Roy S. Smith

Model validation: a connection between robust control and identification

The gap between the models used in control synthesis and those obtained from identification experiments is considered by investigating the connection between uncertain models and data. The model validation problem addressed is: given experimental data and a model with both additive noise and norm-bounded perturbations, is it possible that the model could produce the observed input-output data? This problem is studied for the standard H/sub infinity // mu framework models. A necessary condition for such a model to describe an experimental datum is obtained. For a large class of models in the robust control framework, this condition is computable as the solution of a quadratic optimization problem. >

Control of Deep-Space Formation-Flying Spacecraft; Relative Sensing and Switched Information

Spacecraft formations in deep space can be specified in terms of the relative spacecraft positions and absolute spacecraft orientations. Accurate absolute position measurements are not available, and a formation control design approach, based on relative position information, is presented. This approach exploits the redundancy inherent in ar elative position specification to develop a family of equivalent control topologies for optimal formation control. Switching between topologies provides redundancy and allows combinations of measured relative positions and communicated estimates to be used in the implementation of the optimal formation control. Redundancy can also exist in the actuation specification, and this can be exploited to control the formation centroid, or to implement the minimum fuel control. Information and control switching can be implemented asynchronously without requiring supervisory control. A four-spacecraft, two-dimensional simulation example is used to illustrate the concepts.

The design of H/sub /spl infin// controllers for an experimental non-collocated flexible structure problem

This paper describes results in applying robust control techniques to achieve vibration suppression of an active precision truss structure. The active structure incorporates piezoelectric members which serve as both structural and actuation elements. The problem considered is multiple-input, multiple-output with non-collocated actuators and sensors. Several characterizations of uncertainty are studied and the resulting controllers are compared experimentally. One characterization uses a novel approach involving eigenvalue perturbation descriptions. >

Model Predictive Climate Control of a Swiss Office Building: Implementation, Results, and Cost–Benefit Analysis

This paper reports the final results of the predictive building control project OptiControl-II that encompassed seven months of model predictive control (MPC) of a fully occupied Swiss office building. First, this paper provides a comprehensive literature review of experimental building MPC studies. Second, we describe the chosen control setup and modeling, the main experimental results, as well as simulation-based comparisons of MPC to industry-standard control using the EnergyPlus simulation software. Third, the costs and benefits of building MPC for cases similar to the investigated building are analyzed. In the experiments, MPC controlled the building reliably and achieved a good comfort level. The simulations suggested a significantly improved control performance in terms of energy and comfort compared with the previously installed industry-standard control strategy. However, for similar buildings and with the tools currently available, the required initial investment is likely too high to justify the deployment in everyday building projects on the basis of operating cost savings alone. Nevertheless, development investments in an MPC building automation framework and a tool for modeling building thermal dynamics together with the increasing importance of demand response and rising energy prices may push the technology into the net benefit range.

A Decoupled Feedback Structure for Covertly Appropriating Networked Control Systems

Abstract The use of communication networks for the transmission of control system signals (measurements and actuation) is becoming more widespread. If the communication (or the sensors and actuators themselves) can be modified by a malicious agent then the control of the physical plant can be covertly appropriated. A parameterised decoupling structure is introduced which allows the covert agent a wide range of control actions on the physical plant while remaining undetectable from the point of view of the original networked controller. The designer of the covert agent need only have a model of the physical plant; knowledge of the networked controller is not required. A MIMO process control example (based on the control of irrigation canals) is used to illustrate the concepts.

Continuous-time control model validation using finite experimental data

The application of robust control theory requires models containing unknown, bounded perturbations and unknown, bounded input signals. Model validation is a means of assessing the applicability of a given model with respect to experimental data. This paper develops a theoretical framework, and a computational solution, for the model validation problem in the case where the model, including unknown perturbations and signals, is given in the continuous time domain, yet the experimental datum is a finite, sampled signal. The continuous nature of the unknown components is treated directly with a sampled data lifting theory. This gives results which are valid for any sample period and any datum length. Explicit calculation of whether sufficient data for invalidation has been obtained arises naturally in this framework. A common class of robust control models is treated and leads to a convex matrix optimization problem. A simulation example illustrates the approach.

A generalization of the structured singular value and its application to model validation

The structured singular value /spl mu/ gives a means of assessing the stability and performance of a system under a class of norm-bounded structured perturbations. This paper introduces a generalization of /spl mu/ which guarantees stability with respect to perturbations, where some elements of the perturbation structure are bounded from above and some are bounded from below. This generalization is used to solve the frequency domain robust control model validation problem. Model validation determines whether or not there exist a model in the model set and a bounded noise signal that are consistent with the experimental input-output measurements. A computable convex upper bound to the generalized /spl mu/ problem is presented, and the relationship between the generalized /spl mu/ problem and the standard /spl mu/ and model validation problems is discussed.

Control topologies for deep space formation flying spacecraft

A deep space satellite formation can be specified in terms of the relative satellite positions and absolute satellite orientations. Redundancy in the relative position specification generates a family of control topologies with equivalent stability and reference tracking performance. We give a characterization of this equivalence and show that there exists a control topology which achieves a global tracking objective using only local controllers. This local relative topology can be implemented without requiring communication between the formation spacecraft.

Design of distributed decentralized estimators for formations with fixed and stochastic communication topologies

This paper proposes a solution to the problem of synthesizing distributed decentralized estimators for a formation of agents. The collected dynamics of the formation are modeled by a discrete LTI system. In the considered estimation structure, each agent of the formation carries an estimate of the entire formation state. Agents of the formation can communicate information between each other through unidirectional links modeled with a fixed or a stochastic communication topology. The design procedures are based on a set of convex optimization problems with linear matrix inequalities and result in the suboptimal choice of estimator gains which stabilize the estimation error dynamics and minimize a norm of the estimation error correlation matrix.

Model validation for dynamically uncertain systems

Robust control models describe system uncertainty with both unknown additive signals and unknown dynamic perturbations. These unknown but bounded components lead to a model set description. Model validation is the experimental assessment of the ability of this model set to describe the observed system behaviors. In this paper we consider model validation for H∞ compatible models. This paper provides a detailed presentation of the H∞ model validation problem in the discrete frequency, discrete-time, and sampled-data frameworks. In each case the underlying results and the computational algorithms are discussed. The experimental applicability and the computational consequences are discussed in sufficient detail to give the reader an appreciation of the issues surrounding each model/experiment framework.