检索结果-内蒙古大学图书馆

7th International Conference on Intelligent robotics and Control engineering, IRCE 2024

作者： Yoseph, Abey Tanveer, M. Hassan Voicu, Razvan Cristian Shahid, M. Farrukh Kennesaw State University Department of Robotics & Mechatronics Engineering MariettaGA United States National University of Computer and Emerging Sciences Department of Ai and Data Science Karachi Pakistan

ISBN: (纸本)9798350352399

Machine learning is a critical component of artificial intelligence and computer science that uses data and algorithms to emulate the way that humans learn and predict. Reinforcement learning, one of the many forms of machine learning, trains an agent through trial and error, with successful outcomes being rewarded and unsuccessful outcomes being penalized. The implementation of three reinforcement learning algorithms, Qlearning, Deep Q-Learning, and Double Deep Q-Learning, was tested using a series of ROS (Robot Operating System) packages in the Gazebo simulation environment. The Python openai-ros package and Gym libraries were used to input a robot into a configurable environment and test out the performance of a specific reinforcement learning algorithm. The agent's action space was discretized into three specific actions, move forward, turn left, or turn right, that the robot could take. The reward or punishment was given after each step based on if the current action moved the robot towards the goal state or away from it. For each algorithm, the agent was first pre-trained on an obstacle free environment for 200 episodes and then was placed in an environment with two box-shaped obstacles to avoid. The algorithm/neural network parameters from the pre-training environment were loaded prior to the agent being tested in the new environment. The goal state at each episode was randomly generated and the robot's performance was evaluated using the cumulative reward achieved per episode and the number of times it reached the goal. © 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

RT-Swap: Addressing GPU Memory Bottlenecks for Real-Time Multi-DNN Inference 30

RT-Swap: Addressing GPU Memory Bottlenecks for Real-Time Mul...

引用

30th IEEE Real-Time and Embedded Technology and Applications Symposium, RTAS 2024

作者： Kang, Woosung Lee, Jinkyu Lee, Youngmoon Oh, Sangeun Lee, Kilho Chwa, Hoon Sung DGIST Dept. of Electrical Engineering and Computer Science Korea Republic of Dept. of Computer Science and Engineering Korea Republic of Hanyang University Dept. of Robotics Korea Republic of Ajou University Department of Software and Computer Engineering Korea Republic of School of AI Convergence Soongsil University Korea Republic of

ISBN: (纸本)9798350358414

The increasing complexity and memory demands of Deep Neural Networks (DNNs) for real-Time systems pose new significant challenges, one of which is the GPU memory capacity bottleneck, where the limited physical memory inside GPUs impedes the deployment of sophisticated DNN models. This paper presents, to the best of our knowledge, the first study of addressing the GPU memory bottleneck issues, while simultaneously ensuring the timely inference of multiple DNN tasks. We propose RT-Swap, a real-Time memory management framework, that enables transparent and efficient swap scheduling of memory objects, employing the relatively larger CPU memory to extend the available GPU memory capacity, without compromising timing guarantees. We have implemented RT-Swap on top of representative machine-learning frameworks, demonstrating its effectiveness in making significantly more DNN task sets schedulable at least 72% over existing approaches even when the task sets demand up to 96.2% more memory than the GPU's physical capacity. © 2024 IEEE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Scanning the Issue

引用

IEEE Transactions on Intelligent Transportation Systems 2024年第7期25卷 6251-6275页

作者： Simona Sacone Department of Computer Science Bioengineering Robotics and Systems Engineering University of Genova Genova Italy

The authors provide a meta-review across nearly 100 extant surveys on unmanned aerial vehicles (UAVs) applications and overview papers, extract their key messages, and investigate the extent of being complementary. Based on the findings of their survey analysis, they developed the so-called AERIAL framework, which aggregates the major challenges on the way to a successful application of UAVs for logistics, mobility, and monitoring. The framework covers the following aspects: 1) acceptance and heterogeneities; 2) economy and demand; 3) reliability and risks; 4) infrastructure and optimization; 5) automation and control; and 6) liability and standardization. They believe that their AERIAL framework and meta-review contribute towards a clearer understanding of the scientific UAV landscape challenges and the identification of potential directions for future research studies.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Online 3D Deformable Object Classification for Mobile Cobot Manipulation 56

Online 3D Deformable Object Classification for Mobile Cobot ...

引用

56th International Symposium on robotics, ISR Europe 2023

作者： Nguyen, Khang Dang, Tuan Huber, Manfred Learning and Adaptive Robotics Laboratory Department of Computer Science and Engineering University of Texas at Arlington ArlingtonTX76013 United States

ISBN: (纸本)9783800761418

Vision-based object manipulation in assistive mobile cobots essentially relies on classifying the target objects based on their 3D shapes and features, whether they are deformed or not. In this work, we present an auto-generated dataset of deformed objects specific for assistive mobile cobot manipulation using an intuitive Laplacian-based mesh deformation procedure. We first determine the graspable region of the robot hand on the given object’s mesh. Then, we uniformly sample handle points within the graspable region and perform deformation with multiple handle points based on the robot gripper configuration. In each deformation, we identify the orientation of handle points and prevent self-intersection to guarantee the object’s physical meaning when multiple handle points are simultaneously applied to the mesh at different deformation intensities. We also introduce a lightweight neural network for 3D deformable object classification. Finally, we test our generated dataset on the Baxter robot with two 7-DOF arms, an integrated RGB-D camera, and a 3D deformable object classifier. The result shows that the robot is able to classify real-world deformed objects from point clouds captured at multiple views by the RGB-D camera. The source code is available at https://***/mkhangg/deformable_cobot. © VDE VERLAG GMBH Berlin Offenbach.

关键词： 3D vision deformable object manipulation

来源：评论

学校读者我要写书评

暂无评论

Reactive Multi-agent Coordination using Auction-based Task Allocation and Behavior Trees

Reactive Multi-agent Coordination using Auction-based Task A...

引用

2023 IEEE Conference on Control Technology and Applications, CCTA 2023

作者： Dahlquist, Niklas Lindqvist, Bjorn Saradagi, Akshit Nikolakopoulos, George Luleå University of Technology Robotics and Ai Group Department of Computer Science Electrical and Space Engineering LuleåSE-97187 Sweden

ISBN: (纸本)9798350335446

This article presents an architecture for multi-agent task allocation and task execution, through the unification of a market-inspired task-auctioning system with Behavior Trees for managing and executing lower level behaviors. We consider the scenario with multi-stage tasks, such as 'pick and place', whose arrival times are not known a priori. In such a scenario, a coordinating architecture is expected to be reactive to newly arrived tasks and the resulting rerouting of agents should be dependent on the stage of completion of their current multi-stage tasks. In the novel architecture proposed in this article, a central auctioning system gathers bids (cost-estimates for completing currently available tasks) from all agents, and solves a combinatorial problem to optimally assign tasks to agents. For every agent, it's participation in the auctioning system and execution of an assigned multi-stage task is managed using behavior trees, which switch among several well-defined behaviors in response to changing scenarios. The auctioning system is run at a fixed rate, allowing for newly added tasks to be incorporated into the auctioning system, which makes the solution reactive and allows for rerouting of some agents (subject to the states of the behavior trees). We demonstrate that the proposed architecture is especially well-suited for multistage tasks, where high costs are incurred when rerouting agents who have completed one or more stages of their current tasks. The proposed framework is experimentally validated in multiple scenarios in a lab environment. A video of a demonstration can be found at: https://***/ZdEkoOOlB2g. © 2023 IEEE.

关键词： Cost estimating

来源：评论

学校读者我要写书评

暂无评论

Drone-Based Predictive Structure Analysis: A Review

Drone-Based Predictive Structure Analysis: A Review

引用

2023 International Conference on Sustainable Emerging Innovations in engineering and Technology, ICSEIET 2023

作者： Patil, Trushant Nandan, Durgesh Sharma, Kanhaiya Parashar, Deepak Department of Robotics and Automation Pune India Department of Electronics and Telecommunication Engg. Pune India Department of Computer Science & Engineering Pune India Faculty of Engineering Pune India

ISBN: (纸本)9798350329186

Drone-based predictive structure analysis is a rapidly growing field that has the potential to revolutionize infrastructure and building inspection. In this review of the research, we give a general overview of how drones could increase the effectiveness and precision of structural analysis, particularly for keeping an eye on difficult-to-reach locations. The usage of drones for 3D mapping, inspection, and monitoring of diverse constructions, such as concrete buildings, bridges, tunnels, and dams. The investigations show that drone-based inspection can produce 3D models and high-resolution photos, allowing for remote monitoring and the early identification of possible problems. The findings show that drones may greatly boost infrastructure management's safety and effectiveness, making them an important tool for predictive structural research. © 2023 IEEE.

关键词： Drones

来源：评论

学校读者我要写书评

暂无评论

Robust STL Control Synthesis under Maximal Disturbance Sets 63

Robust STL Control Synthesis under Maximal Disturbance Sets

引用

63rd IEEE Conference on Decision and Control, CDC 2024

作者： Verhagen, Joris Lindemann, Lars Tumova, Jana Kth Royal Institute of Technology Division of Robotics Perception and Learning School of Electrical Engineering and Computer Science Stockholm Sweden University of Southern California Thomas Lord Department of Computer Science Los AngelesCA United States

ISBN: (纸本)9798350316339

This work addresses maximally robust control synthesis under unknown disturbances. We consider a nonlinear system, subject to a Signal Temporal Logic (STL) specification and jointly synthesize the maximal possible disturbance bounds and the corresponding controllers that ensure the STL specification is satisfied under these bounds. Many works have considered STL satisfaction under given bounded disturbances yet, to the authors' best knowledge, this is the first work that aims to maximize the permissible disturbance set and find corresponding maximally robust controllers. We, therefore, introduce disturbance robustness as a model-based robustness metric for STL planning and control synthesis. We extend the notion of disturbance-robust semantics for STL, which is a property of a specification, dynamical system, and controller, and provide an algorithm for maximally robust controllers satisfying an STL specification using Hamilton-Jacobi reachability. We show its soundness and provide a simulation example with an Autonomous Underwater Vehicle (AUV). © 2024 IEEE.

关键词： Control system synthesis

来源：评论

学校读者我要写书评

暂无评论

Edge AI-Driven Video Analytics: A Mobile Deep Learning Framework for Victim Detection in SAR robotics 3

Edge AI-Driven Video Analytics: A Mobile Deep Learning Frame...

引用

3rd FORTEI-International Conference on Electrical engineering, FORTEI-ICEE 2024

作者： Ihsan, Muhammad Putra, Karisma Trinanda Mustar, Muhamad Yusvin Setyawan, Aufa Faiz Chu, Hsueh-Ting Chen, Hsing-Chung Department of Electrical Engineering Universitas Muhammadiyah Yogyakarta Bantul Indonesia Department of Electrical Engineering Center of Artificial Intelligence and Robotics Studies Universitas Muhammadiyah Yogyakarta Bantul Indonesia Department of Computer Science and Information Engineering Asia University Taichung Taiwan

ISBN: (纸本)9798331542207

In post-disaster search and rescue (SAR) missions, it is crucial for robots to distinguish between actual victims and dummy objects, despite their similar characteristics. Edge video analytics have demonstrated exceptional performance in such missions, however, the trade-off between high-performing deep learning models and computational complexity often results in inadequate computational efficiency for SAR robotics. This study tackles this challenge by deploying a deep learning (DL) algorithm, i.e., MobileNet Single Shot Detector (SSD), specifically tailored for embedded systems, to perform object detection and classification tasks on lab-scale SAR robots. The proposed algorithm is developed in Python and implemented on a low-powered edge computing device. The effectiveness of the proposed algorithm in object localization is highlighted by its mAP score of 0.6247 and an average IoU of 0.9036. Experimental tests using a confusion matrix showed that the model accurately predicted 42 dummy objects and 43 victim dolls. The model achieved an overall accuracy of 89.47%, with a precision of 1.0, a recall of 0.8947, and an F1 Score of 0.9444. Real-time testing confirmed the model’s robust performance, maintaining high accuracy and a stable frame rate of 47 FPS. These findings corroborate the efficacy of the MobileNet SSD algorithm in reliably detecting and differentiating objects within embedded SAR robotic operations. © 2024 IEEE.

关键词： Mobile edge computing

来源：评论

学校读者我要写书评

暂无评论

Contact-Rich Task Learning on an Articulated Soft Robot Arm Through Simulation 25th

Contact-Rich Task Learning on an Articulated Soft Robot Ar...

引用

25th Annual Conference on Towards Autonomous Robotic Systems, TAROS 2024

作者： Elstner, Laurenz Kyrkjebø, Erik Stoelen, Martin F. Department of Computer Science Electrical Engineering and Mathematical Sciences Western Norway University of Applied Sciences Førde Norway Fieldwork Robotics Ltd. Cambridge United Kingdom

ISBN: (纸本)9783031720611

Traditional rigid robots face limitations in exploration during learning processes in dynamic and unknown environments where contact and impacts are common. Soft robots that leverage compliant materials, for example as part of variable stiffness actuators (VSAs), offer promising avenues for enhanced adaptability and safety. Deep Reinforcement Learning (RL), while proven to be a powerful tool, requires a lot of training data that is time-consuming to gather. This paper introduces a novel approach to simulating robots driven by antagonistic VSAs, as well as full-task learning with a soft robot arm in three environments with different contact types: path following (no contact), door opening (kinematic constraint), and surface wiping (continuous contact). We show that the tendon-driven robot can learn all three tasks, even if learning in a very low-level actuation space. The stretchable tendons enables it to perform relatively well on the contact-rich tasks. Further, using the proposed variable stiffness controller the robot is able to develop a stable behaviour in terms of forces applied to the environment. The results are promising for extending the approach to the physical robot. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Tendons

来源：评论

学校读者我要写书评

暂无评论

Intelligent Deep Learning Enabled Human Activity Recognition for Improved Medical Services

引用

computer Systems science & engineering 2023年第2期44卷 961-977页

作者： E.Dhiravidachelvi M.Suresh Kumar L.D.Vijay Anand D.Pritima Seifedine Kadry Byeong-Gwon Kang Yunyoung Nam Department of Electronics and Communication Engineering Mohamed Sathak A.J.College of EngineeringChennai603103India Department of Information Technology Sri Sairam Engineering CollegeChennai602109India Department of Robotics Engineering Karunya Institute of Technology and SciencesCoimbatore641114India Department of Mechatronics Engineering Sri Krishna College of Engineering and TechnologyCoimbatore641008India Deparmtent of Applied Data Science Noroff University CollegeKristiansandNorway Department of Information and Communication Engineering Soonchunhyang UniversityAsanKorea Department of Computer Science and Engineering Soonchunhyang UniversityAsanKorea

Human Activity Recognition(HAR)has been made simple in recent years,thanks to recent advancements made in Artiﬁcial Intelligence(AI)*** techniques are applied in several areas like security,surveillance,healthcare,human-robot interaction,and *** wearable sensor-based HAR system includes in-built sensors,human activities can be categorized based on sensor ***,it can also be employed in other applications such as gait diagnosis,observation of children/adult’s cognitive nature,stroke-patient hospital direction,Epilepsy and Parkinson’s disease examination,***-developed Artiﬁcial Intelligence(AI)techniques,especially Deep Learning(DL)models can be deployed to accomplish effective outcomes on HAR *** this motivation,the current research paper focuses on designing Intelligent Hyperparameter Tuned Deep Learning-based HAR(IHPTDL-HAR)technique in healthcare *** proposed IHPTDL-HAR technique aims at recogniz-ing the human actions in healthcare environment and helps the patients in mana-ging their healthcare *** addition,the presented model makes use of Hierarchical Clustering(HC)-based outlier detection technique to remove the ***-HAR technique incorporates DL-based Deep Belief Network(DBN)model to recognize the activities of ***,Harris Hawks Opti-mization(HHO)algorithm is used for hyperparameter tuning of DBN ***,a comprehensive experimental analysis was conducted upon benchmark dataset and the results were examined under different *** experimental results demonstrate that the proposed IHPTDL-HAR technique is a superior per-former compared to other recent techniques under different measures.

关键词： Artiﬁcial intelligence human activity recognition deep learning deep belief network hyperparameter tuning healthcare

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：