Engineering topics

Longman, Richard W.

Publications and source records attributed to Longman, Richard W..

45 records · Page 3

NASA NTRS1990

Optimization of actuator and sensor placement for on-orbit identification in large flexible spacecraft

There has been considerable research on choosing actuator and sensor locations in large flexible spacecraft in order to optimize the controllability and observability of the system, or to maximize some objective function of control system performance. Future large flexible spacecraft may require on-orbit identification of the structure to tune the control system, because such tests cannot be performed in a one-g environment before launch. This indicates that the choice of actuator and sensor locations must serve a dual purpose, for control and for identification. This paper develops concepts for a degree of identifiability and studies placement of actuators and sensors on a free-free beam to optimize such objective functions. The results in this simple situation suggest that in free-free spacecraft structures in orbit, placement for control and placement for identification may often be consistent objectives rather than conflicting objectives.

Bergmann, Martin↗

NASA NTRS1990

Learning control for minimizing a quadratic cost during repetitions of a task

In many applications, control systems are asked to perform the same task repeatedly. Learning control laws have been developed over the last few years that allow the controller to improve its performance each repetition, and to converge to zero error in tracking a desired trajectory. This paper generates a new type of learning control law that learns to minimize a quadratic cost function for tracking. Besides being of interest in its own right, this objective alleviates the need to specify a desired trajectory that can actually be performed by the system. The approach used here is to adapt appropriate methods from numerical optimization theory in order to produce learning control algorithms that adjust the system command from repetition to repetition in order to converge to the quadratic cost optimal trajectory.

Longman, Richard W.↗

NASA NTRS1990

Small gain robustness issues in the p-integrator repetitive controller

The basic theory of the digital p-integrator controller is utilized to modify the command to a stable analog feedback control system in order to produce zero tracking error of a repetitive command. Mapping of continuous time poles and zeros to the images in the discrete time domain is performed. It is shown that for a pole excess of one, with p odd, the p-integrator learning controller will often produce a stable learning process, even for relatively large sample times. For a pole excess of two, a stable learning process can often be achieved for any p as long as the sample time is kept sufficiently short. Therefore, when the pole excess is zero and one utilizes the proper p-integrator controller with p odd, stable performance is often produced, even for relatively large sample times.

Longman, Richard W.↗

NASA NTRS1990

Input/output system identification - Learning from repeated experiments

The paper describes three approaches and possible variations for the determination of the Markov parameters for forced response data using general inputs. It is shown that, when the parameters in the solution procedure are bootstrapped, the results can be obtained very efficiently, but the errors propagate throughout all parameters. By arranging the data in a different form and using singular value decomposition, the resulting identified parameters are more accurate, in the least number of successive experiments, at the expense of a large matrix singular value decomposition. When a recursive procedure is employed, the calculations can be performed very efficiently, but the number of repetitions of the experiments is much greater for a given accuracy than for any of the previous approaches. An alternative formulation is proposed to combine the advantages of each of the approaches.

Juang, Jer-Nan↗

NASA NTRS1989

Recursive form of the eigensystem realization algorithm for system identification

An algorithm is developed for recursively calculating the minimum realization of a linear system from sampled impulse response data. The Gram-Schmidt orthonormalization technique is used to generate an orthonormal basis for factorization of the data matrix. The system matrix thus identified is in upper Hessenberg form, which has advantages for the identification of modal parameters including damping coefficients, frequencies, mode shapes, and modal participation factors. It also has the property that once an element of the system matrix is computed, it is never altered as the dimension of the model is increased in the recursive process. Numerical examples are presented for comparison of the recursive and nonrecursive forms of the eigensystem realization algorithm.

Longman, Richard W.↗

NASA NTRS1989

Variance and bias computation for enhanced system identification

A study is made of the use of a series of variance and bias confidence criteria recently developed for the eigensystem realization algorithm (ERA) identification technique. The criteria are shown to be very effective, not only for indicating the accuracy of the identification results (especially in terms of confidence intervals), but also for helping the ERA user to obtain better results. They help determine the best sample interval, the true system order, how much data to use and whether to introduce gaps in the data used, what dimension Hankel matrix to use, and how to limit the bias or correct for bias in the estimates.

Bergmann, Martin↗

NASA NTRS1988

Variance and bias confidence criteria for ERA modal parameter identification

For the ERA system identification algorithm, perturbation methods are used to develop expressions for variance and bias of the identified modal parameters. Based on the statistics of the measurement noise, the variance results serve as confidence criteria by indicating how likely the true parameters are to lie within any chosen interval about their identified values. This replaces the use of expensive and time-consuming Monte Carlo computer runs to obtain similar information. The bias estimates help guide the ERA user in his choice of which data points to use and how much data to use in order to obtain the best results, performing the trade-off between the bias and scatter. Also, when the uncertainty in the bias is sufficiently small, the bias information can be used to correct the ERA results. In addition, expressions for the variance and bias of the singular values serve as tools to help the ERA user decide the proper modal order.

Longman, Richard W.↗

NASA NTRS1988

A mathematical theory of learning control for linear discrete multivariable systems

When tracking control systems are used in repetitive operations such as robots in various manufacturing processes, the controller will make the same errors repeatedly. Here consideration is given to learning controllers that look at the tracking errors in each repetition of the process and adjust the control to decrease these errors in the next repetition. A general formalism is developed for learning control of discrete-time (time-varying or time-invariant) linear multivariable systems. Methods of specifying a desired trajectory (such that the trajectory can actually be performed by the discrete system) are discussed, and learning controllers are developed. Stability criteria are obtained which are relatively easy to use to insure convergence of the learning process, and proper gain settings are discussed in light of measurement noise and system uncertainties.

Phan, Minh↗

NASA NTRS1988

A variance based confidence criterion for ERA identified modal parameters

The realization theory is developed in a systematic manner for the Eigensystem Realization Algorithm (ERA) used for system identification. First, perturbation results are obtained which describe the linearized changes in the identified parameters resulting from small change in the data. Formulas are then derived that can be used to evaluate the variance of each of the identified parameters, assuming that the noise level is sufficiently low to allow the application of linearized results. These variances can be converted to give confidence intervals for each of the parameters for any chosen confidence level.

Longman, Richard W.↗