Affine Formation Control of General Linear Multi-Agent Systems with Delays

Abstract

This paper studies affine formation control problem of general linear multi-agent systems (MASs) over undirected graph in the presence of time delays. Different delay conditions are considered and predictive observers are designed for the followers to predict the states in the future. Then, affine formation control laws are designed based on the predicted states, and sufficient conditions are derived to ensure the convergence of the formation errors. With an adaptive mechanism, the proposed control laws can be implemented without the usage of global information. Under the proposed protocols, the MASs can form a target formation shape and realize time-varying geometric transformations continuously. Finally, simulations are conducted to verify the effectiveness of the theoretical results.

Triangular formation control with mixed distance and bearing measurements under limited field of view constraints

“…where sgn(⋅) is the signum function. Actually, (11) denotes the sign of z-component of the new vector calculated from the cross product of b * and b. From the right hand rule for cross product, 𝛿 𝜑 is a directional angle starting from b * and ending with b.…”

“…Formation control plays a key role in multi‐agent collaborative control. In recent years, it has received considerable attention due to the wide range of applications, including cooperative transportation, surveillance, and intelligent entertainment performance 6‐11 …”

“…In recent years, it has received considerable attention due to the wide range of applications, including cooperative transportation, surveillance, and intelligent entertainment performance. [6][7][8][9][10][11] Based on the sensing capability and controlled variables, in a recent survey, 12 existing results on formation control can be categorized into absolute position-based, 13 relative displacement-based, [14][15][16] and distance-based approaches. [17][18][19] Position-based strategy control requires that each agent knows its accurate position with respect to a common global coordinate frame.…”

Model‐free learning‐based distributed cooperative tracking control of human‐in‐the‐loop multi‐agent systems

“…Distributed cooperative control of multi‐agent systems (MASs) has deeply permeated various fields, including formation control, 1,2 circuit systems, 3,4 power systems, 5 and more. Distributed tracking control, one of the most fundamental research areas in MASs cooperative control, continues to receive a lot of attention from researchers.…”