Position and Attitude Station-Keeping of a Free-Flying Telerobotic Vehicle

This thesis presents the development of a comprehensive system of state estimation and

closed loop control for the Multimode Proximity Operations Device (MPOD), a neutral

buoyancy telerobot designed to simulate the role of an orbital maneuvering vehicle

(OMV). The state estimator implements a modelless, measurement based, observer

algorithm that outputs a 16 element state vector for use in a closed loop control system.

Sensory inputs to the estimator are of widely differing bandwidths, spanning two orders

of magnitude. The output is an estimate of vehicle state which is sufficiently accurate

and stable to permit closed-loop control of all six degrees-of-freedom.

The closed-loop control system implements a set of PD servo controllers to perform

station-keeping and automated free-flight for the neutral buoyancy telerobot. The

controller uses the state estimate as an input, compensates for disturbances to the

system, and provides stable closed loop response to step inputs in the vehicular set-

point.

Performance of the state estimator as well as the servo controllers are presented and

evaluated. State estimates with the vehicle stationary, present typical standard deviations

of a few centimeters in position and less than 30 milliradians in attitude. Results from

the closed-loop controller include the ability to compensate for a wide range of

disturbances to the system, and well behaved response to large step changes in the

desired vehicle state.