA)

作者： Conor Igoe Swapnil Pande Siddarth Venkatraman Jeff Schneider Machine Learning Department School of Computer Science Carnegie Mellon University Pittsburgh PA United States Robotics Institute School of Computer Science Carnegie Mellon University Pittsburgh PA United States

The successful application of robotic control requires intelligent decision-making to handle the long tail of complex scenarios that arise in real-world environments. Recently, Deep Reinforcement learning (DRL) has provided a data-driven framework to automatically learn effective policies in such complex settings. Since its introduction in 2018, Soft Actor-Critic (SAC) remains as one of the most popular off-policy DRL algorithms and has been used extensively to learn performant robotic control policies. However, in this paper we argue that by relying on the maximum entropy formalism to define learning objectives, previous work introduces a significant bias away from optimal decision making, which often requires near-deterministic behaviour for high-precision tasks. Moreover, we show that when training with the original variants of SAC, overcoming this bias by reducing entropy budgets or entropy coefficients introduces separate issues that lead to slow or unstable learning. We address these shortcomings by treating the entropy coefficient $\alpha$ as a random variable and introduce Multi-Alpha Soft Actor-Critic (MAS). We show how MAS overcomes the stochastic bias of SAC in a variety of robotic control tasks including the CARLA urban-driving simulator, while maintaining the stability and sample efficiency of the original algorithms.

关键词：

来源：评论

学校读者我要写书评

暂无评论

learning Barrier-Certified Polynomial Dynamical Systems for Obstacle Avoidance with Robots

Learning Barrier-Certified Polynomial Dynamical Systems for ...

引用

IEEE International Conference on robotics and Automation (ICRA)

作者： Martin Schonger Hugo T. M. Kussaba Lingyun Chen Luis Figueredo Abdalla Swikir Aude Billard Sami Haddadin Munich Institute of Robotics and Machine Intelligence (MIRMI) Technical University of Munich (TUM) Germany School of Computer Science University of Nottingham UK Omar Al-Mukhtar University (OMU) Albaida Libya Learning Algorithms and Systems Laboratory EPFL Switzerland

ISBN: (数字)9798350384574

ISBN: (纸本)9798350384581

Established techniques that enable robots to learn from demonstrations are based on learning a stable dynamical system (DS). To increase the robots’ resilience to perturbations during tasks that involve static obstacle avoidance, we propose incorporating barrier certificates into an optimization problem to learn a stable and barrier-certified DS. Such optimization problem can be very complex or extremely conservative when the traditional linear parameter-varying formulation is used. Thus, different from previous approaches in the literature, we propose to use polynomial representations for DSs, which yields an optimization problem that can be tackled by sum-of-squares techniques. Finally, our approach can handle obstacle shapes that fall outside the scope of assumptions typically found in the literature concerning obstacle avoidance within the DS learning framework. Supplementary material can be found at the project webpage: https://***/abc-ds

关键词： Shape Scalability Perturbation methods Polynomials Collision avoidance Dynamical systems Task analysis

来源：评论

学校读者我要写书评

暂无评论

GO-VMP: Global Optimization for View Motion Planning in Fruit Mapping

arXiv

引用

arXiv 2025年

作者： Jose, Allen Isaac Pan, Sicong Zaenker, Tobias Menon, Rohit Houben, Sebastian Bennewitz, Maren Bonn-Rhein-Sieg University of Applied Sciences Germany Humanoid Robots Lab University of Bonn Germany Fraunhofer Institute for Intelligent Analysis and Information Systems Germany Humanoid Robots Lab University of Bonn Lamarr Institute for Machine Learning and Artificial Intelligence Center for Robotics University of Bonn Germany

Automating labor-intensive tasks such as crop monitoring with robots is essential for enhancing production and conserving resources. However, autonomously monitoring horticulture crops remains challenging due to their complex structures, which often result in fruit occlusions. Existing view planning methods attempt to reduce occlusions but either struggle to achieve adequate coverage or incur high robot motion costs. We introduce a global optimization approach for view motion planning that aims to minimize robot motion costs while maximizing fruit coverage. To this end, we leverage coverage constraints derived from the set covering problem (SCP) within a shortest Hamiltonian path problem (SHPP) formulation. While both SCP and SHPP are well-established, their tailored integration enables a unified framework that computes a global view path with minimized motion while ensuring full coverage of selected targets. Given the NP-hard nature of the problem, we employ a region-prior-based selection of coverage targets and a sparse graph structure to achieve effective optimization outcomes within a limited time. Experiments in simulation demonstrate that our method detects more fruits, enhances surface coverage, and achieves higher volume accuracy than the motion-efficient baseline with a moderate increase in motion cost, while significantly reducing motion costs compared to the coverage-focused baseline. Real-world experiments further confirm the practical applicability of our approach. Copyright © 2025, The Authors. All rights reserved.

关键词： Fruits

来源：评论

学校读者我要写书评

暂无评论

Innovative Application of Socio-Cultural-System-Inspired Algorithms in IoT and Smart Systems Optimization 5

Innovative Application of Socio-Cultural-System-Inspired Alg...

引用

5th IEEE International Conference on Electro-Computing Technologies for Humanity, NIGERCON 2024

作者： Ikpo, Chibueze V. Chinedu, Paschal Uchenna Ogbimi, Emuejevoke Francis Ikegwu, Anayor Chukwu Nweke, Henry Friday Nwankwo, Wilson Veritas University Department of Software Engineering Bwari Abuja Nigeria Delta State University of Science and Technology Department of Information Systems and Technology Ozoro Nigeria David Umahi Federal University of Health Sciences International Institute for Machine Learning Robotics and Ai Research Ebonyi Nigeria Delta State University of Science and Technology Department of Cyber Security Ozoro Nigeria

ISBN: (纸本)9798331542559

The Odigbo Metaheuristic Optimization Algorithm (OMOA), inspired by the strategic and goal-oriented behaviors of the Ndigbo tribe in Africa, offers a novel approach to solving complex scientific and engineering problems. Drawing on the principles of exploration and exploitation inherent in the traditional Igba apprenticeship system, this algorithm is designed to optimize resource allocation, task scheduling in IoT environments, and feature selection in smart systems. This study evaluates the algorithm's performance through comparative analysis with established optimization methods, demonstrating significant improvements in system responsiveness, resource management, and overall operational efficiency. The findings suggest that the OMOA not only preserves cultural heritage but also provides a powerful tool for modern engineering challenges. © 2024 IEEE.

关键词： Igbo-Socio-Cultural IoT Metaheuristics Optimization Smart Systems

来源：评论

学校读者我要写书评

暂无评论

A deep learning approach for cultural heritage building classification using transfer learning and data augmentation

引用

Journal of Cultural Heritage 2025年 74卷 214-224页

作者： André Luiz Carvalho Ottoni Lara Toledo Cordeiro Ottoni Machine Learning and Robotics Research Group (MLBots) Department of Computing Federal University of Ouro Preto (UFOP) Campus Morro do Cruzeiro Ouro Preto 35400-000 Minas Gerais Brazil Department of Industrial Automation Federal Institute of Minas Gerais (IFMG) Campus Ouro Preto Ouro Preto 35400-000 Minas Gerais Brazil

The detection of architectural components in historic buildings is essential for digital documentation and the conservation process of cultural heritage. In this regard, recent studies have explored artificial intelligence with computer vision to enhance the detection of key components in monuments. However, this field of research still lacks investigation into the influence of using transfer learning and data augmentation to improve the performance of machine learning models. Moreover, the literature still requires research on Artificial Intelligence applied to Brazilian colonial architecture. Thus, this study proposes a new deep learning approach for cultural heritage building classification using transfer learning and data augmentation. For this purpose, the ImageMG dataset is proposed, containing 6449 images of 94 historic buildings from the state of Minas Gerais (Brazil), categorized into five classes: fronton, church, door, window, and tower. Additionally, the influence of using transfer learning to enhance the classification results of the Mobilenet architecture in the task of detecting components of historic buildings is evaluated. The proposed approach also investigates the effects of 64 combinations of data augmentation, utilizing six geometric transformations (zoom, width shift range, height shift range, vertical flip, horizontal flip, and rotation) for generating synthetic images to train the deep learning models. The results showed that the optimization of transfer learning in conjunction with data augmentation demonstrated significant advances in the performance of cultural heritage building classification. Experiments with the ImageMG dataset using transfer learning and vertical flip achieved the best accuracy results in validation (92.37 %), test 1 (90.22 %), and test 2 (87.33 %).

关键词：

来源：评论

学校读者我要写书评

暂无评论

Functional TMS Mapping During Sensorimotor Integration Task 5

Functional TMS Mapping During Sensorimotor Integration Task

引用

5th International Conference Neurotechnologies and Neurointerfaces, CNN 2023

作者： Udoratina, Anna Grigorev, Nikita Savosenkov, Andrey Ermolaev, Denis Maksimenko, Vladimir Gordleeva, Susanna Lobachevsky State University of Nizhny Novgorod Neurotechnology Department Nizhny Novgorod Russia Center for Technologies in Robotics and Mechatronics Components Innopolis University Innopolis Russia Lobachevsky State University of Nizhny Novgorod Mathematics and Mechanics Department Nizhny Novgorod Russia Center for Neurotechnology and Machine Learning Immanuel Kant Baltic Federal University Kaliningrad Russia Neuroscience Research Institute Samara State Medical University Samara Russia

ISBN: (纸本)9798350344059

In the present research, we studied spatiotemporal influence of a single-pulse TMS on the process of sensorimotor integration. We considered how real and sham stimulation of the left or right premotor, motor or sensorimotor cortex, delivered 200 or 800 ms after the demonstration of a visual stimuli, affected response time of subjects during sensorimotor integration task. At this stage of the study, we found that a single TMS of the sensorimotor cortex, delivered 800 ms after stimulus presentation, increases response time. © 2023 IEEE.

关键词： Integration

来源：评论

学校读者我要写书评

暂无评论

Exploration via planning for information about the optimal trajectory 22

Exploration via planning for information about the optimal t...

引用

Proceedings of the 36th International Conference on Neural Information Processing Systems

作者： Viraj Mehta Ian Char Joseph Abbate Rory Conlin Mark D. Boyer Stefano Ermon Jeff Schneider Willie Neiswanger Robotics Institute Machine Learning Department Carnegie Mellon University Princeton University Princeton Plasma Physics Laboratory Computer Science Department Stanford University

ISBN: (纸本)9781713871088

Many potential applications of reinforcement learning (RL) are stymied by the large numbers of samples required to learn an effective policy. This is especially true when applying RL to real-world control tasks, e.g. in the sciences or robotics, where executing a policy in the environment is costly. In popular RL algorithms, agents typically explore either by adding stochasticity to a reward-maximizing policy or by attempting to gather maximal information about environment dynamics without taking the given task into account. In this work, we develop a method that allows us to plan for exploration while taking both the task and the current knowledge about the dynamics into account. The key insight to our approach is to plan an action sequence that maximizes the expected information gain about the optimal trajectory for the task at hand. We demonstrate that our method learns strong policies with 2x fewer samples than strong exploration baselines and 200x fewer samples than model free methods on a diverse set of low-to-medium dimensional control tasks in both the open-loop and closed-loop control settings.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Toward Accurate Online Multi-target Multi-camera Tracking in Real-time

Toward Accurate Online Multi-target Multi-camera Tracking in...

引用

European Signal Processing Conference (EUSIPCO)

作者： Andreas Specker Jürgen Beyerer Fraunhofer IOSB Karlsruhe Germany Vision and Fusion Lab Institute for Anthropomatics and Robotics Karlsruhe Institute of Technology (KIT) Karlsruhe Germany Fraunhofer Center for Machine Learning Karlsruhe Germany

ISBN: (数字)9789082797091

ISBN: (纸本)9781665467995

Multi-target multi-camera tracking is the task of determining the trajectories of objects within a network of cameras. Besides many others, it is a crucial task, e.g., in traffic analysis or law enforcement. Current research mainly focuses on offline algorithms for cross-camera association, which require all video data at once and often lack real-time capability. Thus, such approaches are unsuitable for tracking a criminal's escape route and enabling immediate intervention and arrest. To close this gap, we introduce a novel online multi-target multi-camera tracking method in this work which integrates spatial-temporal information and applies a new probability-based association metric. Moreover, we prove the real-time capability and significantly outperform the baseline on two datasets. In addition, the performance gap to more complex and slower offline methods is discussed to indicate the current state of research.

关键词： Measurement Law enforcement Signal processing algorithms Europe Streaming media Signal processing Real-time systems

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：