Distributed Adaptive Leader-Following Consensus for Nonlinear Multiagent Systems With Actuator Failures Under...

Area of Science:

• Control Theory
• Systems Engineering
• Robotics

Background:

• Multiagent systems (MASs) face challenges in achieving consensus due to actuator failures and complex communication topologies.
• Existing methods often require strong assumptions on communication graphs and lack robustness to unmatched actuator redundancy.

Purpose of the Study:

• To develop a distributed adaptive output-feedback consensus protocol for nonlinear MASs with multiactuator failures.
• To address systems with unmatched redundancy and directed switching graphs, relaxing prior communication constraints.

Main Methods:

• A novel distributed reference generator using neighbor information is designed.
• A reduced-order dynamic gain filter compensates for unmeasured states.
• A recursive design method integrates the generator and filter into a distributed adaptive protocol for failure compensation.

Main Results:

• The proposed scheme relaxes communication graph conditions, allowing disconnections.
• It significantly reduces filter dimension and numerical challenges.
• Achieves semiglobal stability and consensus error convergence unaffected by actuator failures.

Conclusions:

• This work presents the first solution for output-feedback consensus in nonlinear MASs with actuator failures and unmatched redundancy.
• The method offers robustness, reduced complexity, and improved stability guarantees for practical applications.