检索结果-内蒙古大学图书馆

sTetro-D: A deep learning based autonomous descending-stair cleaning robot

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE 2023年 120卷

作者： Prabakaran, Veerajagadheswar Le, Anh Vu Kyaw, Phone Thiha Kandasamy, Prathap Paing, Aung Mohan, Rajesh Elara Singapore Univ Technol & Design ROAR Lab Engn Prod Dev Singapore 487372 Singapore Ton Duc Thang Univ Fac Elect & Elect Engn Commun & Signal Proc Res Grp Ho Chi Minh City 700000 Vietnam Yangon Technol Univ Dept Mechatron Engn Yangon 11101 Myanmar Singapore Univ Technol & Design 8 Somapah Rd Singapore 487372 Singapore

The robots that can perform cleaning in the staircase region are gaining interest and are lined up to release in commercial robot space. Even though several precedents in the literature reported the development of staircase cleaning, their primary focus was on autonomous staircase ascending, which curtails their performance by increasing the power consumption and constraints to achieve full area coverage in a multistory building. The main objective of this research article is to develop a novel autonomous descending staircase-cleaning robot named sTetro-D. The developed robot uses a Deep Convolution Neural Network (DCNN) to autonomously detect a descending staircase, approach it, and perform maximum area coverage. This article presents the technical details of the developed robot and its DCNN-based autonomous descending staircase coverage ability. Also, the developed system was validated in terms of accuracy in detecting the descending staircase, reaching the stairs, and performing maximum area coverage through conducting experimental trials in two real-world scenarios. In all considered scenarios, the developed robotic platform exhibits significantly superior performance in detecting the descending staircase with an average accuracy of 85%, successfully approaching the descending staircase, and achieving 98% area coverage.

关键词： Cleaning robot Staircase traversing robot reconfigurable robot Sensor information fusion Image processing Deep Convolution Neural Network RGBD camera

来源：评论

学校读者我要写书评

暂无评论

Efficient Modelling Methodology for reconfigurable Underwater robots

引用

IFAC-PapersOnLine 2016年第23期49卷 74-80页

作者： Nielsen, Mikkel Cornelius Blanke, Mogens Schjølberg, Ingrid Centre for Autonomous Marine Operations and Systems Department of Engineering Cybernetics Norwegian University of Science and Technology Trondheim Norway Department of Electrical Engineering Technical University of Denmark Lyngby Denmark

This paper considers the challenge of applying reconfigurable robots in an underwater environment. The main result presented is the development of a model for a system comprised of N, possibly heterogeneous, robots dynamically connected to each other and moving with 6 Degrees of Freedom (DOF). This paper presents an application of the Udwadia-Kalaba Equation for modelling the reconfigurable Underwater robots. The constraints developed to enforce the rigid connection between robots in the system is derived through restrictions on relative distances and orientations. To avoid singularities in the orientation and, thereby, allow the robots to undertake any relative configuration the attitude is represented by quaternions. © 2016

关键词： Modular robots Autonomous underwater vehicles Degrees of freedom (mechanics) Modelling methodology Multi body dynamic Quaternion reconfigurable robot Relative distances Udwadia Kalaba Underwater environments Underwater robots

来源：评论

学校读者我要写书评

暂无评论

JL-2: A Mobile Multi-robot System with Docking and Manipulating Capabilities

引用

INTERNATIONAL JOURNAL OF ADVANCED robotIC SYSTEMS 2010年第1期7卷 9-18页

作者： Wang, Wei Yu, Wenpeng Zhang, Houxiang Beihang Univ Inst Robot Beijing Peoples R China Univ Hamburg Dept Informat TAMS Hamburg Germany

This paper presents a new version of the JL series reconfigurable multi-robot system called JL-2. By virtue of the docking manipulator composed of a parallel mechanism and a cam gripper, every mobile robot in the JL-2 system is able to not only perform tasks in parallel, e. g. moving and grasping, but also dock with each other even if there are large misalignments between two robots. A motorized spherical joint is formed between two docked robots to enhance the locomotion capability of JL-2. To fulfill the demands of reconfiguration, a distributed control system and sonar based docking guidance system are designed for the JL-2 prototype. Based on the above design, the JL-2 prototype has been built and successfully demonstrated to confirm the validity and functionality of the proposed capabilities.

关键词： reconfigurable robot Mobile robot Docking mechanism Manipulator Distributed control

来源：评论

学校读者我要写书评

暂无评论

Adaptable reconfigurable and Power-Efficient robots for Health Monitoring 2

Adaptable Reconfigurable and Power-Efficient Robots for Heal...

引用

2nd IEEE International Conference on Advances in Electrical, Computing, Communication and Sustainable Technologies (IEEE ICAECT)

作者： Dasi, Swathi Malladhi, Nagarjuna Tigadi, Arun Majji, Sankararao Patnala, Tulasi Radhika Reddy, V. Prabhakar Pvp Siddhartha Inst Technol Dept ECE Vijayawada India Vardhaman Coll Engn Dept ECE Hyderabad India KLE Dr MS Sheshgiri Coll Engn & Technol Belagavi India GRIET Dept ECE Hyderabad India Shanax Technol Hyderabad India LBRCE Dept EEE Mylavaram India

ISBN: (纸本)9781665411202

A reconfigurable computing system (RCS) is one that can be modified to meet certain functional needs. reconfigurable electronics, such as reconfigurable field programmable gate arrays (FPGAs), make it simple to build such a computer system. Today's computing systems use reconfigurable devices so that they can evolve and adapt to changing application environments while still completing their assigned tasks. First, a reconfigurable robot system is developed, which is aimed at achieving modularity and compactness of the robot modules through the use of a simple yet effective power efficiency estimation-based health and fault detection technique which is used RBIST method. For the purpose of this study, we examined a case study design in the Xilinx ML505 target board to determine the feasibility and complexity of constructing a self-repairing system. The proposed method's efficacy has been demonstrated experimentally.

关键词： Health Monitoring reconfigurable robot RBIST FPGA reconfigurable computing system.

来源：评论

学校读者我要写书评

暂无评论

A model-based design flow for Dynamic Partial reconfigurable FPGAs

A model-based design flow for Dynamic Partial Reconfigurable...

引用

IEEE International Conference on Systems, Man and Cybernetics (SMC)

作者： Giordano, Enrico Di Marco, Federico Pravadelli, Graziano Makarenalabs SRL Verona Italy Univ Verona Dept Comp Sci Verona Italy

ISBN: (纸本)9781728145693

Dynamic Partial reconfigurable FPGAs (DPR-FPGA) are integrated programmable circuits that are dynamically reconfigurable: the designer can reconfigure them runtime, depending on the (dynamic) environment. This paper presents a new model-based design flow for DPR-FPGAs, based on the creation of a specific Extended Finite State Machine (EFSM) which formally describes the hardware. The design flow uses Pyngu, a tool which automatizes the generation of code and simulation through SystemC. This methodology is applied to a reconfigurable robot deployed in different scenarios, which can reconfigure its kinematics in order to overcome obstacles: the kinematics depends on the robots shape, enabling different movement types like trotting, crawling and getting up. Experiments show that the model-based design flow leads to effective adaptable robots design.

关键词： dynamic partial reconfigurable FPGA reconfigurable robot design methodology

来源：评论

学校读者我要写书评

暂无评论

Evaluation of the Travel Efficiency of a Transformable Snake-Like robot Utilizing Infinite Rotation Joint

引用

JOURNAL OF robotICS AND MECHATRONICS 2024年第6期36卷 1328-1338页

作者： Yamano, Akio Kimoto, Tsuyoshi Osaka Metropolitan Univ Grad Sch Engn Dept Aerosp Engn 1-1 Gakuen ChoNaka Ku Sakai Osaka 5998531 Japan

Snake-like robots can achieve flexible movement by simultaneously actuating multiple joints;however, the challenge of high power consumption by driving numerous servomotors under high-load conditions remains. To address this issue, we propose a mechanism that transforms the rear link of a snake-like robot into a wheel-like configuration, enabling a three-wheeled vehicle mode that provides the same traveling speed and efficiency as a wheeled mobile robots on flat surfaces. First, we detail the method for driving the servomotor to achieve undulating locomotion in the snake-like robot with nonuniform link lengths. Next, we propose a method for smoothly switching between the complex- shaped wheel mode and undulating locomotion. Finally, we conduct experiments to assess the travel efficiency in both the undulating mode and the proposed wheel mode across various road surface conditions. Our results demonstrate that the wheel mode achieves higher travel efficiency than the undulating mode on the smooth floors and asphalt.

关键词： biologically-inspired robots snake-like robot reconfigurable robot travel efficiency

来源：评论

学校读者我要写书评

暂无评论

Valid Joint Workspace and Self-aligning Docking Conditions of A reconfigurable Mobile Multi-robots System

Valid Joint Workspace and Self-aligning Docking Conditions o...

引用

International Conference on reconfigurable Mechanisms and robots

作者： Wang, Wei Zhang, Houxiang Zhang, Jianwei Zong, Guanghua Beihang Univ Inst Robot Beijing Peoples R China

ISBN: (纸本)9788889007372

JL-I is a reconfigurable multi-robots system based on parallel mechanisms and cone-shaped docking mechanisms. To reveal how the force cooperation between robots enhances the terrain adaptability of JL-I and to discover the limits of the current reconfiguration mechanism, this paper analyzes the forces arising between robots during cooperation in docking and docked state. The static analysis of the posture-adjusting mechanism yields the force amplifying feature and the valid workspace of this motorized spherical joint consisting of a parallel and serial mechanism. This paper also deduces the self-aligning conditions of the docking mechanism, which is useful to improve the mechanism and clarify the demands on the docking guidance sensors. At last, a series of onsite experiments confirm the analysis and perform two new types of locomotion according to the results of static analysis.

关键词： reconfigurable robot mobile robot

来源：评论

学校读者我要写书评

暂无评论

Drivable workspace analysis of the reconfiguration mechanism in a mobile robots system

Drivable workspace analysis of the reconfiguration mechanism...

引用

IEEE International Conference on robotics and Biomimetics (ROBIO 2007)

作者： Wang, Wei Zhang, Houxiang Ful, Hualei Zhang, Jianwei Zong, Guanghua Beihang Univ Inst Robot Beijing 100083 Peoples R China Univ Hamburg TAMS Dept Informat D-22527 Hamburg Germany

ISBN: (纸本)9781424417612

At first, this paper briefly introduces a novel reconfigurable mobile robots system JL-1 and the serial and parallel mechanism using to perform the reconfiguration actions. Based on the kinematics equations of the serial and parallel mechanism, the static transformation equations of it are concluded according to the virtual work theory, which is useful in determining the drivable workspace of the reconfiguration mechanism. Then the valid workspace of the serial and parallel mechanism of JL-1 is concluded via the transformation matrix of the loads in world coordinates to those on the driving motors according to the real parameters of the motors in JL-1 and the weight of JL-1 robot. The degeneration phenomenon of the workspace is found due to the limited torque of the rotating motor, and a drivable workspace is acquired. At last, by choosing the valid joint path in the drivable workspace, a series of experiments are performed to demonstrate the validity of the theory concluded in this paper.

关键词： workspace reconfigurable robot parallel mechanism serial mechanism

来源：评论

学校读者我要写书评

暂无评论

Reinforcement Learning-Based Complete Area Coverage Path Planning for a Modified hTrihex robot

引用

SENSORS 2021年第4期21卷 1067页

作者： Apuroop, Koppaka Ganesh Sai Le, Anh Vu Elara, Mohan Rajesh Sheu, Bing J. Singapore Univ Technol & Design ROAR Lab Engn Prod Dev Singapore 487372 Singapore Ton Duc Thang Univ Optoelect Res Grp Fac Elect & Elect Engn Ho Chi Minh City 700000 Vietnam Chang Gung Univ Elect Engn & Informat Management Taoyuan 330 Taiwan

One of the essential attributes of a cleaning robot is to achieve complete area coverage. Current commercial indoor cleaning robots have fixed morphology and are restricted to clean only specific areas in a house. The results of maximum area coverage are sub-optimal in this case. Tiling robots are innovative solutions for such a coverage problem. These new kinds of robots can be deployed in the cases of cleaning, painting, maintenance, and inspection, which require complete area coverage. Tiling robots' objective is to cover the entire area by reconfiguring to different shapes as per the area requirements. In this context, it is vital to have a framework that enables the robot to maximize the area coverage while minimizing energy consumption. That means it is necessary for the robot to cover the maximum area with the least number of shape reconfigurations possible. The current paper proposes a complete area coverage planning module for the modified hTrihex, a honeycomb-shaped tiling robot, based on the deep reinforcement learning technique. This framework simultaneously generates the tiling shapes and the trajectory with minimum overall cost. In this regard, a convolutional neural network (CNN) with long short term memory (LSTM) layer was trained using the actor-critic experience replay (ACER) reinforcement learning algorithm. The simulation results obtained from the current implementation were compared against the results that were generated through traditional tiling theory models that included zigzag, spiral, and greedy search schemes. The model presented in the current paper was also compared against other methods where this problem was considered as a traveling salesman problem (TSP) solved through genetic algorithm (GA) and ant colony optimization (ACO) approaches. Our proposed scheme generates a path with a minimized cost at a lesser time.

关键词： reconfigurable robot tiling robots reinforcement learning complete coverage planing energy path planning

来源：评论

学校读者我要写书评

暂无评论

NOVEL LOCKABLE PASSIVE JOINTS FOR JOINT LOCKING-BASED reconfigurable MECHANISMS

NOVEL LOCKABLE PASSIVE JOINTS FOR JOINT LOCKING-BASED RECONF...

引用

ASME International Mechanical Engineering Congress and Exposition (IMECE)

作者： Rosyid, Abdur El-Khasawneh, Bashar Khalifa Univ Ctr Autonomous Robot Syst KUCARS Abu Dhabi U Arab Emirates Khalifa Univ Mech Engn Dept Abu Dhabi U Arab Emirates

ISBN: (纸本)9780791887639

This paper presents a novel design of lockable passive prismatic, revolute, universal, and spherical joints that can be used in a joint locking-based reconfigurable mechanism. The proposed lockable joints offer several advantages, including 1) modularity, 2) small locking force/torque, 3) low-power active trigger, 4) no-slip, and 5) simplicity. All the lockable passive joints proposed in this work provide locking at one or multiple discrete positions, considering that it is the requirement. The modular design, in particular, provides a cost-effective solution for the manufacturing and maintenance of the joints. Designed to have one or more discrete locking positions, the proposed lockable joints have a self-aligning capability that increases the reliability of the locking action. The locking action can be triggered by a low-cost, low-power, short-stroke binary piston such as a solenoid. While the trigger only requires small power, the locking action provides strong locking provided by mechanical locking.

关键词： Lockable joint lockable passive joint reconfigurable mechanism reconfigurable robot

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：