Radhakisan S. Baheti

Efficient Approximation of Kalman Filter for Target Tracking

A Kalman filter in the Cartesian coordinates is described for a maneuvering target when the radar sensor measures range, bearing, and elevation angles in the polar coordinates at high data rates. An approximate gain computation algorithm is developed to determine the filter gains for on-line microprocessor implementation. In this approach, gains are computed for three uncoupled filters and multiplied by a Jacobian transformation determined from the measured target position and orientation. The algorithm is compared with the extended Kalman filter for a typical target trajectory in a naval gun fire control system. The filter gains and the tracking errors for the proposed algorithm are nearly identical to the extended Kalman filter, while the computation requirements are reduced by a factor of four.

Mapping extended Kalman filters onto linear arrays

Techniques for mapping extended Kalman filters onto linear arrays of programmable cells designed for real-time applications are described. First, a general method for mapping a standard (nonsquare root) Kalman filter, where the columns of the covariance matrix are updated in parallel, is introduced. Next, a general method for mapping a factorized (square root) filter, where fast Givens rotations are used to triangularize the prematrix and where rotations of the rows of the prematrix are performed in parallel, is introduced. These mappings are used to implement an extended Kalman filter commonly used in target tracking applications on the Warp computer. The Warp is a commercially available linear array of 10 or more programmable cells connected to an MC68020-based workstation. The Warp implementation of the standard Kalman filter running on 8 Warp cells achieves a measured speedup of 7 over the same filter running on a single cell. The Warp implementation of the factorized filter running on 10 Warp cells achieves a measured speedup of 2. >

Vision processing and control of robotic Arc welding system

A microprocessor-based control system is presented for a Gas Tungsten Arc Welding (GTAW) process to join thin sheet metal parts. The system uses a welding robot, a vision sensor, and an image processor to control the welding torch in real-time. A vision-processing algorithm is developed to compute weld puddle geometry parameters from the noisy image of the molten pool. The weld quality is controlled by regulation of puddle area, puddle width, arc length, and puddle center over the joint. Experimental results indicate successful operation of the control system with heat sink disturbances in the weld fixture.

A new cross correlation algorithm for Volterra kernel estimation of bilinear systems

Correlation analysis is applied to estimate the first- and second-order kernels in a Volterra series expansion of bilinear systems. The kernels are estimated for a simulation model of a nuclear fission process. The method yields good estimates of the first-order kernel under noisy input-output measurements. However, the estimation of the second-order kernel is not satisfactory, due to the presence of higher order Volterra kernels. A new algorithm is developed to identify the parameter matrix B which characterizes the nonlinear part of a bilinear system. The estimation of the second-order kernel is significantly improved.

Fast Mapping Of A Kalman Filter On Warp

A parallel algorithm for solving an n-state Kalman filter on an (n+2)-cell linear array is described. The algorithm is the basis for the mapping of a 9-state target tracking filter on the Warp computer. The Warp implementation is written in a high-level language and achieves a measured speedup of almost 300 over the same filter running on a Sun workstation.

Multivariable frequency domain controller for magnetic suspension and balance systems

The magnetic suspension and balance system for an airplane model in a large wind tunnel is considered. In this system, superconducting coils generate magnetic forces and torques on the magnetized soft iron core of the airplane model. The control system is a position servo where the airplane model, with six degrees of freedom, follows the reference static or dynamic input commands. The controller design, based on the characteristic loci method, minimizes the effects of aerodynamic and inertial cross-couplings, and provides the specified dynamic response.

Adaptive control and calibration of heliostats

A large field of heliostats (steerable mirrors) are oriented to reflect light to a central receiver in solar/electric energy conversion. To achieve high pointing accuracy, each heliostat is calibrated for errors in the heliostat drives and installation. The sun is used as a precision position reference and the heliostat pointing errors are estimated at periodic intervals. A least-squares algorithm is used to estimate the model parameters. Simulation results are given for a typical heliostat configuration with realistic errors. Effectiveness of the error model is analyzed for different mirror pedestal tilts, azimuth and elevation biases and drive wheel radius tolerances.

Second-order correlation method for bilinear system identification

An algorithm is developed to estimate parameters of a class of discrete-time bilinear systems using second-order correlations. The algorithm is simplified for a pseudorandom binary input signal, and least-squares method is used as a second step to improve the estimates. The method is compared with least-squares and maximum-likelihood estimation algorithms for a model of nuclear fission.

A sub-optimal Kalman filter design for target tracking applications

A Kalman filter in the cartesian coordinates is described for a maneuvering target when the radar sensor measures range, bearing, and elevation angles in the polar coordinates at high data rates. An approximate gain computation algorithm is developed to determine the filter gains for on-line microprocessor implementation. In this approach, gains are computed for three uncoupled filters and multiplied by a Jacobian transformation determined from the measured target position and orientation. The algorithm is compared with the extended-Kalman filter for a typical target trajectory in a naval gun fire control system. The filter gains and the tracking errors for the proposed algorithm are nearly identical to the extended-Kalman filter, while the computation requirements can be reduced by a factor of four.

Rapid tuning of a fuel gas saturator control system using Nyquist and bode diagrams

A multivariable controller is designed and implemented for a saturator in a fuel gas process evaluation facility for coal gasification based power plants. The saturator model developed from first principles is represented by a high-order nonlinear state variable dynamic system. The controller is designed so that the closed-loop frequency response is as close as possible, in a least-squares sense, to a desired response. The design method allows the use of a simple controller structure and minimizes interaction in the critical frequency range where the characteristic gains pass near the critical points. The test results describe the operational performance of the control system and offer a practical approach for coordinated control of heat transfer components in a power plant.