检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Bejarano, Federico Pizarro Brunke, Lukas Schoellig, Angela P. The Learning Systems and Robotics Lab University of Toronto Canada The University of Toronto Robotics Institute The Vector Institute for Artificial Intelligence Toronto Canada Germany

Reinforcement learning (RL) controllers are flexible and performant but rarely guarantee safety. Safety filters impart hard safety guarantees to RL controllers while maintaining flexibility. However, safety filters can cause undesired behaviours due to the separation between the controller and the safety filter, often degrading performance and robustness. In this paper, we analyze several modifications to incorporating the safety filter in training RL controllers rather than solely applying it during evaluation. The modifications allow the RL controller to learn to account for the safety filter, improving performance. This paper presents a comprehensive analysis of training RL with safety filters, featuring simulated and real-world experiments with a Crazyflie 2.0 drone. We examine how various training modifications and hyperparameters impact performance, sample efficiency, safety, and chattering. Our findings serve as a guide for practitioners and researchers focused on safety filters and safe RL. 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

Human-like Social Learning for Social Robots: A Systematic Review

Human-like Social Learning for Social Robots: A Systematic R...

引用

IEEE International Workshop on Robot and Human Communication (ROMAN)

作者： Diana Burkart Barbara Bruno Socially Assistive Robotics With Artificial Intelligence (SARAI) Lab Karlsruhe Institute of Technology (KIT) Germany

ISBN: (数字)9798350375022

ISBN: (纸本)9798350375039

Social learning is a learning paradigm aiming to make the process of teaching new skills to an artificial intelligent agent (such as a robot) as close as possible to the process we employ when teaching skills to other humans. Equipping robot companions with the ability to learn from social interactions with their users would enable naive users to effortlessly teach skills to robots and support personalization and long-term use. Aiming to support research on the field, in this article we present a systematic literature review of human-like social learning for social robots, with a specific focus on works employing real robots and real human-robot social interactions. The review examines relevant papers from perspectives including the type of robot used, the task to be learned, and the method and assessment metrics employed, and allows for the identification of open research avenues.

关键词： Measurement Education Social robots Transfer learning Hidden Markov models Learning (artificial intelligence) Reinforcement learning Planning Robots Lenses

来源：评论

学校读者我要写书评

暂无评论

Open-World Semantic Segmentation Including Class Similarity

arXiv

引用

arXiv 2024年

作者： Sodano, Matteo Magistri, Federico Nunes, Lucas Behley, Jens Stachniss, Cyrill Center for Robotics University of Bonn Germany Lamarr Institute for Machine Learning and Artificial Intelligence Germany

Interpreting camera data is key for autonomously acting systems, such as autonomous vehicles. Vision systems that operate in real-world environments must be able to understand their surroundings and need the ability to deal with novel situations. This paper tackles open-world semantic segmentation, i.e., the variant of interpreting image data in which objects occur that have not been seen during training. We propose a novel approach that performs accurate closed-world semantic segmentation and, at the same time, can identify new categories without requiring any additional training data. Our approach additionally provides a similarity measure for every newly discovered class in an image to a known category, which can be useful information in downstream tasks such as planning or mapping. Through extensive experiments, we show that our model achieves state-of-the-art results on classes known from training data as well as for anomaly segmentation and can distinguish between different unknown classes. Copyright © 2024, The Authors. All rights reserved.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

ASPIRe: An Informative Trajectory Planner with Mutual Information Approximation for Target Search and Tracking

ASPIRe: An Informative Trajectory Planner with Mutual Inform...

引用

IEEE International Conference on robotics and Automation (ICRA)

作者： Kangjie Zhou Pengying Wu Yao Su Han Gao Ji Ma Hangxin Liu Chang Liu Department of Advanced Manufacturing and Robotics College of Engineering Peking University National Key Laboratory of General Artificial Intelligence Beijing Institute for General Artificial Intelligence (BIGAI)

ISBN: (数字)9798350384574

ISBN: (纸本)9798350384581

This paper proposes an informative trajectory planning approach, namely, adaptive particle filter tree with sigma point-based mutual information reward approximation (ASPIRe), for mobile target search and tracking (SAT) in cluttered environments with limited sensing field of view. We develop a novel sigma point-based approximation to accurately estimate mutual information (MI) for general, non-Gaussian distributions utilizing particle representation of the belief state, while simultaneously maintaining high computational efficiency. Building upon the MI approximation, we develop the Adaptive Particle Filter Tree (APFT) approach with MI as the reward, which features belief state tree nodes for informative trajectory planning in continuous state and measurement spaces. An adaptive criterion is proposed in APFT to adjust the planning horizon based on the expected information gain. Simulations and physical experiments demonstrate that ASPIRe achieves real-time computation and outperforms benchmark methods in terms of both search efficiency and estimation accuracy.

关键词： Accuracy Target tracking Trajectory planning Robot sensing systems Real-time systems Particle filters Trajectory

来源：评论

学校读者我要写书评

暂无评论

Stochastic-Aware Mamba Diffusion for Pedestrian Trajectory Prediction

Stochastic-Aware Mamba Diffusion for Pedestrian Trajectory P...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Ziyang Ren Ping Wei Haowen Tang Huan Li Jin Yang Jialu Qin National Key Laboratory of Human-Machine Hybrid Augmented Intelligence Institute of Artificial Intelligence and Robotics Xi’an Jiaotong University Xi’an China

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Pedestrian trajectory prediction plays a crucial role in understanding human behavior and intentions. Due to the inherent randomness in human movement, current research constructs trajectories in stochastic space and uses diffusion models to reverse the denoising process. The commonly used denoising model, Transformer, is affected by uneven noise sampling, leading to confusion between temporal consistency and randomness across multiple time points. To address this issue, we propose a novel framework named Stochastic-Aware Mamba Diffusion (SAMD), which combines Stochastic-Aware Mamba with the Diffusion model to predict motion noise and motion states in the stochastic space. It utilizes a temporal aggregator to extract temporal consistency features. We construct a motion-selective state space model, which includes the adaptive transition between consistency motion states and stochastic states for balancing stability and diversity. The motion gating unit activates pertinent information within motion states to extract motion noise for trajectory prediction. Our approach achieves state-of-the-art results on the ETH-UCY and SDD datasets while significantly reducing the computational cost.

关键词： Pedestrians Computational modeling Noise reduction Noise Stochastic processes Predictive models Feature extraction Diffusion models Trajectory Data mining

来源：评论

学校读者我要写书评

暂无评论

A Compact LiDAR Odometry and Mapping with Dynamic Removal

A Compact LiDAR Odometry and Mapping with Dynamic Removal

引用

2024 IEEE International Conference on Unmanned Systems, ICUS 2024

作者： Zhang, Meifeng Jia, Yanpeng Shao, Shiliang Wang, Shiyi College of Artificial Intelligence Baoding University Baoding China The State Key Laboratory of Robotics Shenyang Institute of Automation Chinese Academy of Sciences Shenyang China School of Information Science and Engineering Shenyang Ligong University Shenyang China

ISBN: (纸本)9798350384185

Real-time state estimation and mapping is a fundamental function of unmanned systems. In this paper, we propose a compact 3D Lidar odometry and mapping method (C-LOAM) for real-time 6-DOF pose estimation and map construction for unmanned ground vehicles. Our approach firstly remove dynamic objects from the environment by comparing differences between two consecutive range images. We then extract ground points from the range image. The remaining points are segmented in the range image to obtain robust feature points for the registration stage. After performing two-stage scan-matching, we use the extracted ground points to impose ground constraints and leverage the local feature submap for loop closure detection. By employing various data reuse strategies, we minimize memory consumption, creating a compact and accurate system. Experiments on multiple publicly available datasets demonstrate that our system provides more accurate localization and higher-quality maps compared to other state-of-the-art methods. © 2024 IEEE.

关键词： Mapping

来源：评论

学校读者我要写书评

暂无评论

Harmony: Heterogeneous Multi-Modal Federated Learning through Disentangled Model Training 23

Harmony: Heterogeneous Multi-Modal Federated Learning throug...

引用

21st Annual International Conference on Mobile Systems, Applications and Services, MobiSys 2023

作者： Ouyang, Xiaomin Xie, Zhiyuan Fu, Heming Cheng, Sitong Pan, Li Ling, Neiwen Xing, Guoliang Zhou, Jiayu Huang, Jianwei Chinese University of Hong Kong Hong Kong Michigan State University United States Chinese University of HongKong Shenzhen China Shenzhen Institute of Artificial Intelligence and Robotics for Society China

ISBN: (纸本)9798400701108

Multi-modal sensing systems are increasingly prevalent in real-world applications such as health monitoring and autonomous driving. Most multi-modal learning approaches need to access users' raw data, which poses significant concerns to users' privacy. Federated learning (FL) provides a privacy-aware distributed learning framework. However, current FL approaches have not addressed the unique challenges of heterogeneous multi-modal FL systems, such as modality heterogeneity and significantly longer training delay. In this paper, we propose Harmony, a new system for heterogeneous multi-modal federated learning. Harmony disentangles the multi-modal network training in a novel two-stage framework, namely modality-wise federated learning and federated fusion learning. By integrating a novel balance-aware resource allocation mechanism in modality-wise FL and exploiting modality biases in federated fusion learning, Harmony improves the model accuracy under non-i.i.d. data distributions and speeds up system convergence. We implemented Harmony on a real-world multi-modal sensor testbed deployed in the homes of 16 elderly subjects for Alzheimer's Disease monitoring. Our evaluation on the testbed and three large-scale public datasets of different applications show that, Harmony outperforms by up to 46.35% accuracy over state-of-the-art baselines and saves up to 30% training delay. © 2023 Owner/Author(s).

关键词： Testbeds

来源：评论

学校读者我要写书评

暂无评论

An Underdetermined Wideband 2-D DOA Estimation Algorithm via Acoustic Vector-Sensor Coprime Array

An Underdetermined Wideband 2-D DOA Estimation Algorithm via...

引用

2023 IEEE International Conference on Signal Processing, Communications and Computing, ICSPCC 2023

作者： Han, Xueyan Liu, Meiqin Zhang, Senlin College of Information Engineering Binzhou University Binzhou China Institute of Artificial Intelligence and Robotics Xi'An Jiaotong University Xian China College of Electrical Engineering Zhejiang University Hangzhou China

ISBN: (纸本)9798350316728

In this paper, the issue of underdetermined two-dimensional (2-D) direction-of-arrival (DOA) estimation of underwater wideband signals is investigated by using an acoustic vector-sensor extended coprime array. Firstly, a novel virtual extended acoustic vector-sensor array is constructed and the model of the received wideband signals is derived under this array. Then, a binary selection matrix is designed to decrease the computational complexity by reducing dimensions. Finally, an accurate wideband DOA estimation is achieved by solving an optimization problem based on joint sparse recovery theory in the virtual domain, which fully uses the joint sparsity between different sub-bands. Some experimental simulations are carried out to confirm the effectiveness of the proposed algorithm. Some experimental simulations are carried out to confirm the effectiveness of the proposed algorithm. © 2023 IEEE.

关键词： Direction of arrival

来源：评论

学校读者我要写书评

暂无评论

Probabilistic Visibility-Aware Trajectory Planning for Target Tracking in Cluttered Environments

Probabilistic Visibility-Aware Trajectory Planning for Targe...

引用

American Control Conference (ACC)

作者： Han Gao W. Pengying S. Yao Kangjie Zhou M. Ji Hangxin Liu Chang Liu Department of Advanced Manufacturing and Robotics College of Engineering Peking University National Key Laboratory of General Artificial Intelligence Beijing Institute for General Artificial Intelligence (BIGAI)

ISBN: (数字)9798350382655

ISBN: (纸本)9798350382662

Target tracking has numerous significant civilian and military applications, and maintaining the visibility of the target plays a vital role in ensuring the success of the tracking task. Existing visibility-aware planners primarily focus on keeping the target within the limited field of view of an onboard sensor and avoiding obstacle occlusion. However, the negative impact of system uncertainty is often neglected, rendering the planners delicate to uncertainties in practice. To bridge the gap, this work proposes a model predictive control (MPC)-based trajectory planner for real-time visibility-aware and safe target tracking in the presence of system uncertainty. For more accurate target motion prediction, we introduce the concept of belief-space probability of detection (BPOD) to measure the predictive visibility of the target under stochastic robot and target states. An Extended Kalman Filter variant incorporating BPOD is developed to predict target belief state under uncertain visibility within the planning horizon. To reach real-time trajectory planning, we propose a computationally efficient algorithm to uniformly calculate both BPOD and the chance-constrained collision risk by utilizing linearized signed distance function (SDF), and subsequently solve the MPC problem by sequential convex programming. Extensive simulation results with benchmark comparisons show the capacity of the proposed approach to robustly maintain the visibility of the target under high system uncertainty. The practicality of the proposed trajectory planner is validated by real-world experiments.

关键词： Uncertainty Target tracking Trajectory planning Stochastic systems Simulation Stochastic processes Robot sensing systems

来源：评论

学校读者我要写书评

暂无评论

Learning Goal-oriented Bimanual Dough Rolling Using Dynamic Heterogeneous Graph Based on Human Demonstration

Learning Goal-oriented Bimanual Dough Rolling Using Dynamic ...

引用

2024 IEEE International Conference on robotics and Biomimetics, ROBIO 2024

作者： Liu, Junjia Li, Chenzui Wang, Shixiong Dong, Zhipeng Lam, Tin Lun Calinon, Sylvain Li, Miao Chen, Fei T-Stone Robotics Institute The Chinese University of Hong Kong Department of Mechanical and Automation Engineering Hong Kong Hong Kong School of Science and Engineering The Chinese University of Hong Kong Shenzhen Institute of Artificial Intelligence and Robotics for Society Shenzhen China Sylvain Calinon is with the Idiap Research Institute Martigny Switzerland School of Microelectronics Institute of Technological Sciences Wuhan University Wuhan China

ISBN: (纸本)9781665481090

Soft object manipulation poses significant chal-lenges for robots, requiring effective techniques for state representation and manipulation policy learning. State representation involves capturing the dynamic changes in the environment, while manipulation policy learning focuses on establishing the relationship between robot actions and state transformations to achieve specific goals. To address these challenges, this research paper introduces a novel approach: a dynamic heterogeneous graph-based model for learning goal-oriented soft object manipulation policies. The proposed model utilizes graphs as a unified representation for both states and policy learning. By leveraging the dynamic graph, we can extract crucial information regarding object dynamics and manipulation policies. Furthermore, the model facilitates the integration of demonstrations, enabling guided policy learning. To evaluate the efficacy of our approach, we designed a dough rolling task and conducted experiments using both a differentiable simulator and a real-world humanoid robot. Additionally, several ablation studies were performed to analyze the effect of our method, demonstrating its superiority in achieving human-like behavior. © 2024 IEEE.

关键词： Adversarial machine learning

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：