Hierarchical control of cooperative nonlinear dynamical systems

    Authors

    Y. J. Xu; M. Xin; J. A. Wang;S. Jayasuriya

    Comments

    Authors: contact us about adding a copy of your work at STARS@ucf.edu

    Abbreviated Journal Title

    Int. J. Control

    Keywords

    cooperative control; hierarchical control; formation control; VMC; obstacle avoidance; MODEL-PREDICTIVE CONTROL; PSEUDOSPECTRAL METHOD; CAMOUFLAGE; CONSENSUS; MOTION; Automation & Control Systems

    Abstract

    Cooperative control methods that are scalable with low computational cost are crucial for networked dynamical systems to respond quickly in unknown or cluttered environments. In an attempt to make the problem tractable, many existing cooperative controls are designed with oversimplified assumptions and/or without the capabilities of rapidly handling different environmental and dynamical constraints. In this article, proposed is a two-level hierarchical, cooperative control framework using a divide-and-conquer strategy so that challenges can be separately handled at different levels. It is scalable and has low computational cost. Based on a simplified homogeneous double-integrator dynamic model, the top-level planner first computes cooperative trajectories satisfying obstacle avoidance requirements. Then at the lower level, state and control constraints, nonlinear dynamics and self-collision/obstacle avoidance as related to the real system are addressed through a bio-inspired fast trajectory planning algorithm. The stability of the overall hierarchical structure is proven. Two examples, a differential-drive ground vehicle formation control and an unmanned aerial vehicle formation flight, are used to illustrate the advantages of the proposed hierarchical framework.

    Journal Title

    International Journal of Control

    Volume

    85

    Issue/Number

    8

    Publication Date

    1-1-2012

    Document Type

    Article

    Language

    English

    First Page

    1093

    Last Page

    1111

    WOS Identifier

    WOS:000305256900010

    ISSN

    0020-7179

    Share

    COinS