Robert A. Paz

A Reliable Control Design for Discrete-Time Systems

This paper presents results on the design of reliable control systems which are characterized by the ability to withstand the outage of any sensor in a predefined subset or of any actuator in a predefined subset of control loops.

Decentralized H ∞ control

This paper presents a method based on the use of algebraic Riccati equations for designing a decentralized control law. The designs provide both greater computational ease as well as improved H∞ performance over previous designs. Results are given in both the continuous and discrete time settings.

Ripple-free tracking with robustness

Despite the previous efforts in deadbeat control, a need still exists to offer a generic design methodology for both performance and robustness. We propose a hybrid, two-degree-of-freedom (2DOF) controller for the fixed-order constrained optimization problem addressing performance and robustness specifications. The controller is given in terms of the solution of two Diophantine equations. Simple parameterization of these Diophantine equations allows for straight-forward optimization and constraint satisfaction for both performance and robustness measures.

Robust Stabilization and Disturbance Attenuation for Discrete-Time Systems with Structured Uncertainty

This paper presents a method based on the use of discrete algebraic Riccati equations for designing state and output feedback control laws for plants with structured uncertainty. The designs provide both robust stability and disturbance rejection with uniform H¿ norm bounds. The design method consists of incorporating information on the plant uncertainty into the Riccati equations used for nominal H¿ suboptimal designs.

ℋ∞control in discrete time: state feedback control and norm bounds

H ∞ norm bounds are useful in control design for disturbance rejection. For discrete systems, H ∞ norm optimizing controls can be characterized via the solution of the Discrete Algebraic Riccati Equation, (DARE). In addition, the DARE can be used to characterize controls that guarantee a uniform H ∞ norm bound. In this paper one identifies an entire class of state feedback controls that guarantee an H ∞ norm bound, and provides new lower bounds for the H ∞ norm of stable systems, as well as for the best achievable H ∞ norm using state feedback controls

Autobalancing control for a reduced gravity simulator

Physical simulation of a reduced gravity condition significantly aids astronaut training, biomechanics researchers, neuro-rehabilitation practitioners as well as other potential applications. This paper provides means for automatically balancing a simple reduced gravity platform.

A testbed for testing an active body support system for locomotion training

This paper describes the design of an experimental testbed used for nonhuman test of an active body-weight support system for treadmill based locomotion rehabilitation. The body support system is designed to off load not only partial weight but also inertia force (dynamic load) of the patient who is supported by the system during neural training. To avoid human subject test in the early phase of the technology development, a cam-slider based testbed is designed to facilitate the test of the active cable control of the body support system. The goal of the test is to see if the testbed can generate an accurate periodical motion pattern mimicking a human walking gait. The paper describes the design of the testbed hardware and a simulation test of the dynamic performance of the testbed.

Nonhuman test of a new active body support system for improving locomotion training

This paper describes the results of a nonhuman test of the concept of a new active body-weight support (BWS) system for improving treadmill-based locomotion rehabilitation. The BWS system can offload not only partial body weight but also partial body mass and thus reduce the dynamic load of the patient who is supported by the system during locomotion training. Due to safety requirements, a cam-slider based testbed has been designed to facilitate the nonhuman test of the new BWS method before human subject testing is developed. The nonhuman test results demonstrated that the new method performs satisfactorily as designed for.

Robust Stabilization and Disturbance Attenuation for Systems with Structured Nonlinearities

This paper presents a method based on the use of covariance control for parametrising a set of fixed-order, linear, disturbance-attenuating output feedback control laws for plants with structured nonlinearities. The disturbance rejection is measured by the induced B2 norm of the operator from the disturbance vector to the regulated output vector. The results are also provided for static output feedback and estimator designs.

Simple computational methods for polynomial interpolations

Polynomial interpolation is the general approach to many control designs. Designs using the root locus, for example, directly apply polynomial interpolations. It is shown that such interpolations can be easily calculated using polynomial root finders. While this may appear obvious, it has previously not been approached this way. These results may be used in computer-aided control-system design software.