Integration of ROS and RT tasks using message pipe mechanism on Xenomai for telepresence robot

被引:12
作者
Delgado, R. [1 ]
You, B. -J. [2 ]
Han, M. [2 ]
Choi, B. W. [1 ]
机构
[1] Seoul Natl Univ Sci & Technol, Dept Elect & Informat Engn, Seoul, South Korea
[2] Ctr Human Ctr Interact Coexistence, Seoul, South Korea
基金
新加坡国家研究基金会;
关键词
Robots - Computer operating systems - Visual communication;
D O I
10.1049/el.2018.5560
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
A promising approach integrating non-real-time (NRT) robot operating system (ROS) packages and RT tasks is proposed to enhance the development of RT robot control applications. Since ROS alone does not provide RT properties essential for achieving precise control period in manipulating multiple devices and complicated software, Xenomai, an RT extension of Linux is adapted. However, using NRT ROS packages inside RT tasks triggers mode switching that causes inability to satisfy critical temporal constraints. To address this issue, a message pipe mechanism termed cross-domain datagram protocol (XDDP) is applied. In comparison to traditional inter-task mechanisms, XDDP provides a communication interface between RT and NRT tasks. This greatly improves robot application development utilising ROS tools and packages with RT tasks on the Xenomai domain ensuring priority-based scheduling and deterministic response in a multitasking environment. Feasibility of the proposed method was validated for practical use by realisation on the open embedded controller for a telepresence robot. Experiments were conducted to actuate the mobile base of the robot using ROS navigation packages. The results indicate that the robot accomplishes its objectives while satisfying RT constraints.
引用
收藏
页码:127 / 128
页数:2
相关论文
共 7 条
[1]  
[Anonymous], 2014, P SIGGRAPH AS 2014 A
[2]  
[Anonymous], LECT NOTES COMPUTER
[3]   Embedded ROS [J].
Bouchier, Paul .
IEEE ROBOTICS & AUTOMATION MAGAZINE, 2013, 20 (02) :17-19
[4]   Bio-Inspired Embedded Vision System for Autonomous Micro-Robots: The LGMD Case [J].
Hu, Cheng ;
Arvin, Farshad ;
Xiong, Caihua ;
Yue, Shigang .
IEEE TRANSACTIONS ON COGNITIVE AND DEVELOPMENTAL SYSTEMS, 2017, 9 (03) :241-254
[5]  
Marin-Plaza P., 2018, Journal of Advanced Transportation, V2018, DOI DOI 10.1155/2018/6392697
[6]   Performance evaluation of real-time mechanisms on open embedded hardware platforms [J].
Shin U.C. ;
Choi B.W. .
Journal of Institute of Control, Robotics and Systems, 2017, 23 (01) :60-66
[7]   RT-ROS: A real-time ROS architecture on multi-core processors [J].
Wei, Hongxing ;
Shao, Zhenzhou ;
Huang, Zhen ;
Chen, Renhai ;
Guan, Yong ;
Tan, Jindong ;
Shao, Zili .
FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE, 2016, 56 :171-178