online read us now
Paper details
Number 3 - September 2022
Volume 32 - 2022
Decentralized multi-agent formation control with pole-region placement via cone-complementarity linearization
Antonio González, Antonio Sala, Leopoldo Armesto
Abstract
An output-feedback decentralised formation control strategy is pursued under pole-region constraints, assuming that the
agents have access to relative position measurements with respect to a set of neighbors in a graph describing the sensing
topology. No communication between the agents is assumed; however, a shared one-way communication channel with a
pilot is needed for steering tasks. Each agent has a separate copy of the same controller. A virtual structure approach is presented for the formation steering as a whole; actual formation control is established via cone-complementarity linearization algorithms for the appropriate matrix inequalities. In contrast to other research where only stable consensus is pursued, the proposed method allows us to specify settling-time, damping and bandwidth limitations via pole regions. In addition, a full methodology for the decoupled handling of steering and formation control is provided. Simulation results in the example section illustrate the approach.
Keywords
multi-agent autonomous system, formation control, decoupling, linear matrix inequalities, cone-complementarity linearization