Properties of interconnected negative imaginary systems and extension to formation‐containment control of networked multi‐UAV systems with experimental validation results

Yu‐hsiang Su, Parijat Bhowmick, Alexander Lanzon

Research output: Contribution to journalArticlepeer-review

16 Downloads (Pure)

Abstract

This paper extends the properties of a positive feedback interconnection of two negative imaginary (NI) systems to multi-agent NI systems and proposes a new formation-containment control methodology relying on the characteristic loci technique. Inspired by recent applications of NI and passivity-based control techniques in the multi-agent systems (MAS) domain, a new formation-tracking and containment control scheme is developed for a class of networked multi-UAV systems. The proposed scheme offers a two-stage and two-loop control configuration where the inner loop uses a cascaded PID controller to ensure stable hovering of the UAVs, and the outer loop deploys a distributed “mixed” SNI and strictly passive controller to achieve the formation-containment objectives. This scheme works with a dynamic output feedback control strategy; hence, it offers advantages when the full-state measurement is not possible. In contrast to the well-known Lyapunov theory-based cooperative control schemes, the present one exploits the characteristic loci technique to prove the formation-tracking and containment phenomena theoretically. The paper also provides experimental validation results on a fleet of Crazyflie 2.1 nano quadcopters.

Original languageEnglish
JournalAsian Journal of Control
Early online date4 Dec 2023
DOIs
Publication statusE-pub ahead of print - 4 Dec 2023

Keywords

  • characteristics loci
  • consensus
  • formation-containment control
  • multi-UAV systems
  • negative imaginary systems
  • positive feedback

Fingerprint

Dive into the research topics of 'Properties of interconnected negative imaginary systems and extension to formation‐containment control of networked multi‐UAV systems with experimental validation results'. Together they form a unique fingerprint.

Cite this