Nato RL - Academia.edu (original) (raw)

Papers by Nato RL

An adaptive LQG (Linear-Quadratic-Gaussian) self-tuning controller for ARMAX (AutoRegressive Movi... more An adaptive LQG (Linear-Quadratic-Gaussian) self-tuning controller for ARMAX (AutoRegressive Moving Average) plants is developed. Its feedback updating scheme is based upon an RLS (Recursive Least Squares) parameter estimation of T+1 independent output predictors, T??n, n being the plant order. The seemingly heavy numerical burden is actually limited by the common updating gain shared by all estimates in the independent predictors. The proposed regulator turns out to be a refinement of the MUSMAR self tuner that has proved to posses attractive properties of robustness.

Physica Status Solidi B-basic Solid State Physics, 2001

Subject classification: 71.35.Cc; 73.21.La

Ocean Modelling, 2006

A primitive equation model and a statistical predictor are coupled by data assimilation in order ... more A primitive equation model and a statistical predictor are coupled by data assimilation in order to combine the strength of both approaches. In this work, the system of two-way nested models centred in the Ligurian Sea and the satellite-based ocean forecasting (SOFT) system predicting the sea surface temperature (SST) are used. The data assimilation scheme is a simplified reduced order Kalman filter based on a constant error space. The assimilation of predicted SST improves the forecast of the hydrodynamic model compared to the forecast obtained by assimilating past SST observations used by the statistical predictor. This study shows that the SST of the SOFT predictor can be used to correct atmospheric heat fluxes. Traditionally this is done by relaxing the model SST towards the climatological SST. Therefore, the assimilation of SOFT SST and climatological SST are also compared. (A. Barth). Ocean Modelling 13 (2006) 255-270 www.elsevier.com/locate/ocemod models with different imperfections (and different strengths). Data assimilation provides the statistical frame for merging different model results. Statistical predictors have been applied in ocean forecasting as an alternative or complementary approach to physical models. Various applications of statistical predictors have been tested and numerous implementations have been applied to the El Niñ o Southern Oscillation (ENSO). To predict the state of the system the analogue prediction methods (e.g. Sugihara and May, 1990) look for the similar past states, the so-called analogues. The evolution is inferred from the evolution of the analogues. Based on analogues, Drosdowsky (1994) predicted the Southern Oscillation Index time series. Van den Dool (1994) used constructed analogues to forecast surface weather over the US from monthly 700 mb height. The approach of Keppenne and Ghil (1992) relies on a maximum entropy method to predict the Southern Oscillation Index. A linear inverse model was also used to predict sea surface temperature (SST) anomalies in the tropical Indo-Pacific region . proposed an empirical orthogonal function (EOF) iteration scheme to predict El Niño 3 SST anomalies and the Southern Oscillation Index. Another class of statistical predictor is based on a canonical correlation analysis. This method was also used to predict the 3-month mean SST in several regions of the tropical Pacific and Indian oceans (Barnston and Ropelewski, 1992), total precipitation for island stations in the tropical Pacific (He and and temperature and precipitation in Hawaii and Alaska . developed an ENSO statistical prediction method, which is based on the optimal combination of persistence, month-to-month trend of initial conditions, and climatology.

IEEE Journal of Oceanic Engineering, 2004

A new coherent reverberation model developed at the Naval Research Laboratory, Washington, DC, an... more A new coherent reverberation model developed at the Naval Research Laboratory, Washington, DC, and the Supreme Allied Commander Atlantic is exercised in the 17-750-Hz band to estimate the degree of non-Rayleighness of shallow-water reverberation envelopes as a function of waveguide multipath, system bandwidth, directivity, and frequency. Findings suggest that reverberation from diffuse, but non-Gaussian, scatterer distributions is significantly more Rayleigh for multipath environments than for equivalent environments excited by a single or small number of modes or for broadside receiver array processing that extracts narrow angles of reception. These findings suggest that the problem of non-Rayleigh reverberation in shallow-water waveguides can be ameliorated through the use of tuned ensonification and reception schemes, which retain high probabilities of detection while reducing the associated probability of false alarm.

Journal of The Acoustical Society of America, 2003

The BiStaR bistatic reverberation model has been benchmarked against analytic formulas for bistat... more The BiStaR bistatic reverberation model has been benchmarked against analytic formulas for bistatic reverberation in isovelocity range dependent waveguides. Studies conducted using BiStaR to evaluate the effects of introducing refraction into the benchmark scenarios indicate that downward refraction can have a tremendous effect on the time-angle evolution of mono- and bistatic reverberation, especially in shoaling directions. Here recent advances in analytic formulas for propagation and reverberation in refractive range dependent waveguides are exercised to evaluate analytically the predicted changes in RL caused by the different propagation characteristics. The analytic results are compared with the numerical results and conclusions are drawn regarding the propagation regimes and mechanisms that cause the strong deviations from the isovelocity predictions. [Work supported by ONR and SACLANTCEN.]

We provide an explanation for the poor performance of the minimum-variance distortionless-respons... more We provide an explanation for the poor performance of the minimum-variance distortionless-response (MVDR) beam former when applied to one particular technique for geoacoustic inversion of marine ambient noise. Previous studies have shown that using the Bartlett beam former one can estimate the reflection loss (RL) as a function of frequency and grazing angle, and use it to estimate the seabed geoacoustic parameters. Despite its superior performance in direction-of-arrival estimation, the MVDR beam former's results for RL estimation are less reliable than those yielded by the Bartlett beam former. One important difference between the two beam formers lies in the fact that the MVDR beam patterns for the two look directions ±φ can be markedly different, whereas the Bartlett beam patterns are always symmetric with respect to the broadside look direction. By estimating the RL from simulated acoustic fields we first present a method of quantifying and visualizing the asymmetry, and then show that it is responsible for the anomalies in the MVDR-based RL estimate.

An attempt to reconcily performance and robustness in adaptive control is discussed in the framew... more An attempt to reconcily performance and robustness in adaptive control is discussed in the framework of algorithms with moderate complexity. The basic idea is to recognize that ARMAX plants, when properly controlled, can be described by ARX models. Thus, standard RLS identification can be used. In addition, instead of relying on a single estimated plant predictor, a set of separately

An adaptive LQG (Linear-Quadratic-Gaussian) self-tuning controller for ARMAX (AutoRegressive Movi... more An adaptive LQG (Linear-Quadratic-Gaussian) self-tuning controller for ARMAX (AutoRegressive Moving Average) plants is developed. Its feedback updating scheme is based upon an RLS (Recursive Least Squares) parameter estimation of T+1 independent output predictors, T??n, n being the plant order. The seemingly heavy numerical burden is actually limited by the common updating gain shared by all estimates in the independent predictors. The proposed regulator turns out to be a refinement of the MUSMAR self tuner that has proved to posses attractive properties of robustness.

Physica Status Solidi B-basic Solid State Physics, 2001

Subject classification: 71.35.Cc; 73.21.La

Ocean Modelling, 2006

A primitive equation model and a statistical predictor are coupled by data assimilation in order ... more A primitive equation model and a statistical predictor are coupled by data assimilation in order to combine the strength of both approaches. In this work, the system of two-way nested models centred in the Ligurian Sea and the satellite-based ocean forecasting (SOFT) system predicting the sea surface temperature (SST) are used. The data assimilation scheme is a simplified reduced order Kalman filter based on a constant error space. The assimilation of predicted SST improves the forecast of the hydrodynamic model compared to the forecast obtained by assimilating past SST observations used by the statistical predictor. This study shows that the SST of the SOFT predictor can be used to correct atmospheric heat fluxes. Traditionally this is done by relaxing the model SST towards the climatological SST. Therefore, the assimilation of SOFT SST and climatological SST are also compared. (A. Barth). Ocean Modelling 13 (2006) 255-270 www.elsevier.com/locate/ocemod models with different imperfections (and different strengths). Data assimilation provides the statistical frame for merging different model results. Statistical predictors have been applied in ocean forecasting as an alternative or complementary approach to physical models. Various applications of statistical predictors have been tested and numerous implementations have been applied to the El Niñ o Southern Oscillation (ENSO). To predict the state of the system the analogue prediction methods (e.g. Sugihara and May, 1990) look for the similar past states, the so-called analogues. The evolution is inferred from the evolution of the analogues. Based on analogues, Drosdowsky (1994) predicted the Southern Oscillation Index time series. Van den Dool (1994) used constructed analogues to forecast surface weather over the US from monthly 700 mb height. The approach of Keppenne and Ghil (1992) relies on a maximum entropy method to predict the Southern Oscillation Index. A linear inverse model was also used to predict sea surface temperature (SST) anomalies in the tropical Indo-Pacific region . proposed an empirical orthogonal function (EOF) iteration scheme to predict El Niño 3 SST anomalies and the Southern Oscillation Index. Another class of statistical predictor is based on a canonical correlation analysis. This method was also used to predict the 3-month mean SST in several regions of the tropical Pacific and Indian oceans (Barnston and Ropelewski, 1992), total precipitation for island stations in the tropical Pacific (He and and temperature and precipitation in Hawaii and Alaska . developed an ENSO statistical prediction method, which is based on the optimal combination of persistence, month-to-month trend of initial conditions, and climatology.

IEEE Journal of Oceanic Engineering, 2004

A new coherent reverberation model developed at the Naval Research Laboratory, Washington, DC, an... more A new coherent reverberation model developed at the Naval Research Laboratory, Washington, DC, and the Supreme Allied Commander Atlantic is exercised in the 17-750-Hz band to estimate the degree of non-Rayleighness of shallow-water reverberation envelopes as a function of waveguide multipath, system bandwidth, directivity, and frequency. Findings suggest that reverberation from diffuse, but non-Gaussian, scatterer distributions is significantly more Rayleigh for multipath environments than for equivalent environments excited by a single or small number of modes or for broadside receiver array processing that extracts narrow angles of reception. These findings suggest that the problem of non-Rayleigh reverberation in shallow-water waveguides can be ameliorated through the use of tuned ensonification and reception schemes, which retain high probabilities of detection while reducing the associated probability of false alarm.

Journal of The Acoustical Society of America, 2003

The BiStaR bistatic reverberation model has been benchmarked against analytic formulas for bistat... more The BiStaR bistatic reverberation model has been benchmarked against analytic formulas for bistatic reverberation in isovelocity range dependent waveguides. Studies conducted using BiStaR to evaluate the effects of introducing refraction into the benchmark scenarios indicate that downward refraction can have a tremendous effect on the time-angle evolution of mono- and bistatic reverberation, especially in shoaling directions. Here recent advances in analytic formulas for propagation and reverberation in refractive range dependent waveguides are exercised to evaluate analytically the predicted changes in RL caused by the different propagation characteristics. The analytic results are compared with the numerical results and conclusions are drawn regarding the propagation regimes and mechanisms that cause the strong deviations from the isovelocity predictions. [Work supported by ONR and SACLANTCEN.]

We provide an explanation for the poor performance of the minimum-variance distortionless-respons... more We provide an explanation for the poor performance of the minimum-variance distortionless-response (MVDR) beam former when applied to one particular technique for geoacoustic inversion of marine ambient noise. Previous studies have shown that using the Bartlett beam former one can estimate the reflection loss (RL) as a function of frequency and grazing angle, and use it to estimate the seabed geoacoustic parameters. Despite its superior performance in direction-of-arrival estimation, the MVDR beam former's results for RL estimation are less reliable than those yielded by the Bartlett beam former. One important difference between the two beam formers lies in the fact that the MVDR beam patterns for the two look directions ±φ can be markedly different, whereas the Bartlett beam patterns are always symmetric with respect to the broadside look direction. By estimating the RL from simulated acoustic fields we first present a method of quantifying and visualizing the asymmetry, and then show that it is responsible for the anomalies in the MVDR-based RL estimate.

An attempt to reconcily performance and robustness in adaptive control is discussed in the framew... more An attempt to reconcily performance and robustness in adaptive control is discussed in the framework of algorithms with moderate complexity. The basic idea is to recognize that ARMAX plants, when properly controlled, can be described by ARX models. Thus, standard RLS identification can be used. In addition, instead of relying on a single estimated plant predictor, a set of separately