D
Dennis S. Bernstein
Researcher at University of Michigan
Publications - 876
Citations - 29606
Dennis S. Bernstein is an academic researcher from University of Michigan. The author has contributed to research in topics: Adaptive control & Control theory. The author has an hindex of 70, co-authored 847 publications receiving 26704 citations. Previous affiliations of Dennis S. Bernstein include Northrop Grumman Corporation & Harris Corporation.
Papers
More filters
Journal ArticleDOI
Gradient-, Ensemble-, and Adjoint-Free Data-Driven Parameter Estimation
Ankit Goel,Dennis S. Bernstein +1 more
TL;DR: Nonlinear estimation methods, such as the extended Kalman filter, unscented Kalman filters, and ensemble Kalman Filter, can be used for parameter estimation by viewing the unknown parameters as cons as well as linear models.
Proceedings ArticleDOI
Adaptive tracking using ARMARKOV/Toeplitz models
TL;DR: In this article, an adaptive algorithm for the MIMO tracking problem is developed for the ARMARKOV/Toeplitz models, and the parameter matrix of the compensator is updated online by means of a gradient algorithm.
Proceedings ArticleDOI
An inner-loop/outer-loop architecture for an adaptive missile autopilot
TL;DR: This work uses retrospective cost adaptive control with a constant forgetting factor (CFF), variable forgetting factors (VFF), and Kalman Filter (KF) to control a planar missile with nonlinear dynamics and aerodynamics.
Journal ArticleDOI
Stable ?2-optimal controller synthesis
TL;DR: In this article, the authors considered a fixed-structure stable ℋ2-optimal controller synthesis using a multiobjective optimization technique which provides a trade-off between closed-loop performance and the degree of controller stability.
Proceedings ArticleDOI
Sliding window recursive quadratic optimization with variable regularization
TL;DR: In this article, a sliding-window variable-regularization recursive least squares algorithm is presented. But this algorithm does not support time-varying regularization in the weighting as well as what is being weighted.