检索结果-内蒙古大学图书馆

Image Processing Challenges (ICIPC), IEEE International Conference on

作者： Vinayak Nageli Arun Kumar S Rama Krishna Sai S Gorthi Arshad Jamal Indian Institute of Technology (IIT) Tirupati Centre for Artificial Intelligence and Robotics (CAIR) Bangalore

This work proposes a solution for detection and tracking of drones in distorted Infrared (IR) images. Our proposed solution combines stats-of-the-art object detection and tracking techniques with distortion classification technique to achieve robust performance in highly distorted videos. There are two main steps in our proposed solution. First, an online drones detection using state-of-the-art YOLOv8 object detection framework is combined with Distortion Classification network. Second, we use the detection proposals from YOLOv8 for temporal association to track the drones using popular state-of-the-art trackers namely, ByteTrack and Strong-SORT. Our solution was evaluated using the ICIP-2023 challenge dataset for drones detection and tracking. The ICIP-2023 challenge dataset consists of IR images and videos with challenging attributes likes different types of distortions, varying size of drones and abrupt drone motion etc. with complex scenes. Our solution works in all the challenging scenario captured in the dataset and our submission achieved the best performance among all entries in the challenge, setting a benchmark for the ICIP-2023 challenge dataset with detection Accuracy of 95% and mAP of 98% for drones and Tracking HOTA of 0.68 and IDF1 0.77 with real-time operation at 35FPS.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Fabrication of Sensorsless Automated Guided Vehicle for Small Warehouse Rack Management

Fabrication of Sensorsless Automated Guided Vehicle for Smal...

引用

2025 IEEE International Conference on Computational, Communication and Information Technology, ICCCIT 2025

作者： Sahu, Neelesh Kumar Chaturvedi, Piyush Gupta, Vaishali Patel, Ruchi Verma, Ashok Shrivastava, Megha Gyan Ganga Institute of Technology and Sciences Artificial Intelligence and Robotics MP Jabalpur India IPS Academy Institute of Engineering & Science Department of Computer Sc. & Engg. MP Indore India Gyan Ganga Institute of Technology and Sciences MP Jabalpur India Medicaps University Department of Mathematics MP Indore India

ISBN: (纸本)9798331512965

Automated guided vehicles (AGVs) are an essential part of today's logistics networks because they save time, reduce wear and maintenance expenses, and maximize efficiency in route *** creation of AGVthat is capable of effectively managing warehouse racks without the utilization of sensors is the primary emphasis of this project. In order to travel and carry out duties, the AGVruns on predetermined pathways and makes use of a button-based interface. For the purpose of offering an efficient and costeffective solution for the automation of small warehouses, it is outfitted with a lever mechanism that is operated by a servo motor and features the ability to connect and move racks. © 2025 IEEE.

关键词： Automatic guided vehicles

来源：评论

学校读者我要写书评

暂无评论

Enhancing Exploration with Diffusion Policies in Hybrid Off-Policy RL: Application to Non-Prehensile Manipulation

arXiv

引用

arXiv 2024年

作者： Le, Huy Gabriel, Miroslav Hoang, Tai Neumann, Gerhard Vien, Ngo Anh Institute for Anthropomatics and Robotics Karlsruhe Institute of Technology Karlsruhe76131 Germany Bosch Center for Artificial Intelligence Renningen71272 Germany

Learning diverse policies for non-prehensile manipulation is essential for improving skill transfer and generalization to out-of-distribution scenarios. In this work, we enhance exploration through a two-fold approach within a hybrid framework that tackles both discrete and continuous action spaces. First, we model the continuous motion parameter policy as a diffusion model, and second, we incorporate this into a maximum entropy reinforcement learning framework that unifies both the discrete and continuous components. The discrete action space, such as contact point selection, is optimized through Q-value function maximization, while the continuous part is guided by a diffusion-based policy. This hybrid approach leads to a principled objective, where the maximum entropy term is derived as a lower bound using structured variational inference. We propose the Hybrid Diffusion Policy algorithm (HyDo) and evaluate its performance on both simulation and zero-shot sim2real tasks. Our results show that HyDo encourages more diverse behavior policies, leading to significantly improved success rates across tasks - for example, increasing from 53% to 72% on a real-world 6D pose alignment task. Project page: https://***/hydo © 2024, CC BY-NC-ND.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

DEALING WITH NON-STATIONARITY IN MARL VIA TRUST-REGION DECOMPOSITION 10

DEALING WITH NON-STATIONARITY IN MARL VIA TRUST-REGION DECOM...

引用

10th International Conference on Learning Representations, ICLR 2022

作者： Li, Wenhao Wang, Xiangfeng Jin, Bo Sheng, Junjie Zha, Hongyuan School of Computer Science and Technology East China Normal University Shanghai China Shenzhen Institute of Artificial Intelligence and Robotics for Society Shenzhen China

Non-stationarity is one thorny issue in cooperative multi-agent reinforcement learning (MARL). One of the reasons is the policy changes of agents during the learning process. Some existing works have discussed various consequences caused by non-stationarity with several kinds of measurement indicators. This makes the objectives or goals of existing algorithms are inevitably inconsistent and disparate. In this paper, we introduce a novel notion, the δ-stationarity measurement, to explicitly measure the non-stationarity of a policy sequence, which can be further proved to be bounded by the KL-divergence of consecutive joint policies. A straightforward but highly non-trivial way is to control the joint policies' divergence, which is difficult to estimate accurately by imposing the trust-region constraint on the joint policy. Although it has lower computational complexity to decompose the joint policy and impose trust-region constraints on the factorized policies, simple policy factorization like mean-field approximation will lead to more considerable policy divergence, which can be considered as the trust-region decomposition dilemma. We model the joint policy as a pairwise Markov random field and propose a trust-region decomposition network (TRD-Net) based on message passing to estimate the joint policy divergence more accurately. The Multi-Agent Mirror descent policy algorithm with Trust region decomposition, called MAMT, is established by adjusting the trust-region of the local policies adaptively in an end-to-end manner. MAMT can approximately constrain the consecutive joint policies' divergence to satisfy δ-stationarity and alleviate the non-stationarity problem. Our method can bring noticeable and stable performance improvement compared with baselines in cooperative tasks of different complexity. © 2022 ICLR 2022 - 10th International Conference on Learning Representationss. All rights reserved.

关键词： Markov processes

来源：评论

学校读者我要写书评

暂无评论

MSNNet: Multi-scale-abled networks with multi-groups channel attention for vibrational spectrum matching

引用

Vibrational Spectroscopy 2023年 129卷

作者： Fan, Youpeng Fang, Yongchun The Institute of Robotics and Automatic Information System College of Artificial Intelligence Nankai Uni versity Tianjin300350 China

In recent years, deep learning methods have exhibited superior performance in many applications in vibrational spectroscopy especially compared with conventional machine learning methods such as Support Vector Machine (SVM) and Partial Least Squares (PLS). On the other hand, almost all of these methods employ such as LeNet and ResNet as their backbones which are not designed for processing spectra originally. In this paper, we demonstrate that these widely-used neural network models are insufficient when dealing with vibrational spectra. Particularly they lack the ability to process multi-scale information hidden in vibrational spectra. To address this, we further propose a novel multi-scale-abled network named MSNNet which combines a modified Res2Net with the Multi-Groups Channel Attention (MGCA). Experiments on two publicly available spectrum datasets show that the proposed MSNNet significantly outperforms existing methods. Moreover, using the Lorentz model we also generat a number of synthetic datasets where the scale in spectra can be deliberately controlled. Results on these synthetic datasets not only validate the superiority of the proposed method, but also support our hypothesis that the scale information can play a vital role in recognition of vibrational spectra. © 2023 Elsevier B.V.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Vision Mamba: Efficient Visual Representation Learning with Bidirectional State Space Model

arXiv

引用

arXiv 2024年

作者： Zhu, Lianghui Liao, Bencheng Zhang, Qian Wang, Xinlong Liu, Wenyu Wang, Xinggang School of EIC Huazhong University of Science & Technology China Institute of Artificial Intelligence Huazhong University of Science & Technology China Horizon Robotics China Beijing Academy of Artificial Intelligence China

Recently the state space models (SSMs) with efficient hardware-aware designs, i.e., the Mamba deep learning model, have shown great potential for long sequence modeling. Meanwhile building efficient and generic vision backbones purely upon SSMs is an appealing direction. However, representing visual data is challenging for SSMs due to the position-sensitivity of visual data and the requirement of global context for visual understanding. In this paper, we show that the reliance on self-attention for visual representation learning is not necessary and propose a new generic vision backbone with bidirectional Mamba blocks (Vim), which marks the image sequences with position embeddings and compresses the visual representation with bidirectional state space models. On ImageNet classification, COCO object detection, and ADE20k semantic segmentation tasks, Vim achieves higher performance compared to well-established vision transformers like DeiT, while also demonstrating significantly improved computation & memory efficiency. For example, Vim is 2.8× faster than DeiT and saves 86.8% GPU memory when performing batch inference to extract features on images with a resolution of 1248×1248. The results demonstrate that Vim is capable of overcoming the computation & memory constraints on performing Transformer-style understanding for high-resolution images and it has great potential to be the next-generation backbone for vision foundation models. Code and models are released at https://***/hustvl/Vim 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

Video Self-Supervised Cross-Pathway Training Based on Slow and Fast Pathways

Video Self-Supervised Cross-Pathway Training Based on Slow a...

引用

2023 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2023

作者： Li, Jie Yang, Jing Jiang, Zhou Chen, Yuehai Du, Shaoyi School of Automation Science and Engineering Xi'an Jiaotong University Xi'an China School of Software Engineering Xi'an Jiaotong University Xi'an China Institute of Artificial Intelligence and Robotics Xi'an Jiaotong University Xi'an China

ISBN: (纸本)9798350337020

In the field of video self-supervised learning, contrastive instance learning methods suffer from a lack of semantic information, resulting in inadequate generalization in downstream tasks. Although optical flow can provide some semantic information, it requires significant computational cost prior to training. To address this, we propose a Video self-supervised Cross-pathway training model based on Slow and Fast pathways (VCSF). This model separately extracts temporal and spatial features from pure RGB video frames, and uses the complementary representations of the two pathways to conduct cross-pathway training. Additionally, we propose a motion perception module in the low-frame-rate space to enhance the network's ability to perceive rapidly changing human motion. We conducted extensive experiments in downstream missions of UCF101 and HMDB51, and obtained state-of-the-art results in models using the UCF101 data set for self-supervised pre-training, including motion recognition and nearest neighbor retrieval. © 2023 IEEE.

关键词： Cross-pathway training Self-supervised learning Slow and Fast pathways

来源：评论

学校读者我要写书评

暂无评论

PM_(2.5) probabilistic forecasting system based on graph generative network with graph U-nets architecture

引用

Journal of Central South University 2025年第1期32卷 304-318页

作者： LI Yan-fei YANG Rui DUAN Zhu LIU Hui School of Mechatronic Engineering Hunan Agricultural UniversityChangsha 410128China Institute of Artificial Intelligence and Robotics(IAIR) Key Laboratory of Traffic Safety on Track of Ministry of EducationSchool of Traffic and Transportation EngineeringCentral South UniversityChangsha 410075China

Urban air pollution has brought great troubles to physical and mental health,economic development,environmental protection,and other *** the changes and trends of air pollution can provide a scientific basis for governance and prevention *** this paper,we propose an interval prediction method that considers the spatio-temporal characteristic information of PM_(2.5)signals from multiple stations.K-nearest neighbor(KNN)algorithm interpolates the lost signals in the process of collection,transmission,and storage to ensure the continuity of *** generative network(GGN)is used to process time-series meteorological data with complex *** graph U-Nets framework is introduced into the GGN model to enhance its controllability to the graph generation process,which is beneficial to improve the efficiency and robustness of the *** addition,sparse Bayesian regression is incorporated to improve the dimensional disaster defect of traditional kernel density estimation(KDE)interval *** the support of sparse strategy,sparse Bayesian regression kernel density estimation(SBR-KDE)is very efficient in processing high-dimensional large-scale *** PM_(2.5)data of spring,summer,autumn,and winter from 34 air quality monitoring sites in Beijing verified the accuracy,generalization,and superiority of the proposed model in interval prediction.

关键词： PM_(2.5)interval forecasting graph generative network graph U-Nets sparse Bayesian regression kernel density estimation spatial-temporal characteristics

来源：评论

学校读者我要写书评

暂无评论

InteractionNet: Joint Planning and Prediction for Autonomous Driving with Transformers

arXiv

引用

arXiv 2023年

作者： Fu, Jiawei Shen, Yanqing Jian, Zhiqiang Chen, Shitao Xin, Jingmin Zheng, Nanning The Institute of Artificial Intelligence and Robotics Xi'an Jiaotong University Shaanxi Xi’an710049 China

Planning and prediction are two important modules of autonomous driving and have experienced tremendous advancement recently. Nevertheless, most existing methods regard planning and prediction as independent and ignore the correlation between them, leading to the lack of consideration for interaction and dynamic changes of traffic scenarios. To address this challenge, we propose InteractionNet, which leverages transformer to share global contextual reasoning among all traffic participants to capture interaction and interconnect planning and prediction to achieve joint. Besides, InteractionNet deploys another transformer to help the model pay extra attention to the perceived region containing critical or unseen vehicles. InteractionNet outperforms other baselines in several benchmarks, especially in terms of safety, which benefits from the joint consideration of planning and forecasting. The code will be available at https://***/fujiawei0724/InteractionNet. Copyright © 2023, The Authors. All rights reserved.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

New Software Bionic Haptic Actuator Design Based on Barometric Array

New Software Bionic Haptic Actuator Design Based on Baromet...

引用

International Conference on Applied Nonlinear Dynamics, Vibration, and Control, ICANDVC 2023

作者： Yu, Zige Li, Sai Lin, Mengying Hu, Hang Li, Yingying Lei, Qian Huang, Zixin School of Electrical and Information Engineering Wuhan Institute of Technology Wuhan430205 China Hubei Key Laboratory of Digital Textile Equipment Wuhan Textile University Wuhan430200 China Institute of Robotics and Automatic Information Systems College of Artificial Intelligence Nankai University Tianjin300350 China

ISBN: (纸本)9789819705535

To address the limitations of single-point tactile perception in soft biomimetic actuators, such as fixed positioning, limited coverage area, low resolution at long distances from the geometric center, and challenges in maintaining high sensitivity and stability over a broad range, a novel soft biomimetic tactile actuator based on a pneumatic array is designed and manufactured. Using liquid silicone rubber and BMP280 pressure sensor, the actuator was crafted by an injection molding process. Silicone gel was injected into the pressure sensor array, and vacuum extraction created a sealed space within the sensor cavity. The pressure excitation applied on the surface of the actuator could be converted into electrical signals, enabling tactile pressure detection. An STM32 microprocessor is utilized for building haptic information acquisition and processing system, with sensitivity assessed using the Kalman filter algorithm. Experimental results show that the soft biomimetic tactile actuator has high sensitivity and minimal repetition error, which can effectively detects changes in tactile force output, mitigates noise interference, and enhances the precision and stability of physiological pressure data feedback in physiotherapy applications. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Bionics

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：