Multi-agent flocking formation driven by distributed control with topological specifications

Abstract

In this paper, a class of multi-agent flocking control algorithms are contrived, which can specify communication networking, inter-agent distance and position-pattern subgrouping for second-order particle multi-agent systems, by embedding Olfati-Saber's algorithm with what we call the topological specification matrices/vectors. In other words, by equipping Olfati-Saber's algorithms with more weighting factors, the steady-state formation behaviours in a concerned multi-agent system can be manoeuvered while collision-free flocking is induced. More specifically, we create novel collective potential functions with the topological specification matrices/vectors to evaluate multi-agent dynamics and achieve the expected steady-state features in terms of graph connectivity, distance keeping, position-pattern-related subgrouping and obstacle avoidance. Moreover, formation convergence can be improved by integral specification of position difference. Numerical simulations well demonstrate effectiveness of the proposed algorithms.

Keywords:

• Multi-agent flocking
• graph connectivity
• inter-agent distance
• pattern-related subgroup
• potential function
• formation convergence

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This study is supported by the National Nature Science Foundation of China [grant number 61573001].