检索结果-内蒙古大学图书馆

arXiv 2022年

作者： Ming, Nan Feng, Yi Fan, Rui MIAS Research Group Robotics & Artificial Intelligence Laboratory Department of Control Science & Engineering Tongji University Shanghai201804 China

The state-of-the-art (SoTA) surface normal estimators (SNEs) generally translate depth images into surface normal maps in an end-to-end fashion. Although such SNEs have greatly minimized the trade-off between efficiency and accuracy, their performance on spatial discontinuities, e.g., edges and ridges, is still unsatisfactory. To address this issue, this paper first introduces a novel multi-directional dynamic programming strategy to adaptively determine inliers (co-planar 3D points) by minimizing a (path) smoothness energy. The depth gradients can then be refined iteratively using a novel recursive polynomial interpolation algorithm, which helps yield more reasonable surface normals. Our introduced spatial discontinuity-aware (SDA) depth gradient refinement strategy is compatible with any depth-to-normal SNEs. Our proposed SDA-SNE achieves much greater performance than all other SoTA approaches, especially near/on spatial discontinuities. We further evaluate the performance of SDA-SNE with respect to different iterations, and the results suggest that it converges fast after only a few iterations. This ensures its high efficiency in various robotics and computer vision applications requiring realtime performance. Additional experiments on the datasets with different extents of random noise further validate our SDA-SNE's robustness and environmental adaptability. Our source code, demo video, and supplementary material are publicly available at ***/SDA-SNE. Copyright © 2022, The Authors. All rights reserved.

关键词： Dynamic programming

来源：评论

学校读者我要写书评

暂无评论

Systematic Study of Optical Turbulence in Rayleigh-Bénard Underwater Convection

Systematic Study of Optical Turbulence in Rayleigh-Bénard U...

引用

2023 Propagation Through and Characterization of Atmospheric and Oceanic Phenomena, pcAOP 2023 in Optica Imaging Congress - Part of Imaging and Applied Optics Congress 2023

作者： O'Malley, Owen Avramov-Zamurovic, Svetlana Ferlic, Nathaniel A. Judd, K. Peter Weapons Robotics and Control Engineering Department United States Naval Academy 597 McNair Rd. Hopper Hall AnnapolisMD21402 United States 22347 Cedar Point Road Patuxent RiverMD20670 United States United States Naval Research Laboratory 4555 Overlook Ave. SW WashingtonDC20375 United States

Synchronized optical wavefront and intensity measurements are used to characterize the optical turbulence generated by Rayleigh-Bénard convection. We find refractive index structure constant and the shape of the ... 详细信息

ISBN: (纸本)9781957171289

关键词： Refractive index

来源：评论

学校读者我要写书评

暂无评论

Incremental Joint Learning of Depth, Pose and Implicit Scene Representation on Monocular Camera in Large-scale Scenes

arXiv

引用

arXiv 2024年

作者： Deng, Tianchen Wang, Nailin Wang, Chongdi Yuan, Shenghai Wang, Jingchuan Wang, Danwei Chen, Weidong The Institute of Medical Robotics Department of Automation Shanghai Jiao Tong University Key Laboratory of System Control and Information Processing Ministry of Education Shanghai200240 China The School of Electrical and Electronic Engineering Nanyang Technological University Singapore

Dense scene reconstruction for photo-realistic view synthesis has various applications, such as VR/AR, autonomous vehicles. However, most existing methods have difficulties in large-scale scenes due to three core challenges: (a) inaccurate depth input. Accurate depth input is impossible to get in real-world large-scale scenes. (b) inaccurate pose estimation. Existing methods are not robust enough with the growth of cumulative errors in large scenes and long sequence. (c) insufficient scene representation capability. A single global radiance field lacks the capacity to effectively scale to large-scale scenes. To this end, we propose an incremental joint learning framework, which can achieve accurate depth, pose estimation, and large-scale scene reconstruction. A vision transformer-based network is adopted as the backbone to enhance performance in scale information estimation. For pose estimation, a feature-metric bundle adjustment (FBA) method is designed for accurate and robust camera tracking in large-scale scenes. In terms of implicit scene representation, we propose an incremental scene representation method to construct the entire large-scale scene as multiple local radiance fields to enhance the scalability of 3D scene representation. Extended experiments have been conducted to demonstrate the effectiveness and accuracy of our method in depth estimation, pose estimation, and large-scale scene reconstruction. The code will be open-sourced upon paper acceptance. © 2024, CC BY-NC-SA.

关键词： Depth perception

来源：评论

学校读者我要写书评

暂无评论

Process Assurance for Object Detection Through Deep Neural Networks to Accomplish the Autonomous Aerial Refueling Task

Process Assurance for Object Detection Through Deep Neural N...

引用

Automated Vehicle Validation Conference (IAVVC), IEEE International

作者： Danielle Clement Sarah Mottino Donald Costello Aeronautics Division Lockheed Martin Corporation Fort Worth Texas United States Weapons Robotics and Control Engineering Department United States Naval Academy Annapolis Maryland United States

This paper presents a high level discussion on certifying of a deep neural network to build situational awareness for an autonomous vehicle to complete the aerial refueling task through object detection verification a...

关键词：

来源：评论

学校读者我要写书评

暂无评论

Event-Triggered Attitude controller Design for Unmanned Aerial Vehicles under Cyber Attacks

Event-Triggered Attitude Controller Design for Unmanned Aeri...

引用

IEEE International Symposium on Industrial Electronics (ISIE)

作者： Seungyong Han Xuyang Guo Yongsik Jin Yongseob Lim Suneel Kumar Kommuri Korea Atomic Energy Research Institute Daejeon Republic of Korea Department of Control Science and Engineering Tongji University Shanghai China Electronics and Telecommunications Research Institute Daegu Republic of Korea Department of Robotics and Mechatronics Engineering Daegu Gyeongbuk Institute of Science and Technology Daegu Republic of Korea Department of Electrical and Electronic Engineering Xi’an Jiaotong-Liverpool University Suzhou China

ISBN: (数字)9798350394085

ISBN: (纸本)9798350394092

This paper proposes an event-triggered attitude controller design method for unmanned aerial vehicles (UAVs) under cyber attacks. Versatile UAVs are capable of transmitting the measurement or receiving control signals through network channels. In the controller design for such UAVs, it is required to consider two distinctive features. The first one is that the strategy of updating the control signal is important for saving the limited energy consumption. The other one is that the UAVs are vulnerably exposed to threats of malicious attacks through network channels. For enhanced energy efficiency and improved stability, the event-triggered mechanism (ETM) and the attitude controller are simultaneously designed. By utilizing Lyapunov- Krasovskii functionals (LKFs), the sufficient conditions for the co-design are derived in terms of linear matrix inequalities (LMIs). For a 3-degrees of freedom (DOF) quadcopter system, the simulation results are presented to validate the proposed attitude control method.

关键词： Sufficient conditions Attitude control Simulation Autonomous aerial vehicles Stability analysis Numerical models Resource management

来源：评论

学校读者我要写书评

暂无评论

Time-Optimal Path Tracking for Cooperative Manipulators: A Convex Optimization Approach

arXiv

引用

arXiv 2023年

作者： Haghshenas, Hamed Hansson, Anders Norrlöf, Mikael The Division of Automatic Control Department of Electrical Engineering Linköping University Linköping581-83 Sweden ABB Robotics Division Västerås721-36 Sweden

This paper studies the time-optimal path tracking problem for a team of cooperating robotic manipulators carrying an object. Considering the problem for rigidly grasped objects, we show that it can be cast as a convex optimization problem and solved efficiently with a guarantee of optimality. When formulating the problem, we avoid using a particular wrench distribution and exploit the full actuation available to the system. Then, we consider the problem for grasps using frictional forces and show that this problem also, under a force-closure grasp assumption, can be formulated as a convex optimization problem and solved efficiently and to optimality. To ensure a firm grasp, internal forces have been taken into account in this approach. © 2023, CC BY.

关键词： Convex optimization

来源：评论

学校读者我要写书评

暂无评论

Full-body Grasping Strategy for Planar Underactuated Soft Manipulators using Passivity-based control

Full-body Grasping Strategy for Planar Underactuated Soft Ma...

引用

IEEE International Conference on Soft robotics (RoboSoft)

作者： Hoang Chu Brandon J. Caasenbrood Mahboubeh Keyvanara Irene A. Kuling Henk Nijmeijer Department of Mechanical Engineering Dynamics and Control Group Eindhoven University of Technology Eindhoven MB The Netherlands 4TU Consortium Dutch Soft Robotics University of Twente Enschede Netherlands

Soft robotics is a branch of robotics that aims to emulate nature by exploring so-called soft materials. By utilizing the embedded softness, various degrees of dexterity in grasping can be achieved without the need for advanced controllers. However, when compared to nature (and modern rigid robots), comparable levels of dexterity and object manipulation are still lacking. For example, when considering the elephant's trunk, whole-body manipulation and sensory feedback are explored to achieve simultaneous, robust, and adaptive grasping. In this work, we incorporate closed-loop control into soft robotic grasping. Using passivity-based control, we achieve whole-body grasping for planar, slender, soft manipulators with torque actuation. Our approach also accounts for the underactuation present in these systems and adapts the grasping strategy accordingly. Furthermore, we explore damping injection without velocity measurements to enhance the attenuation of undesired oscillatory motion. The performance of the closed-loop system is evaluated through simulation and experiments.

关键词： Damping Torque Grasping Soft robotics Robot sensing systems Manipulators Sensors

来源：评论

学校读者我要写书评

暂无评论

Learning a Better control Barrier Function

arXiv

引用

arXiv 2022年

作者： Dai, Bolun Krishnamurthy, Prashanth Khorrami, Farshad Control/Robotics Research Laboratory Electrical & Computer Engineering Department Tandon School of Engineering New York University BrooklynNY11201 United States

control barrier functions (CBFs) are widely used in safety-critical controllers. However, constructing a valid CBF is challenging, especially under nonlinear or non-convex constraints and for high relative degree systems. Meanwhile, finding a conservative CBF that only recovers a portion of the true safe set is usually possible. In this work, starting from a "conservative" handcrafted CBF (HCBF), we develop a method to find a CBF that recovers a reasonably larger portion of the safe set. Since the learned CBF controller is not guaranteed to be safe during training iterations, we use a model predictive controller (MPC) to ensure safety during training. Using the collected trajectory data containing safe and unsafe interactions, we train a neural network to estimate the difference between the HCBF and a CBF that recovers a closer solution to the true safe set. With our proposed approach, we can generate safe controllers that are less conservative and computationally more efficient. We validate our approach on two systems: a second-order integrator and a ball-on-beam. © 2022, CC BY.

关键词： controllers

来源：评论

学校读者我要写书评

暂无评论

Distributed Event-Triggered Formation control for a Multi-robotic Fish System

Distributed Event-Triggered Formation Control for a Multi-ro...

引用

International Conference on Guidance, Navigation and control, ICGNC 2022

作者： Dai, Shijie Wu, Zhengxing Tan, Min Yu, Junzhi School of Artificial Intelligence University of Chinese Academy of Sciences Beijing100049 China State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences Beijing100190 China State Key Laboratory for Turbulence and Complex System Department of Advanced Manufacturing and Robotics BIC-ESAT College of Engineering Peking University Beijing100871 China

ISBN: (纸本)9789811966125

This paper investigates the formation control problem for a multi-robotic fish system and a distributed event-triggered-based formation control framework is proposed. The framework consists of a communication topology, a distributed formation control law, and a dynamic event-triggered communication mechanism. In particular, a position-based formation control law is presented to drive the multi-robotic fish system to an anticipated configuration based on local measurements while keeping pace with a moving target. Meanwhile, a dynamic event-triggered mechanism is utilized to measure importance of the interactive information and decide the communication timing. The proposed formation control method shows its merits to greatly reduce communication consumption with a limited loss of formation control performance. Finally, simulations with quantitative comparisons are presented to verify the effectiveness of the proposed framework. This formation control framework provides a solid foundation for future marine cooperative control of the multi-robotic fish system. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Multi agent systems

来源：评论

学校读者我要写书评

暂无评论

Mecanum wheel mobile platform for rapid prototyping of the autonomous control algorithms

引用

IFAC-PapersOnLine 2024年第9期58卷 183-188页

作者： Kacper Jaroszewicz Wojciech Walendziuk Roman Trochimczuk Adam Wolniakowski Radosław Tomaszewski Zdenek Slanina Doctoral School of Bialystok University of Technology Bialystok Poland Department of Electrotechnics Power Electronics and Electrical Power Engineering Bialystok University of Technology Bialystok Poland Department of Automatic Control and Robotics Bialystok University of Technology Bialystok Poland VSB – Technical University of Ostrava 17. listopadu 2172/15 Ostrava - Poruba Czech Republic

This article describes the author’s design of the Mecanum wheel mobile robots (MWR), which can be used as a test platform for experimental research into the development and implementation of autonomous systems. The paper contains a literature review of the subject matter and describes the mechanical actuator subsystem and electronic and information subsystem of the mobile robot solution. The robot was navigated using a minicomputer equipped with a Jetson Nano chip. The robot’s navigation system was implemented using the ROS environment through high-level software. Software modularity is desirable in circuits intended for research and rapid prototyping. This allows for easier modification and maintenance. The microcontroller has been implemented with a DC drive control system and incremental encoder support to estimate robot position using odometry. The project utilizes the Jetson Nano for processing data from distance sensors and monitoring the robot’s performance. The mobile platform is equipped with a RGB-D camera (Microsoft Kinect 360), the robot can use this to perform 3D mapping of the environment. A LiDAR sensor is used to collect data from a two-dimensional space. Laboratory tests were carried out to test the validity of the solutions adopted in the design of the Mecanum-wheeled mobile robot’s navigation system configuration.

关键词： autonomous systems Mecanum wheels robot (MWR) Microsoft Kinect 360 LiDAR sensor

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：