检索结果-内蒙古大学图书馆

Grid-level Multiclass Soiling Classification Model for Driving systems

Journal of Institute of Control, robotics and systems 2024年第10期30卷 1076-1081页

作者： Kim, Taewoo Hwang, Youngbae Department of Intelligent Systems and Robotic Chungbuk National University Korea Republic of

This study proposes a soiling detection algorithm to identify and locate contamination in vehicle camera lenses. Research on AI applications that utilize cameras and distance sensors in driving systems is directly related to the advancement of autonomous driving systems., Detecting soiling, such as mud and water droplets, is a particularly critical issue. Traditional methods using piezoelectric, ultrasonic, and thermal sensors can introduce additional design and maintenance complexity. Therefore, this study aims to reduce this complexity by utilizing existing surround-view cameras installed on vehicles without additional sensors. The proposed algorithm employs image-processing techniques and a lightweight neural network architecture based on ResNet18 to detect lens contamination in real-time across various driving environments. Experiments were conducted using 5,000 images from the WoodScape Soiling Dataset. The input images were divided into a 16 × 16 grid and classified into four labels: opaque, semi-transparent, transparent, and clean. The proposed grid-level multiclass soiling classification model demonstrated effective performance in detecting contamination, including mud, water droplets, and foggy dust. This study is expected to enhance the safety and convenience of driving systems. © ICROS 2024.

关键词： Camera lenses

来源：评论

学校读者我要写书评

暂无评论

Seeing Through the Grass: Semantic Pointcloud Filter for Support Surface Learning

引用

IEEE robotics and Automation Letters 2023年第11期8卷 7687-7694页

作者： Li, Anqiao Yang, Chenyu Frey, Jonas Lee, Joonho Cadena, Cesar Hutter, Marco Eth Zurich Robotic Systems Lab Zurich8092 Switzerland Max Planck Institute for Intelligent Systems Tübingen72076 Germany

Mobile ground robots require perceiving and understanding their surrounding support surface to move around autonomously and safely. The support surface is commonly estimated based on exteroceptive depth measurements, e.g., from LiDARs. However, the measured depth fails to align with the true support surface in the presence of high grass or other penetrable vegetation. In this work, we present the semantic pointcloud filter (SPF), a convolutional neural network (CNN) that learns to adjust LiDAR measurements to align with the underlying support surface. The SPF is trained in a semi-self-supervised manner and takes as an input a LiDAR pointcloud and RGB image. The network predicts a binary segmentation mask that identifies the specific points requiring adjustment, along with estimating their corresponding depth values. To train the segmentation task, 464 distinct images are manually labeled into rigid and non-rigid terrain. The depth estimation task is trained in a self-supervised manner by utilizing the future footholds of the robot to estimate the support surface based on a Gaussian process. Our method can correctly adjust the support surface prior to interacting with the terrain and is extensively tested on the quadruped robot ANYmal. We show the qualitative benefits of SPF in natural environments for elevation mapping and traversability estimation compared to using raw sensor measurements and existing smoothing methods. Quantitative analysis is performed in various natural environments, and an improvement by 48% RMSE is achieved within a meadow terrain. © 2016 IEEE.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Bayesian network structure learning by dynamic programming algorithm based on node block sequence constraints

引用

CAAI Transactions on Intelligence Technology 2024年第6期9卷 1605-1622页

作者： Chuchao He Ruohai Di Bo Li Evgeny Neretin School of Electronics and Information Engineering Xi'an Technological UniversityXi'anChina School of Electronic Information Northwestern Polytechnical UniversityXi'anChina School of Robotic and Intelligent Systems Moscow Aviation InstituteMoscowRussia

The use of dynamic programming(DP)algorithms to learn Bayesian network structures is limited by their high space complexity and difficulty in learning the structure of large-scale ***,this study proposes a DP algorithm based on node block sequence *** proposed algorithm constrains the traversal process of the parent graph by using the M-sequence matrix to considerably reduce the time consumption and space complexity by pruning the traversal process of the order graph using the node block *** results show that compared with existing DP algorithms,the proposed algorithm can obtain learning results more efficiently with less than 1%loss of accuracy,and can be used for learning larger-scale networks.

关键词： Bayesian network(BN) dynamic programming(DP) node block sequence strongly connected component(SCC) structure learning

来源：评论

学校读者我要写书评

暂无评论

An Adaptive Neuro-Fuzzy Controller to Enhance Power Sharing in Distributed Energy Resources Applications

引用

IEEE Access 2025年 13卷 94667-94680页

作者： Seyedmohammad Hasheminasab Mohamad Alzayed Hicham Chaoui Department of Electronics Intelligent Robotic and Energy Systems (IRES) Research Group Carleton University Ottawa Canada

In inverter-interfaced microgrids, droop control techniques are essential for regulating active and reactive power exchange. However, their performance is compromised by the varying impedance of the feeder and the slow response to dynamic load changes, leading to power-sharing inaccuracies. This paper proposes an Adaptive Neuro-Fuzzy Inference System (ANFIS)-based virtual impedance controller to address this issue and enhance active and reactive power sharing. The proposed controller dynamically adjusts a virtual voltage to compensate for impedance mismatches, modifying the reference voltage of the inverter. This enables precise power tracking with minimal deviation from the defined reference values and a faster response under transient conditions, including startup and external disturbances. The ANFIS framework integrates fuzzy logic and neural networks, eliminating the limitations of manual and separate tuning in conventional controllers and improving performance in nonlinear systems. The controller’s performance is validated on an IEEE 39-bus test system under various scenarios, including charging, discharging, and transient disturbances. It is tested with three battery sizes (1 MW, 96 kW, and 75 kW) under the same controller setup to assess scalability. Training with per-unit data ensures scalability across different battery capacities and distributed generators. The results are compared to traditional methods to demonstrate the controller’s superior effectiveness.

关键词： Impedance Voltage control Power system stability Microgrids Reactive power Control systems PI control Batteries Adaptive systems Tuning

来源：评论

学校读者我要写书评

暂无评论

Human-Centric Concept for a Reconfigurable robotic System Enabling Low-Volume Assembly of Photonic and Quantum Modules

Human-Centric Concept for a Reconfigurable Robotic System En...

引用

2025 IEEE/SICE International Symposium on System Integration, SII 2025

作者： Wicht, Andreas Franke, Tobias Hahn, Alina Hakansson, Nils Kurbis, Christian Smol, Robert Hulin, Thomas Eiband, Thomas Lehner, Peter Muhlbauer, Maximilian Nottensteiner, Korbinian Pietschmann, Richard Thaler, Bernhard Thaler, Diana Bosse, Jurgen Berlin12489 Germany Institute of Robotics and Mechatronics Wessling82234 Germany Germany Technical University of Munich Sensor Based Robotic Systems and Intelligent Assistance Systems Garching85748 Germany Robo-Technology GmbH Puchheim82178 Germany

ISBN: (纸本)9798331531614

This paper presents a novel concept for a reconfigurable robotic system specifically designed to meet the demands of hybrid integration for miniaturized photonic and quantum System-in-Packages (SiPs). The proposed solution introduces a distinctive approach to ultra-high-resolution, multi-telerobotic assembly and inspection. By integrating key Industry 5.0 principles, it establishes a human-centric control framework that minimizes both physical and cognitive stress while ensuring the human operator remains in full control at all *** robotic system features eight robots working simultaneously within a compact footprint of just 5×10 cm2. A comprehensive digital twin framework constitutes a central element of the robotic system. It encompasses the robotic workcell, the SiP under assembly, and the components to be integrated, ensuring precise adherence to design specifications. Key functionalities include automated path planning in a multi-robotic environment, collision avoidance in a densely packed workcell, and virtual fixtures to guide teleoperation, enhancing the operator's control and interaction through advanced and intuitive human-machine interfaces (HMI). The proposed system meets the critical demands of ultra-high-resolution assembly for complex, high-value SiPs, providing high flexibility and ease of operation for small-batch manufacturing. © 2025 IEEE.

关键词： System-in-package

来源：评论

学校读者我要写书评

暂无评论

Lunar Rover's Path Tracking Control Strategy Using Zero-Order Optimization Method

Lunar Rover's Path Tracking Control Strategy Using Zero-Orde...

引用

2024 IEEE International Conference on Control Science and systems Engineering, ICCSSE 2024

作者： Cao, Jianfei Wang, Yaobing Yu, Jiyang Beijing Institute of Spacecraft System Engineering Beijing Key Laboratory of Intelligent Space Robotic Systems Technology and Applications Beijing China Beijing Institute of Spacecraft System Engineering Beijing China

ISBN: (纸本)9798331517199

The development of lunar rovers is critical for human exploration of the lunar surface. These rovers will need to travel longer distances at higher speeds, and the efficient control of their paths is essential for mission success. Model Predictive Control (MPC) is a path tracking method with foresight and robustness, capable of predicting vehicle movement and solving optimization problems to minimize cost functions. However, the complexity of lunar environments necessitates simplified models, such as kinematic models, to reduce computational cost. The study proposes a special MPC method that uses zeroth-order gradient optimization to directly solve the nonlinear model and generate precise control commands, significantly reducing computational burden and meeting the computational cost requirements of lunar rover control. © 2024 IEEE.

关键词： Cost functions

来源：评论

学校读者我要写书评

暂无评论

MADDPG Based Radar Interference Resource Allocation Decision

MADDPG Based Radar Interference Resource Allocation Decision

引用

2024 China Automation Congress, CAC 2024

作者： Jiang, Sen Wang, Bo School of HDU-ITMO Research Center for Intelligent Systems and Robotics Hangzhou Dianzi University Hangzhou310018 China School of Automation International Joint Research Laboratory for Autonomous Robotic Systems Hangzhou Dianzi University Hangzhou310018 China

ISBN: (纸本)9798350368604

For the problem of jamming penetration under the one-to-many scenario with jammer, a real-time radar interference resource allocation method based on the Multi-agent deep deterministic policy gradient (MADDPG) algorithm is proposed. This method describes the interference resource allocation model as a Markov Decision Process (MDP), and designed the state space, action space and reward function for interference resource allocation problem. The jammer is divided into three heterogeneous agents, each responsible for selecting jamming targets, jamming patterns, and jamming power, respectively. These three agents coordinate to enhance the convergence ability of the high-dimensional action space algorithm. Simulation results indicate that the jamming resource allocation method designed in this paper provides better jamming effects on radar positions, higher stability, and effectively improves the operational performance of the jammer. © 2024 IEEE.

关键词： Radar interference

来源：评论

学校读者我要写书评

暂无评论

Learning with 3D Rotations: a Hitchhiker's Guide to SO(3) 41

Learning with 3D Rotations: a Hitchhiker's Guide to SO(3)

引用

41st International Conference on Machine Learning, ICML 2024

作者： Geist, A. René Frey, Jonas Zhobro, Mikel Levina, Anna Martius, Georg Max Planck Institute for Intelligent Systems Autonomous Learning Group Germany University of Tübingen Distributed Intelligence Lab Germany Robotic Systems Lab Switzerland University of Tübingen Self-organization of Neuronal Networks Group Germany

Many settings in machine learning require the selection of a rotation representation. However, choosing a suitable representation from the many available options is challenging. This paper acts as a survey and guide through rotation representations. We walk through their properties that harm or benefit deep learning with gradient-based optimization. By consolidating insights from rotation-based learning, we provide a comprehensive overview of learning functions with rotation representations. We provide guidance on selecting representations based on whether rotations are in the model's input or output and whether the data primarily comprises small angles. The project code is available at: ***/martius-lab/hitchhiking-rotations. Copyright 2024 by the author(s)

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

ETHcavation: A Dataset and Pipeline for Panoptic Scene Understanding and Object Tracking in Dynamic Construction Environments

arXiv

引用

arXiv 2024年

作者： Terenzi, Lorenzo Nubert, Julian Eyschen, Pol Roth, Pascal Fei, Simin Jelavic, Edo Hutter, Marco Robotic Systems Lab ETH Zürich Switzerland MPI for Intelligent Systems Stuttgart Germany

Construction sites are challenging environments for autonomous systems due to their unstructured nature and the presence of dynamic actors, such as workers and machinery. This work presents a comprehensive panoptic scene understanding solution designed to handle the complexities of such environments by integrating 2D panoptic segmentation with 3D LiDAR mapping. Our system generates detailed environmental representations in real-time by combining semantic and geometric data, supported by Kalman Filter-based tracking for dynamic object detection. We introduce a fine-tuning method that adapts large pre-trained panoptic segmentation models for construction site applications using a limited number of domain-specific samples. For this use case, we release a first-of-its-kind dataset of 502 hand-labeled sample images with panoptic annotations from construction sites. In addition, we propose a dynamic panoptic mapping technique that enhances scene understanding in unstructured environments. As a case study, we demonstrate the system's application for autonomous navigation, utilizing real-time RRT* for reactive path planning in dynamic scenarios. The dataset1 and code2 for training and deployment are publicly available to support future research. Copyright © 2024, The Authors. All rights reserved.

关键词： Kalman filters

来源：评论

学校读者我要写书评

暂无评论

Enhancing GAN-Based Motion Data Augmentation Through Dynamic Time Warping Distance Filtering 6

Enhancing GAN-Based Motion Data Augmentation Through Dynamic...

引用

6th International Conference on Artificial Intelligence in Information and Communication, ICAIIC 2024

作者： Yoon, Junwon Chung, Hyun-Joon Kang, Jeon-Seong Kim, Jung-Jun Jeon, Kwang-Woo Kim, Seungwoo Shim, Myounghoon Ryu, Jae-Kwan Korea Institute of Robotics Technology Convergence AI Robotics RD Division Seoul06372 Korea Republic of Unmanned/Intelligent Robotic Systems RD LIG Nexl Seongnam13488 Korea Republic of

ISBN: (纸本)9798350344349

Motion capture data is crucial but creating a large dataset can be challenging due to complexities in acquisition. Generative Adversarial Network (GAN)-based motion data augmentation offers a potential solution to this issue. However, GANs often struggle with learning from limited data, resulting in poor quality output. In this study, we propose a Dynamic Time Warping (DTW) filtering method that filters out generated data significantly deviating from real-world examples. Through this approach, we have achieved an improvement in the fidelity of the generated data, even with dataset size constraints, as evidenced by an increase in action recognition accuracy. © 2024 IEEE.

关键词： Generative adversarial networks

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：