Trajectory control of nonlinear, time-varying, discrete-time systems using discrete-time sliding control

The Deep Submergence Laboratory at the Woods Hole Oceanographic Institution is currently developing supervisory control system methodologies for underwater vehicles and manipulators. An effective low-level control system is a key component in the successful implementation of the high-level supervisory control scheme. The design of this low-level control system for underwater vehicles and manipulators is particularly challenging because of the fluidic environment in which they operate. These environmental effects, when compounded by the uncertain, nonlinear dynamics that characterize such systems, require the use of control techniques beyond traditional methods to ensure reliable and predictable performance. A new discrete-time control methodology, based on continuous-time sliding control concepts, is presented in order to explicitly account for the digital implementation of the controller. The methodology directly addresses the control of systems with nonlinear, time-varying dynamics with uncertain parameters. Quantifiable modeling-performance trade-offs are obtained while accounting for the presence of high-frequency unmodeled system dynamics. The discrete-time sliding control method is demonstrated in simulation and compared to a continuous-time sliding controller.