检索结果-内蒙古大学图书馆

IEEE TRANSACTIONS ON INTELLIGENT VEHICLES 2024年第1期9卷 2318-2332页

作者： Li, Dachuan Liu, Bowen Huang, Zijian Hao, Qi Zhao, Dezong Tian, Bin Southern Univ Sci & Technol Res Inst Trustworthy Autonomous Syst Shenzhen 518055 Peoples R China Southern Univ Sci & Technol Dept Comp Sci & Engn Shenzhen 518057 Peoples R China Univ Glasgow James Watt Sch Engn Glasgow G12 8QQ Lanark Scotland Chinese Acad Sci Inst Automat State Key Lab Management & Control Complex Syst Beijing 100190 Peoples R China Univ Chinese Acad Sci Sch Artificial Intelligence Beijing 100190 Peoples R China

Conventional perception-planning pipelines of autonomous vehicles (AV) utilize deep learning (DL) techniques that typically generate deterministic outputs without explicitly evaluating their uncertainties and trustworthiness. Therefore, the downstream decision-making components may generate unsafe outputs leading to system failure or accidents, if the preceding perception component provides highly uncertain information. To mitigate this issue, this article proposes a coherent safe perception-planning framework that quantifies and transfers DL-based perception uncertainties. Following the Bayesian Deep Learning paradigm, we design a probabilistic 3D object detector that extracts objects from LiDAR point clouds while quantifying the corresponding aleatoric and epistemic uncertainty. A chance-constrained motion planner is designed to formulate an explicit link between DL-based perception uncertainties and operation risk and generate safe and risk-bounding trajectories. The proposed framework is validated through various challenging scenarios in the CARLA simulator. Experiment results demonstrate that our framework can effectively capture the uncertainties in DL, and generate trajectories that bound the risk under DL perception uncertainties. It also outperforms counterpart approaches without explicitly evaluating the uncertainties of DL-based perception.

关键词： Uncertainty Planning Three-dimensional displays Decision making Detectors Trajectory Probabilistic logic Autonomous driving motion planning object detection Bayesian deep learning uncertainty quantification

来源：评论

学校读者我要写书评

暂无评论

Backstepping-based parallel control for cascaded nonlinear systems 1

Backstepping-based parallel control for cascaded nonlinear s...

引用

1st IEEE International Conference on Digital Twins and Parallel Intelligence, DTPI 2021

作者： Lu, Jingwei Wei, Qinglai Zhou, Tianmin Han, Liyuan Wang, Fei-Yue School of Artificial Intelligence University of Chinese Academy of Sciences Beijing China The State Key Laboratory for Management and Control of Complex Systems Institute of Automation Beijing China

ISBN: (纸本)9781665433372

In this study, a novel nonlinear parallel control method is proposed for cascaded nonlinear systems using the backstepping technique. Unlike the existing state feedback control methods, the control input is taken into the feedback system. First, an augmented system is constructed to facilitate the constructing the Lyapunov function. Then, the backstepping technique can be applied to obtain the nonlinear parallel control law, and the stability analysis is shown using the Lyapunov theory. Finally, a simulation is conducted to demonstrate the effectiveness of the proposed parallel control method. © 2021 IEEE.

关键词： Backstepping

来源：评论

学校读者我要写书评

暂无评论

Graph Attention Memory for Visual Navigation 4

Graph Attention Memory for Visual Navigation

引用

4th International Conference on Data-Driven Optimization of complex systems, DOCS 2022

作者： Li, Dong Zhang, Qichao Zhao, Dongbin Noah's Ark Lab Huawei Beijing China Institute of Automation Chinese Academy of Sciences State Key Laboratory of Management and Control for Complex System China University of Chinese Academy of Sciences School of Artificial Intelligence Beijing China

ISBN: (数字)9781665459822

ISBN: (纸本)9781665459822

Visual navigation in complex environments is inefficient with traditional reactive policy or general-purposed recurrent policy due to the long-term memory problem. To address this issue, this paper proposes a graph attention memory (GAM) architecture consisting of memory construction module, graph attention module, and control module. The memory construction module builds the topological graph based on supervised learning by taking the exploration prior. Then, guided attention features to reach the goal are extracted with the graph attention module. Finally, the deep reinforcement learning based control module makes decisions based on visual observations and guided attention features. In addition, the convergence of the proposed GAM module for recurrent attention operation is analyzed in this paper. We evaluate GAM-based navigation system in two complex 3D ViZDoom environments. Experimental results show that the GAM-based navigation system outperforms all baselines in both success rate and navigation efficiency, and significantly improves the generalization. © 2022 IEEE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Activity and Relationship Modeling Driven Weakly Supervised Object Detection 25

Activity and Relationship Modeling Driven Weakly Supervised ...

引用

25th International Conference on Pattern Recognition (ICPR)

作者： Li, Yinlin Qian, Yang Yang, Xu Zhang, Yuren Chinese Acad Sci Inst Automat State Key Lab Management & Control Complex Syst Beijing Peoples R China ByteDance Beijing Peoples R China

ISBN: (纸本)9781728188089

This paper presents a weakly supervised object detection method based on activity label and relationship modeling, which is motivated by the assumption that configuration of human and object are similar in same activity, and joint modeling of human, active object and activity could leverage the recognition of them. Compared to most weakly supervised method taking object as independent instance, firstly, active human and object proposals are learned and filtered based on class activation map of multi-label classification. Secondly, a spatial relationship prior including relative position, scale, overlaps etc are learned dependent on action. Finally, a multistream object detection framework integrating the spatial prior and pairwise ROI pooling are proposed to jointly learn the object and action class. Experiments are conducted on HICODET dataset, and our approach outperforms the state of the art weakly supervised object detection methods.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Toward Fair and Thrilling Autonomous Racing: Governance Rules and Performance Metrics for the Autonomous One

IEEE TRANSACTIONS ON INTELLIGENT VEHICLES

引用

IEEE TRANSACTIONS ON INTELLIGENT VEHICLES 2023年第8期8卷 3974-3982页

作者： Li, Bai Fang, Yile Xu, Tian'ao Ma, Siji Wang, Haorui Wang, Yazhou Li, Xinyuan Zhang, Tantan Bian, Xuepeng Wang, Fei-Yue State Key Lab Adv Design & Mfg Vehicle Body Changsha 410082 Peoples R China Hunan Univ Coll Mech & Vehicle Engn Changsha 410082 Peoples R China Swiss Fed Inst Technol Dept Mech & Proc Engn CH-8092 Zurich Switzerland Macau Univ Sci & Technol Fac Innovat Engn Macau 999078 Peoples R China Beijing Inst Fash Technol Coll Fash Accessory Art Engn Coll Beijing 100105 Peoples R China Tencent Com Inc Tencent Automat Drive Lab Beijing 100193 Peoples R China Chinese Acad Sci Inst Automat State Key Lab Management & Control Complex Syst Beijing 100190 Peoples R China

This letter is the second report from a series of IEEE TIV's Decentralized and Hybrid Workshops (DHWs) on Intelligent Vehicles for Education (IV4E). It outlines the prospect of The Autonomous One (TAO), a future autonomous racing competition modeled after Formula One to advance IV4E by pushing the boundaries of artificial intelligence. Existing autonomous races face challenges, including compromised fairness and low participant enthusiasm. These issues limit spectator engagement, thereby reducing the races' educational value. To prevent similar flaws in organizing TAO, our focus lies in setting governance rules from an organizer's perspective. In these DHWs, we analyzed the rules of existing autonomous races and suggested rule-making guidelines for TAO. To improve fairness, we recommend a balanced scoring system with rigorous monitoring of participating teams' performances. To discourage consistent victories by a single team, we suggest that leading teams share their source codes, thereby setting the championship-level performance as the baseline in the next season and accelerating the evolution of all participating teams' abilities. For enhancing suspense and spectator interest, an on-site adaptive rewarding scheme should be deployed to create thrilling turnarounds. Our strategies aim to maintain fairness, increase spectator interest, inspire competitiveness, and ultimately contribute to the advancement of IV4E in organizing TAO.

关键词： Autonomous racing intelligent vehicles for education (IV4E) the autonomous one (TAO)

来源：评论

学校读者我要写书评

暂无评论

Parallel control-Based Event-Triggered Optimal control for Constrained Discrete-Time Nonlinear systems

Parallel Control-Based Event-Triggered Optimal Control for C...

引用

2021 China Automation Congress, CAC 2021

作者： Lu, Jingwei Bai, Tianxiang Wei, Qinglai Zhou, Tianmin Wang, Fei-Yue School of Artificial Intelligence University of Chinese Academy of Sciences Beijing China The State Key Laboratory for Management and Control of Complex Systems Institute of Automation Beijing China

ISBN: (纸本)9781665426473

This study proposes a new event-triggered optimal control (ETOC) method for discrete-time (DT) constrained nonlinear systems. First, a new triggering condition is proposed. We show the asymptotic stability of the closed-loop system using the proposed triggering condition and analyze the degeneration degree of the real performance index. Second, to perform the proposed ETOC method effectively, parallel control (PC) combined with adaptive dynamic programming (ADP) is applied. Finally, the validity of the ETOC method is validated by a simulation. © 2021 IEEE

关键词： Asymptotic stability

来源：评论

学校读者我要写书评

暂无评论

A Virtual Ultrasonography Simulator for Skill Training Using Magnetic-Inertial Probe Tracking

引用

IEEE-ASME TRANSACTIONS ON MECHATRONICS 2024年第1期29卷 445-454页

作者： Wang, Heng Dong, Shuda Yang, Qi Han, Jiancheng He, Ze'an He, Yihua Wang, Shuangyi South China Univ Technol Shien Ming Wu Sch Intelligent Engn Guangzhou 511442 Peoples R China Capital Med Univ Maternal Fetal Med Ctr Fetal Hear Beijing Anzhen Hosp Echocardiog Med Ctr Beijing 100029 Peoples R China Chinese Acad Sci Inst Automat State Key Lab Management & Control Complex Syst Beijing 100190 Peoples R China Chinese Acad Sci Hong Kong Inst Sci & Innovat Ctr Artificial Intelligence & Robot Hong Kong Peoples R China

Simulation-based ultrasonography training has been proven to be effective for novice practitioners to acquire operation skills. However, existing simulators are cumbersome, expensive, and sometimes fail to provide intuitive and interactive training experience. In this article, a compact, inexpensive, and interactive ultrasonography simulator is developed using magnetic-inertial 6-degree of freedom (DoF) probe pose tracking and virtual ultrasound image generation. Embedded with magnetic and inertial sensors, the sham ultrasound probe is tracked by fusing magnetic measurement of a single electromagnet embedded in the phantom and inertial measurement of the probe. Then the virtual scan plane is determined according to the estimated probe pose to slice the digital anatomical model for virtual image generation, providing visual feedback for skill training. The probe tracking accuracy is experimentally validated to be sufficiently accurate for correct image generation with the RMS position error of 3.67 mm and average orientation error of 0.74 degrees. The interactive performance of the simulator is evaluated by volunteer tests, showing shortened task completion time (40.68 s versus 50.46 s) and slightly improved accuracy of reaching target views compared to the slider control method. It is also shown that an inexperienced operator can learn to successfully acquire standard and diagnostic views of fetal hearts with congenital diseases. Due to intuitive freehand interaction, accurate probe tracking, and real-time visual guidance of ultrasound images and three-dimensional (3-D) virtual scene of the probe and anatomy, the proposed simulator can provide immersive training experience for quick ultrasound skill acquirement.

关键词： Probes Training Magnetic resonance imaging Phantoms Sensors Solid modeling Magnetic sensors

来源：评论

学校读者我要写书评

暂无评论

Bilateral coordination control of flexible master-slave manipulators using a partial differential equation model

引用

JOURNAL OF VIBRATION AND control 2021年第13-14期27卷 1561-1572页

作者： Li, Le Yang, Hongjun Liu, Jinkun Beihang Univ Sch Automat Sci & Elect Engn Xueyuan Rd 37 Beijing 100191 Peoples R China Chinese Acad Sci Inst Automat State Key Lab Management & Control Complex Syst Beijing Peoples R China

In this study, we evaluate the coordination tracking control problem of a flexible master-slave teleoperation system. The system under consideration is based on a dynamic model described by a set of partial differential equations. Existing research on bilateral controllers is based on teleoperation systems composed of rigid master robots and rigid or flexible slave robots. In this work, we consider teleoperation systems with flexible master and slave robots. We dynamically model flexible master-slave manipulators using partial differential equations. Based on the dynamic model, a bilateral coordination controller is developed to realize the coordination angle tracking and vibration suppression of flexible master-slave manipulators. The teleoperation system is proven to be asymptotically stable under the control scheme. Numerical simulation results illustrate that the proposed controller is effective.

关键词： Bilateral teleoperation system flexible manipulator coordination control vibration control partial differential equation model

来源：评论

学校读者我要写书评

暂无评论

ULSM: Underground Localization and Semantic Mapping with Salient Region Loop Closure under Perceptually-Degraded Environment

ULSM: Underground Localization and Semantic Mapping with Sal...

引用

IEEE/RSJ International Conference on Intelligent Robots and systems (IROS)

作者： Wang, Junhui Tian, Bin Zhang, Rui Chen, Long Chinese Acad Sci Inst Automat State Key Lab Management & Control Complex Syst Beijing Peoples R China Univ Chinese Acad Sci Sch Artificial Intelligence Beijing Peoples R China Waytous Inc Beijing Peoples R China

ISBN: (数字)9781665479271

ISBN: (纸本)9781665479271

Simultaneous Localization and Mapping (SLAM) has greatly assisted in exploring perceptually-degraded underground environments, such as human-made tunnels, mine tunnels, and caves. However, the recurring sensor failures and spurious loop closures in these scenes bring significant challenges to applying SLAM. This paper proposes an architecture for underground localization and semantic mapping (ULSM) that promotes the robustness of odometry estimation and map-building. In this architecture, a two-stage robust motion compensation method is proposed to adapt to sensor-failure situations. The proposed salient region loop closure detection contributes to avoiding spurious loop closures. Meanwhile, the 2D pose as the initial value for point cloud registration is estimated without additional input. We also design a multi-robot cooperative mapping scheme based on descriptors of the salient region. Extensive experiments are conducted on datasets collected in the Tunnel Circuit of DARPA Subterranean Challenge.

关键词： Location awareness Point cloud compression Simultaneous localization and mapping Semantics Estimation Robustness Motion compensation

来源：评论

学校读者我要写书评

暂无评论

Sora for Senarios Engineering of Intelligent Vehicles: V&V, C&C, and Beyonds

IEEE TRANSACTIONS ON INTELLIGENT VEHICLES

引用

IEEE TRANSACTIONS ON INTELLIGENT VEHICLES 2024年第2期9卷 3117-3122页

作者： Li, Xuan Miao, Qinghai Li, Lingxi Ni, Qinghua Fan, Lili Wang, Yutong Tian, Yonglin Wang, Fei-Yue Peng Cheng Lab Shenzhen 518000 Peoples R China Univ Chinese Acad Sci Sch Artificial Intelligence Beijing 100049 Peoples R China Indiana Univ Purdue Univ Indianapolis IUPUI Purdue Sch Engn & Technol Dept Elect & Comp Engn Indianapolis IN 46202 USA Macau Univ Sci & Technol Fac Innovat Engn Macau 999078 Peoples R China Beijing Inst Technol Sch Informat & Elect Beijing 100081 Peoples R China Chinese Acad Sci Inst Automat State Key Lab Management & Control Complex Syst Beijing 100190 Peoples R China

The advent of Scenarios Engineering (SE) paves the way to a new era of intelligent vehicles (IVs), driven by Artificial Intelligence (AI)-enabled strategies. It aims at shaping the IVs to be a form that is more relevant to the underlying scenario, thereby accomplishing validation, verification (V&V), and calibration, certification (C&C) of each vehicle. However, such improved capabilities relies on the accumulation and analysis of an unprecedented volume of scenarios. Recently, Sora and other video generation models have opened up new horizons for Imaginative Intelligence. As an extension of TIV-DHW (Distributed/Decentralized Hybrid Workshop) on SE, this letter discusses the potential of Sora to change the scenario generation process by reducing physical shooting, increasing extreme scenario generation, thereby enabling more comprehensive training and testing of IVs. This letter also analyzes the limitations of Sora in accurately model physics and understand cause and effect, which may affect its effectiveness in SE applications. Last, through a comprehensive outlook, this letter aims to provide a potential direction for the development of Sora-like AI technology, thereby promoting the safety, efficiency, reliability, and sustainability of IVs.

关键词： Intelligent vehicles Indexes Biological system modeling Vehicle dynamics Visualization Transformers Training data Sora text-to-video model scenarios engineering artificial intelligence intelligent vehicles

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：