检索结果-内蒙古大学图书馆

2023 China Automation Congress, CAC 2023

作者： Deng, Siyi Xia, Yuchen Zuo, Weiliang Xin, Jingmin Zhou, Sanping Zheng, Nanning Institute of Artificial Intelligence and Robotics Xi'An Jiaotong University National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center for Visual Information and Applications Xi'an China

ISBN: (纸本)9798350303759

Parking space recognition technology is of great significance for intelligent parking lot management. To address the shortcomings of current parking lot management methods, we conducted research and practice on parking space recognition based on image processing and deep learning methods, resulting in a comprehensive and effective parking space recognition method. By preprocessing the parking lot images to divide the parking area and using the VGG16 neural network for parking space recognition training, we processed the parking lot video with the trained model. The experimental results show that most of the parking spaces in the video can be accurately identified, with an accuracy rate of 98.32%. The link to the parking space recognition video is: https://***/9xMph9USgW4. © 2023 IEEE.

关键词： Image processing

来源：评论

学校读者我要写书评

暂无评论

Exploiting Cross-Modal Cost Volume for Multi-sensor Depth Estimation 17th

Exploiting Cross-Modal Cost Volume for Multi-sensor Depth E...

引用

17th Asian Conference on Computer Vision, ACCV 2024

作者： Kim, Janghyun Shin, Ukcheol Heo, Seokyong Park, Jinsun Department of Information Convergence Engineering Pusan National University Busan Korea Republic of Robotics Institute Carnegie Mellon University Pittsburgh United States School of Computer Science and Engineering Pusan National University Busan Korea Republic of

ISBN: (纸本)9789819609680

Single-modal depth estimation has shown steady improvement over the years. However, relying solely on a single imaging sensor such as RGB and near-infrared (NIR) cameras can result in unreliable and erroneous depth estimation, particularly in challenging lighting conditions such as low-light or sudden lighting change scenarios. Thereby, several approaches have leveraged multiple sensors for robust depth estimation. However, the effective fusion method that maximally utilizes multi-modal sensor information still requires further investigation. With this in mind, we propose a multi-modal cost volume fusion strategy with cross-modal attention, incorporating information from both cross-spectral and single-modality pairs. Our method initially constructs low-level cost volumes that consist of modality-specific (ı.e., single modality) and modality-invariant (ı.e., cross-spectral) volumes from multi-modal sensors. These cost volumes are then gradually fused using bidirectional cross-modal fusion and unidirectional LiDAR fusion to generate a multi-sensory cost volume. Furthermore, we introduce a straightforward domain gap reduction approach to learn modality-invariant features and depth refinement techniques through cost volume-guided propagation. Experimental results demonstrate that our method achieves SOTA (State-of-the-Art) performance under diverse environmental changes. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Depth perception

来源：评论

学校读者我要写书评

暂无评论

SUNOD: Synthetic Underwater Non-Natural Object Detection Dataset

SUNOD: Synthetic Underwater Non-Natural Object Detection Dat...

引用

2023 China Automation Congress, CAC 2023

作者： Xia, Yuchen Deng, Siyi Li, Yuxi Zuo, Weiliang Xin, Jingmin Zheng, Nanning Institute of Artificial Intelligence and Robotics Xi'an Jiaotong University National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center for Visual Information and Applications Xi'an China

ISBN: (纸本)9798350303759

Deep-learning-based non-natural object detection in complex underwater environments is widely needed in many application scenarios. However, obtaining image data of underwater non-natural objects is difficult, so there is basically no relevant dataset. Besides, the underwater environments and lighting conditions are complex, which will lead to image quality degradation and increase the difficulty of detecting. For this purpose, we build a Synthetic Underwater Non-natural Object Detection (SUNOD) dataset. We first utilize a free software, Unreal Engine 5 (UE5), to create a realistic three-dimensional virtual underwater environment containing 12 kinds of non-natural objects. Then we take pictures of these non-natural objects and manually label 2,840 images. To make SUNOD more challenging and be able to create object detection models with more generalization capabilities and better robustness, we expand it to contain 72,943 images by 26 kinds of data augmentation methods. In addition, we report the performance of 5 YOLOv8 object detection models with different volumes on SUNOD. SUNOD can be used both to fine-tune pretrained models to accomplish the underwater non-natural object detection task and to be a benchmark to evaluate new models for this task. This study provides the first synthetic dataset for underwater non-natural object detection, which contributes to the research on the intelligence of submarines, exploration robots, and other underwater devices. © 2023 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Can LLMs Learn From Mistakes? An Empirical Study on Reasoning Tasks

Can LLMs Learn From Mistakes? An Empirical Study on Reasonin...

引用

2024 Conference on Empirical Methods in Natural Language Processing, EMNLP 2024

作者： An, Shengnan Ma, Zexiong Cai, Siqi Lin, Zeqi Zheng, Nanning Lou, Jian-Guang Chen, Weizhu National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center of Visual Information and Applications Institute of Artificial Intelligence and Robotics Xi'an Jiaotong University China Microsoft United States Peking University China

ISBN: (纸本)9798891761681

Towards enhancing the chain-of-thought (CoT) reasoning of large language models (LLMs), much existing work has revealed the effectiveness of straightforward learning on annotated/generated CoT paths. However, there is less evidence yet that reasoning capabilities can be enhanced through a reverse learning process, i.e., learning from potential mistakes in reasoning. To investigate whether LLMs can learn from mistakes, we construct mistake-correction datasets, using GPT-4 to identify and correct the mistakes in inaccurate CoTs. With these mistake-correction datasets, we fine-tune open-source LLMs and arrive at the following conclusions. (1) LLMs can indeed learn from mistakes to enhance their CoT reasoning performances. (2) Compared to CoT data, the mistake-correction data provides additional knowledge on the explanations and reasons for the potential mistakes in CoTs, which consistently contributes to the effectiveness of learning from mistakes. (3) Evolution techniques, especially the correction-centric evolution we introduced, can further enhance the effectiveness of learning from mistakes. © 2024 Association for Computational Linguistics.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

The Accuracy Analysis of Tool center Point Calibration via Condition Number and Minimum Eigenvalue

The Accuracy Analysis of Tool Center Point Calibration via C...

引用

2023 IEEE International Conference on Real-Time Computing and robotics, RCAR 2023

作者： Gao, Ziqi Li, Yingli Luo, Yang Jiang, Yunxiang Zhang, Bi Zhao, Xingang Chen, Lingkai Zhao, Yiwen Institutes for Robotics and Intelligent Manufacturing Chinese Academy of Sciences Shenyang110169 China University of Chinese Academy of Sciences Beijing100049 China Shenyang Institute of Automation Chinese Academy of Sciences State Key Laboratory of Robotics Shenyang110016 China Research Institute of Interdisciplinary Innovation Zhejiang Lab Research Center for Intelligent Robotics Hangzhou311100 China Zhejiang Engineering Research Center for Intelligent Robotics Hangzhou311100 China

ISBN: (纸本)9798350327182

The accuracy of TCP (tool center point) calibration is vital to the operation accuracy of the robot. In this paper, we analyzed the accuracy analysis of TCP calibration of the regression matrix via condition number and minimum eigenvalue. It is theoretically proved that a regression matrix with a large condition number and a small eigenvalue will aggravate the influence of the observation error of the measurement point on TCP calibration accuracy and increase the TCP calibration error. To verify our analysis, simulation calibration experiments were carried out and statistical analysis was carried out on the calibration results. The result shows that the mathematical expectation and the variance of calibration error at minimum, median, and maximum matrix CA condition number are 1.36mm (±0.60mm), 1.57mm (± 0.75mm) and 2.94mm (±2.26mm) respectively, and the mathematical expectation and the variance of calibration error at minimum, median, and maximum eigenvalue are 3.33mm (± 2.32mm), 1.54mm (±0.72mm) and 1.16mm (± 0.49mm), respectively. © 2023 IEEE.

关键词： Calibration

来源：评论

学校读者我要写书评

暂无评论

Map Completion for SLAM Systems Based on Neural Radiance Field

Map Completion for SLAM Systems Based on Neural Radiance Fie...

引用

2023 China Automation Congress, CAC 2023

作者： Deng, Siyi Zuo, Weiliang Xia, Yuchen Xin, Jingmin Zhou, Sanping Zheng, Nanning Institute of Artificial Intelligence and Robotics Xi' An Jiaotong University National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center for Visual Information and Applications Xi'an China

ISBN: (纸本)9798350303759

Simultaneous localization and mapping (SLAM) systems aim to enhance the capabilities of robots in complex environments by building maps of unknown surroundings. Creating a dense and detailed map can significantly improve the robot's perception of the environment and enable it to perform more challenging tasks. However, traditional algorithms often struggle to balance localization, tracking, and mapping. Recently, neural radiance fields-based 3D reconstruction has emerged as an efficient and accurate method, offering a potential solution for map completion in SLAM systems. In our work, we integrate neural radiance fields with the SLAM system, providing a real-time solution for tracking, localization, and initial scene reconstruction, along with offline map completion. Experimental results demonstrate that our approach sacrifices some tracking accuracy to achieve complete map construction. © 2023 IEEE.

关键词： Mapping

来源：评论

学校读者我要写书评

暂无评论

Disturbance rejection for biped robots during walking and running using control moment gyroscopes

引用

IET Cyber-Systems and robotics 2022年第4期4卷 268-282页

作者： Haochen Xu Zhangguo Yu Xuechao Chen Chencheng Dong Huanzhong Chen Qiang Huang School of Mechatronical Engineering Beijing Institute of TechnologyBeijingChina Beijing Advanced Innovation Center for Intelligent Robotics and Systems Beijing Institute of TechnologyBeijingChina

Keeping balance in movement is an important premise for biped robots to complete various ***,the balance control of biped robots mainly depends on the cooperation of various joints of the robot's *** robots move faster,the adjustment allowance of joints is reduced,and the robot's anti-disturbance ability will inevitably *** solve this problem,the control moment gyroscope(CMG)is creatively used as an auxiliary stabilisation device for fully actuated biped robots and the CMG assistance strategy,which can be integrated into the biped's balance control framework,is *** strategy includes model predictive control module,distribution module,and CMG precession *** the command of it,CMGs can effectively assist the robot in resisting impact and returning to initial positions in *** results of anti-impact simulation on the walking and running biped robot prove that,with the help of CMGs,the robot's ability to resist disturbance and remain stable is significantly *** cover image is based on the Original Article Disturbance rejection for biped robots during walking and running using control moment gyroscopes by Haochen Xu et al.,https://***/10.1049/csy2.12070.

关键词： anti-impact biped robot control moment gyroscope model predictive control

来源：评论

学校读者我要写书评

暂无评论

Improved Capon Estimator for High-Resolution DOA Estimation and Its Statistical Analysis

引用

IEEE/CAA Journal of Automatica Sinica 2023年第8期10卷 1716-1729页

作者： Weiliang Zuo Jingmin Xin Changnong Liu Nanning Zheng Akira Sano IEEE the National Key Laboratory of Human-Machine Hybrid Augmented Intelligence the National Engineering Research Center for Visual Information and Applicationsand the Institute of Artificial Intelligence and RoboticsXi’an Jiaotong UniversityXi’an 710049China the Data Center China Construction BankBeijing 100033China the Department of System Design Engineering Keio UniversityYokohama 223-8522Japan

Despite some efforts and attempts have been made to improve the direction-of-arrival(DOA)estimation performance of the standard Capon beamformer(SCB)in array processing,rigorous statistical performance analyses of these modified Capon estimators are still *** paper studies an improved Capon estimator(ICE)for estimating the DOAs of multiple uncorrelated narrowband signals,where the higherorder inverse(sample)array covariance matrix is used in the Capon-like cost *** establishing the relationship between this nonparametric estimator and the parametric and classic subspace-based MUSIC(multiple signal classification),it is clarified that as long as the power order of the inverse covariance matrix is increased to reduce the influence of signal subspace components in the ICE,the estimation performance of the ICE becomes equivalent to that of the MUSIC regardless of the signal-to-noise ratio(SNR).Furthermore the statistical performance of the ICE is analyzed,and the large-sample mean-squared-error(MSE)expression of the estimated DOA is *** the effectiveness and the theoretical analysis of the ICE are substantiated through numerical examples,where the Cramer-Rao lower bound(CRB)is used to evaluate the validity of the derived asymptotic MSE expression.

关键词： Capon beamformer direction-of-arrival(DOA)estimation large-sample mean-squared-error(MSE) subspace-based methods uniform linear array

来源：评论

学校读者我要写书评

暂无评论

Breaking through the Learning Plateaus of In-context Learning in Transformer 41

Breaking through the Learning Plateaus of In-context Learnin...

引用

41st International Conference on Machine Learning, ICML 2024

作者： Fu, Jingwen Yang, Tao Wang, Yuwang Lu, Yan Zheng, Nanning National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center for Visual Information and Applications Institute of Artificial Intelligence and Robotics Xi'an Jiaotong University China Tsinghua University China Microsoft Research Asia China

In-context learning, i.e., learning from context examples, is an impressive ability of Transformer. Training Transformers to possess this in-context learning skill is computationally intensive due to the occurrence of learning plateaus, which are periods within the training process where there is minimal or no enhancement in the model's in-context learning capability. To study the mechanism behind the learning plateaus, we conceptually separate a component within the model's internal representation that is exclusively affected by the model's weights. We call this the "weights component", and the remainder is identified as the "context component". By conducting meticulous and controlled experiments on synthetic tasks, we note that the persistence of learning plateaus correlates with compromised functionality of the weights component. Recognizing the impaired performance of the weights component as a fundamental behavior that drives learning plateaus, we have developed three strategies to expedite the learning of Transformers. The effectiveness of these strategies is further confirmed in natural language processing tasks. In conclusion, our research demonstrates the feasibility of cultivating a powerful in-context learning ability within AI systems in an eco-friendly manner. Copyright 2024 by the author(s)

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Applying Modified ZMP Gait Planning on Biped Robot Walking 42

Applying Modified ZMP Gait Planning on Biped Robot Walking

引用

42nd Chinese Control Conference, CCC 2023

作者： Zhiyong, Jiang Hongrun, Lu Chunlei, Wang Dongxiao, Ma The 21st Research Institute of China Electronics Technology Group Corporation Robotics Engineering Center Shanghai200233 China

ISBN: (纸本)9789887581543

In this paper, A biped walking algorithm based on modified ZMP gait planning has been introduced. In conventional gait planning, some transient changes appear in the robot's coronal plane. One efficient method for this problem is adding a planning point in the ZMP gait. In addition, the sagittal and coronal movements are planned to move simultaneously. Human-moving postures is utilized to optimize the robot's gait. Besides, the quintic polynomial interpolation is applied in gait planning to eliminate the transient changes. Finally, the trajectories of the joint angles are obtained by the particle swarm optimization algorithm, which would be implemented on the biped robot. The test results show that the proposed modified gait planning can effectively improve the naturalness and stability of biped walking. © 2023 Technical Committee on Control Theory, Chinese Association of Automation.

关键词： Robot programming

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：