检索结果-内蒙古大学图书馆

AirShot: Efficient Few-Shot Detection for Autonomous Exploration

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Wang, Zihan Li, Bowen Wang, Chen Scherer, Sebastian Institute for Imaging Data and Communications School of Engineering The University of Edinburgh United Kingdom AirLab Robotics Institute Carnegie Mellon University PittsburghPA15213 United States Lab Institute for Artificial Intelligence and Data Science Department of Computer Science and Engineering University at Buffalo NY14260 United States

Few-shot object detection has drawn increasing attention in the field of robotic exploration, where robots are required to find unseen objects with a few online provided examples. Despite recent efforts have been made to yield online processing capabilities, slow inference speeds of low-powered robots fail to meet the demands of real-time detection-making them impractical for autonomous exploration. Existing methods still face performance and efficiency challenges, mainly due to unreliable features and exhaustive class loops. In this work, we propose a new paradigm AirShot, and discover that, by fully exploiting the valuable correlation map, AirShot can result in a more robust and faster few-shot object detection system, which is more applicable to robotics community. The core module Top Prediction Filter (TPF) can operate on multi-scale correlation maps in both the training and inference stages. During training, TPF supervises the generation of a more representative correlation map, while during inference, it reduces looping iterations by selecting top-ranked classes, thus cutting down on computational costs with better performance. Surprisingly, this dual functionality exhibits general effectiveness and efficiency on various off-the-shelf models. Exhaustive experiments on COCO2017, VOC2014, and SubT datasets demonstrate that TPF can significantly boost the efficacy and efficiency of most off-the-shelf models, achieving up to 36.4% precision improvements along with 56.3% faster inference speed. Code and Data are at: https://***/ImNotPrepared/AirShot. Copyright © 2024, The Authors. All rights reserved.

关键词： Efficiency

Seafloor Imaging Maps by Incorporating Side Scan Sonar Technology on Hybrid Autonomous Underwater Glider

学校读者我要写书评

暂无评论

Seafloor Imaging Maps by Incorporating Side Scan Sonar Techn...

International Seminar on Research of Information Technology and Intelligent Systems (ISRITI)

作者： Ardian Budi Kusuma Atmaja Egi Muhammad Idris Hidayat Bambang Riyanto Trilaksono Simon Siregar Natsir Habibullah School of Electrical Engineering and Informatics Institut Teknologi Bandung Bandung Indonesia Department Computer Engineering Faculty of Applied Science Doctoral Program of Electrical Engineering Institut Teknologi Bandung Telkom University

This research aims to construct a two-dimensional image to represent an underwater geometry map with a Side Scan Sonar (SSS) mounted on a Hybrid Autonomous Underwater Glider (HAUG). Building the underwater map has two stages of the process. The first stage is preprocessing which includes Time-varying gains (TVG), slant range correction, and ground range correction. The second is HAUG navigation, the movement of HAUG orientation to the angle and distance of SSS readings and plotting the geometry of the grid map. With these two stages, a two-dimensional grid map will be formed. Meanwhile, the 3D plot utilizes slant range and ground range correction to obtain the surface height of the bottom of the pool. From the experiments conducted to conduct mapping in the pool, two-dimensional and three-dimensional visual forms of the pool floor and pool walls were produced. The results of the estimated measurement of the width of the pool obtained 14.05 meters and the height of the pool 3.199 meters with an SSS beam angle of 58 degrees.

关键词：

Choosing a Suitable Requirement Prioritization Method: A Survey

学校读者我要写书评

暂无评论

SSRN

SSRN 2023年

作者： Mahmoud Alhenawi, Esra'a Awawdeh, Shatha Khurma, Ruba Abu Castillo, Pedro A. Department of Computer Science King Abdullah II School of Information Technology The University of Jordan Amman Jordan Applied Science Private University Jordan Faculty of Information Technology Computer Science department Al-Ahliyya Amman University Amman Jordan Department of Computer Engineering Automation and Robotics University of Granada Granada Spain

Software requirements prioritization plays a crucial role in software development. It can be viewed as the process of ordering requirements by determining which requirements must be done first and which can be done later. Powerful requirements prioritization techniques are of paramount importance to finish the implementation on time and budget. Many factors affect requirement prioritization such as stakeholder expectations, complexity, dependency, scalability, risk and cost. Therefore, finding the proper order of requirements is a challenging process. Hence, different types of requirements prioritization techniques have been developed to support this task. In this survey we propose a novel classification that can classify the prioritization techniques under two major classes: relative and exact prioritization techniques class where each class is divided into two subclasses. We also provide an overview about fifteen different requirements prioritization techniques that are classified under our proposed classification. Moreover, we make a comparison between methods that are related to the same subclass to analyze their strengths and weakness. Based on the comparison results, the properties for each proposed subclass of techniques are identified. Depending on these properties, we present some recommendations to help project managers in the process of selection the most suitable technique to prioritize requirements based on their projects characteristics. © 2023, The Authors. All rights reserved.

关键词： Requirements engineering

Investigation of the Base Oil Properties in Optimizing the Demulsification 16

学校读者我要写书评

暂无评论

Investigation of the Base Oil Properties in Optimizing the D...

16th International engineering and Computing Research Conference, EURECA 2021

作者： Low, C.M. Ilia Anisa, A.N. Chua, B.L. Abdurahman, N.H. Chemical Engineering Program School of Computer Science and Engineering Taylor's University Lake Campus No. 1 Jalan Taylor's Subang Jaya47500 Malaysia Faculty of Chemical & Natural Resources Engineering Universiti Malaysia Pahang Lebuhraya Tun Razak Gambang Pahang26300 Malaysia

In this research, the correlation of the physical properties from different types of base oil- Group I and Group II base oils - to the demulsification process with variables such as mixing speed (500-1500 rpm), water content (20-80%) and volume of formulated surfactant (1 - 5 ml) were investigated. The water-in-oil emulsion was prepared using a formulated surfactant, and the demulsification was observed via a Bottle Test method. The optimization of demulsification was obtained from the correlation of the respective variables using the Central Composite Design (CCD) for Response Surface Methodology (RSM). Results showed that the Group I base oil emulsions have a lower viscosity and higher density than the Group II base oil, which has higher emulsion stability. The higher demulsification was observed by reducing the mixing speed, increasing the water content, and reducing the volume of the formulated surfactant for both groups of the base oil. The optimum of the water separation w obtained from the Group I base oil was 46.73% at 515 rpm, 76% water content, and 1 ml volume of the formulated surfactant. Meanwhile, 99.29% of water separation was found for the Group II base oil at 520 rpm, with 75% water content and 2 ml volume of the formulated surfactant. © Published under licence by IOP Publishing Ltd.

关键词： Demulsification

IA-TIGRIS: An Incremental and Adaptive Sampling-Based Planner for Online Informative Path Planning

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Moon, Brady Suvarna, Nayana Jong, Andrew Chatterjee, Satrajit Yuan, Junbin Scherer, Sebastian Robotics Institute School of Computer Science Carnegie Mellon University PittsburghPA United States GRASP Lab the University of Pennsylvania PhiladelphiaPA United States Mechanical Engineering Department Carnegie Mellon University PittsburghPA United States

Planning paths that maximize information gain for robotic platforms has wide-ranging applications and significant potential impact. To effectively adapt to real-time data collection, informative path planning must be computed online and be responsive to new observations. In this work, we present IA-TIGRIS, an incremental and adaptive sampling-based informative path planner that can be run efficiently with onboard computation. Our approach leverages past planning efforts through incremental refinement while continuously adapting to updated world beliefs. We additionally present detailed implementation and optimization insights to facilitate real-world deployment, along with an array of reward functions tailored to specific missions and behaviors. Extensive simulation results demonstrate IA-TIGRIS generates higher-quality paths compared to baseline methods. We validate our planner on two distinct hardware platforms: a hexarotor UAV and a fixed-wing UAV, each having unique motion models and configuration spaces. Our results show up to a 41% improvement in information gain compared to baseline methods, suggesting significant potential for deployment in real-world applications. Project website and video at ***. Copyright © 2025, The Authors. All rights reserved.

关键词： Robot programming

Multilevel Particle Filters for Partially Observed McKean-Vlasov Stochastic Differential Equations

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Awadelkarim, Elsiddig Jasra, Ajay Applied Mathematics and Computational Science Program Computer Electrical and Mathematical Sciences and Engineering Division King Abdullah University of Science and Technology Thuwal23955-6900 Saudi Arabia School of Data Science The Chinese University of Hong Kong Shenzhen China

In this paper we consider the filtering problem associated to partially observed McKean-Vlasov stochastic differential equations (SDEs). The model consists of data that are observed at regular and discrete times and the objective is to compute the conditional expectation of (functionals) of the solutions of the SDE at the current time. This problem, even the ordinary SDE case is challenging and requires numerical approximations. Based upon the ideas in [3, 12] we develop a new particle filter (PF) and multilevel particle filter (MLPF) to approximate the afore-mentioned expectations. We prove under assumptions that, for ϵ > 0, to obtain a mean square error of O(ϵ2) the PF has a cost per-observation time of O(ϵ−5) and the MLPF costs O(ϵ−4) (best case) or O(ϵ−4 log(ϵ)2) (worst case). Our theoretical results are supported by numerical experiments. © 2024, CC BY.

关键词： Mean square error

Efficient Diffusion-Driven Corruption Editor for Test-Time Adaptation

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Oh, Yeongtak Lee, Jonghyun Choi, Jooyoung Jung, Dahuin Hwang, Uiwon Yoon, Sungroh Department of Electrical and Computer Engineering Seoul National University Korea Republic of Interdisciplinary Program in Artificial Intelligence Seoul National University Korea Republic of School of Computer Science and Engineering Soongsil University Korea Republic of Division of Digital Healthcare Yonsei University Korea Republic of

Test-time adaptation (TTA) addresses the unforeseen distribution shifts occurring during test time. In TTA, performance, memory consumption, and time consumption are crucial considerations. A recent diffusion-based TTA approach for restoring corrupted images involves image-level updates. However, using pixel space diffusion significantly increases resource requirements compared to conventional model updating TTA approaches, revealing limitations as a TTA method. To address this, we propose a novel TTA method that leverages an image editing model based on a latent diffusion model (LDM) and fine-tunes it using our newly introduced corruption modeling scheme. This scheme enhances the robustness of the diffusion model against distribution shifts by creating (clean, corrupted) image pairs and fine-tuning the model to edit corrupted images into clean ones. Moreover, we introduce a distilled variant to accelerate the model for corruption editing using only 4 network function evaluations (NFEs). We extensively validated our method across various architectures and datasets including image and video domains. Our model achieves the best performance with a 100 times faster runtime than that of a diffusion-based baseline. Furthermore, it is three times faster than the previous model updating TTA method that utilizes data augmentation, making an image-level updating approach more feasible. ∗ © 2024, CC BY.

关键词：

Enhancing Precision Agriculture Through Human-in-the-Loop Planning and Control

学校读者我要写书评

暂无评论

Enhancing Precision Agriculture Through Human-in-the-Loop Pl...

IEEE International Conference on Automation science and engineering (CASE)

作者： Shankar A. Deka Sujet Phodapol Andreu Matoses Gimenez Victor Nan Fernandez-Ayala Rufus Wong Pian Yu Xiao Tan Dimos V. Dimarogonas Department of Electrical Engineering and Automation School of Electrical Engineering Aalto University Espoo Finland Division of Decision and Control Systems School of EECS Royal Institute of Technology (KTH) Stockholm Sweden Department of Cognitive Robotics Faculty of Mechanical Engineering TU Delft Delft Netherlands Department of Computer Science University of Oxford Oxford UK

ISBN: (数字)9798350358513

ISBN: (纸本)9798350358520

In this paper, we introduce a ROS based framework designed for the planning and control of robotic systems within the context of precision agriculture, with an emphasis on human-in-the-loop capabilities. Utilizing Linear Temporal Logic to articulate complex task specifications, our algorithm creates high-level robotic plans that are not only correct by design but also adaptable in real time by human operators. This dual-focus approach ensures that while humans have the flexibility to modify the high-level plan on-the-fly or even take over low-level control of the robots, the system inherently safeguards against any human actions that could potentially breach the predefined task specifications. We demonstrate our algorithm within the dynamic and challenging environment of a real vineyard, where the collaboration between human workers and robots is critical for tasks such as harvesting and pruning, and show the practical applicability and robustness of our software. This work marks a pioneering application of formal methods to complex, real-world agricultural environments.

关键词： Precision agriculture Heuristic algorithms Software algorithms Robustness Human in the loop Software Real-time systems Planning Robots Multi-agent systems

Federated Learning While Providing Model as a Service: Joint Training and Inference Optimization

学校读者我要写书评

暂无评论

Federated Learning While Providing Model as a Service: Joint...

IEEE Annual Joint Conference: INFOCOM, IEEE computer and Communications Societies

作者： Pengchao Han Shiqiang Wang Yang Jiao Jianwei Huang School of Information Engineering Guangdong University of Technology Guangzhou China IBM T. J. Watson Research Center NY USA Department of Computer Science and Technology Tongji University Shanghai China School of Science and Engineering Shenzhen Institute of Artificial Intelligence and Robotics for Society The Chinese University of Hong Kong Shenzhen Shenzhen China

ISBN: (数字)9798350383508

ISBN: (纸本)9798350383515

While providing machine learning model as a service to process users’ inference requests, online applications can periodically upgrade the model utilizing newly collected data. Federated learning (FL) is beneficial for enabling the training of models across distributed clients while keeping the data locally. However, existing work has overlooked the coexistence of model training and inference under clients’ limited resources. This paper focuses on the joint optimization of model training and inference to maximize inference performance at clients. Such an optimization faces several challenges. The first challenge is to characterize the clients’ inference performance when clients may partially participate in FL. To resolve this challenge, we introduce a new notion of age of model (AoM) to quantify client-side model freshness, based on which we use FL’s global model convergence error as an approximate measure of inference performance. The second challenge is the tight coupling among clients’ decisions, including participation probability in FL, model download probability, and service rates. Toward the challenges, we propose an online problem approximation to reduce the problem complexity and optimize the resources to balance the needs of model training and inference. Experimental results demonstrate that the proposed algorithm improves the average inference accuracy by up to 12%.

关键词： Training Adaptation models Accuracy Federated learning Computational modeling Approximation algorithms Data models