Manoeuvrable multi-agent formation control via time-varying flocking and iterative learning tracking

Abstract

Manoeuvrable (or time-varying) formation control of second-order integral multi-agent networks with single virtual leaders is contrived by incorporating time-varying weighting matrices into flocking algorithms. Existence and properties are examined, including control bounded-ness, average trajectory and formation manoeuvrability with respect to the weighting matrices and virtual leader reference input. More specifically, it is shown that under the time-varying flocking algorithms, the multi-agents ultimately run into lattice formations and their kinematical features in terms of orientation, scaling, subspace specification and homogeneous/heterogeneous consensus can be manoeuvred. Moreover, by gain-scheduling assignment for the leader-average dynamics via the variation-of-constant formula and iterative learning optimisation, trajectory-tracking manoeuvrable formation control can be achieved. Technical advantages include: flexible formations rather than rigid ones can be dealt with; formation and trajectory tracking can be designed separately and implemented simultaneously; diverse formations are achievable by selecting the weighting matrices; trajectory-tracking control is independent of trajectory modelling. Numerical examples are illustrated to show the main results.

Keywords:

• Multi-agent flocking
• formation control
• kinematical manoeuvrability
• trajectory tracking
• gain scheduling
• iterative learning

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This study was supported jointly by the National Nature Science Foundation of China [grant numbers 61573001, 61703098], the Natural Science Foundation of Jiangsu Province [grant number BK20160699] and the Fundamental Research Funds for the Central Universities [grant number 2017B07214].

Notes on contributors

J. Zhou

J. Zhou received the B.S. degree in Radio Engineering from Sichuan University, Chendu, China, in 1984; the M.S. degree in Information and Control from Lanzhou University, Lanzhou, China, in 1987 and the Ph.D. degree in Electrical Engineering from Kyoto University, Kyoto, Japan, respectively. Currently, he works as a professor with the Department of Control Engineering at the School of Energy and Electrical Engineering, Hohai University, Nanjing, China. His latest research topics include: nonlinear/hybrid systems and control, robustness performance synthesis, neural, multi-agent and distributed sensor networks, stabilisation of multi-machine power systems, harmonic analysis of periodic systems and control.

R. Huang

R. Huang received the B.S. degree in automation from Nanjing Technology University, Nanjing, China in 2017, and received the master degree in Engineering, from Hohai University, Nanjing, China. Currently working as a software engineer at Unisoc. Her research interests include control theory and applications in multi-agent networks.

H. Q. Huang

H. Q. Huang received the bachelor's degree in automation from the China University of Mining and Technology, China, in 2007, the master's degree in Agricultural Mechanization Engineering from Nanjing Agricultural University, China, in 2010, and the Ph.D. degree in instrument science and technology from Southeast University, China, in 2015. From 2015 to 2017, he was a Postdoctoral Researcher with Southeast University. He then joined Hohai University, China, where he is an Associate Professor with the School of Energy and Electrical Engineering. His main research interests include navigation technology applied to underwater glider, inertial navigation, and filtering methods. He received Bronze Medal at the 43th Geneva International Exhibition of Inventions for his Underwater Navigation System Design.