中国科学院重庆绿色智能技术研究院机构知识库

QQ客服 官方微博 反馈留言

	Real-time cooperative kinematic control for multiple robots in distributed scenarios with dynamic neural networks
	Liu, Mei1,2; Zhang, Jiazheng1,2; Shang, Mingsheng1,2
	2022-06-28
摘要	This work investigates the real-time cooperative kinematic control for multiple robots in distributed sce-narios. The information interaction among robots is described as a directed topology in which each robotic manipulator is connected to specified nodes. Besides, a performance index of minimum joint velocity norm is introduced to regulate the robots' motion. Combining with the performance index and integrating the related constraints, a control scheme is developed to guide the behavior of robots with an error feedback term leveraged to improve the real-time execution accuracy of tasks. Moreover, considering that the involved parameters are time-dependent, a dynamic neural-network-assisted solver is constructed to solve the control scheme online. In the end, simulations and illustrative experiments on the virtual robot experiment platform (V-REP) are conducted by a set of UR5 robots to execute tasks in a switchable directed topology, where the corresponding results demonstrate the effectiveness of the pro-posed scheme. (c) 2021 Elsevier B.V. All rights reserved.
关键词	Multiple redundant manipulators Cooperative control Motion planning Dynamic neural network
DOI	10.1016/j.neucom.2021.12.038
发表期刊	NEUROCOMPUTING
ISSN	0925-2312
卷号	491页码:621-632
通讯作者	Shang, Mingsheng(msshang@cigit.ac.cn)
收录类别	SCI
WOS记录号	WOS:000796026700003
语种	英语