This dissertation addresses the design and stability analysis of decentralized, cooperative
control laws for multivehicle systems. Advances in communication, navigation,
and surveillance systems have enabled greater autonomy in multivehicle systems, and
there is a shift toward decentralized, cooperative systems for computational efficiency
and robustness. In a decentralized control scheme, control inputs are determined
onboard each vehicle; therefore, decentralized controllers are more efficient for large
numbers of vehicles, and the system is more robust to communication failures and
reconfiguration.
The design of decentralized, cooperative control laws is explored for a nonlinear
vehicle model that can be represented in a double-integrator form. Cooperative controllers
are functions of spacing errors with respect to other vehicles in the system,
where the communication structure defines the information that is available to each
vehicle. Control inputs are selected to achieve internal stability, or zero steady-state
spacing errors, between vehicles in the system.
Closed-loop equations of motion for the cooperative system can be written in a
structural form, where damping and stiffness matrices contain control gains acting on
the velocity and positions of the vehicles, respectively. The form of the stiffness matrix
is determined by the communication structure, where different communication structures yield different control forms. Communication structures are compared using
two structural analysis tools: modal cost and frequency-response functions, which
evaluate the response of the multivehicle systems to disturbances. The frequency-response
information is shown to reveal the string stability of different cooperative
control forms.
The effects of time delays in the feedback states of the cooperative control laws
on system stability are also investigated. Closed-loop equations of motion are modeled
as delay differential equations, and two stability notions are presented: delay-independent
and delay-dependent stability.
Lastly, two additional cooperative control forms are investigated. The first control
form spaces vehicles along an arbitrary path, where distances between vehicles
are constant for a given spacing parameter. This control form shows advantages over
spacing vehicles using control laws designed in an inertial frame. The second control
form employs a time-based spacing scheme, which spaces vehicles at constant-time
intervals at a desired endpoint. The stability of these control forms is presented.
| Identifer | oai:union.ndltd.org:tamu.edu/oai:repository.tamu.edu:1969.1/ETD-TAMU-2009-05-696 |
| Date | 16 January 2010 |
| Creators | Weitz, Lesley A. |
| Contributors | Hurtado, John E. |
| Source Sets | Texas A and M University |
| Language | en_US |
| Detected Language | English |
| Type | Book, Thesis, Electronic Dissertation |
| Format | application/pdf |
Page generated in 0.0019 seconds