An output feedback structured model reference adaptive control law has been developed for spacecraft rendezvous and docking problems. The effect of bounded errors on controller performance is studied in detail. Output can represent an aggregation sensor calibration errors, systematic bias, or some stochastic disturbances present any real measurements state estimates. the laws stable, tracking relative position attitude trajectories evaluated, considering unmodeled external as well parametric...

A Gaussian-mixture-model approach is proposed for accurate uncertainty propagation through a general nonlinear system. The transition probability density function approximated by finite sum of Gaussian functions which the parameters (mean and covariance) are propagated using linear theory. Two different approaches introduced to update weights components Gaussian-mixture model first method updates such that they minimize integral square difference between true forecast its Gaussian-sum...

A nonlinear filter is developed by representing the state probability density function a finite sum of Gaussian kernels whose mean and covariance are propagated from one time-step to next using linear system theory methods such as extended Kalman or unscented filter. The novelty in proposed method that weights updated at every time-step, solving convex optimization problem posed requiring approximation satisfy Fokker-Planck-Kolmogorov equation for continuous-time dynamical systems...

Reinforced concrete is subjected to deterioration due aging, increased load, and natural hazards. To minimize the maintenance costs increase operation lifetime, researchers practitioners are increasingly interested in improving current nondestructive evaluation technologies or building advanced structural health monitoring strategies. Acoustic emission methods offer an attractive solution for evaluation/structural of reinforced structures. In particular, development cracks large interest...

This paper presents an attitude estimation method from uncertain observations of inertial sensors, which is highly robust against different uncertainties. The proposed covariance inflated multiplicative extended Kalman filter (CI-MEKF) takes the advantage non-singularity in MEKF as well a novel inflation (CI) approach to fuse inconsistent information. CI compensates undesired effect magnetic distortion and body acceleration (as inherent biases magnetometer accelerometer sensors data,...

Two new recursive approaches have been developed to provide accurate estimates for posterior moments of both parameters and system states while making use the generalized polynomial-chaos framework uncertainty propagation. The main idea method is expand random state input parameter variables involved in a stochastic differential/difference equation polynomial expansion. These polynomials are associated with prior probability density function parameters. Later, Galerkin projection used obtain...

The main objective of this paper is to present the development computational methodology, based on Gaussian mixture model, that enables accurate propagation probability density function through mathematical models for orbit propagation. key idea approximate associated with states by a sum kernels. unscented transformation used propagate each kernel locally nonlinear dynamical models. Furthermore, convex optimization problem formulated forcing model approximation satisfy Kolmogorov equation...

This paper presents a methodology to estimate the probability of conjunction between two space objects when density functions for orbital state vector are significantly non-Gaussian in nature. The proposed method makes use recently developed conjugate unscented transformation with principle maximum entropy collision objects. points used efficiently propagate statistical moments corresponding each object, which then reconstruct miss-distance function by entropy. is integrated over required...

This paper presents a computationally efficient approach to evaluate multidimensional expectation integrals. Specifically, certain nonproduct cubature points are constructed that exploit the symmetric structure of Gaussian and uniform density functions. The proposed can be used as an alternative Gauss–Hermite (GH) Gauss–Legendre quadrature rules, but with significantly fewer number while maintaining same order accuracy when integrating polynomial functions in space. advantage newly developed...

An approach for nonlinear propagation of orbit uncertainties is discussed while making use the Fokker-Planck-Kolmogorov Equation (FPKE). The central idea to replace evolution initial conditions a dynamical system with probability density function (pdf) state variables. transition pdf corresponding vector approximated by using nite Gaussian mixture model. mean and covariance dierent components model are propagated through an Unscented Kalman Filter (UKF). Furthermore, unknown amplitudes found...

This paper presents an extension to the unscented transformation evaluate expectation integrals in general N-dimensional space by satisfying higher order moment equations. New sets of sigma points are defined satisfy equations up eighth order. The proposed methodology can be used as efficient alternative Gaussian quadrature rule with significantly reduced number function evaluations but without any loss accuracy. Numerical simulation results illustrates effectiveness computing high dimension...

An optimal linear attitude estimator is presented for the case of a single-point real-time estimation spacecraft using minimum-element parameterization: Rodrigues (or Gibbs) vector g. The optimality criterion, which does not coincide with Wahba's constrained rigorously quadratic and unconstrained. singularity, occurs when principal angle π, can easily be avoided by one rotation. accuracy tests show that proposed method provides precision comparable those fully complying Wahba definition....

In contrast to traditional wireless sensor network (WSN) applications that perform only data collection and aggregation, the new generation of information processing such as pursuit-evasion games, tracking, evacuation, disaster relief require in-network storage querying. Due resource limitations WSNs, it is challenging implement querying in a distributed, lightweight, resilient, energy-efficient manner. We address these challenges by exploiting location geometry propose an framework, namely,...

This paper discusses the development of a computationally efficient approach to generate optimal feedback control laws for infinite time problems by solving corresponding Hamilton–Jacobi–Bellman (HJB) equation. The solution process consists iteratively linear generalized HJB (GHJB) equation starting with an admissible stable controller. collocation methods are exploited solve GHJB in specified domain interest. Recently developed nonproduct quadrature method known as Conjugate Unscented...

This paper focuses on the problem of managing or tasking a network sensors to accurately track number objects while using information theoretic sensor performance metrics. The mathematical formulation optimally group mutual as utility measure is discussed along with relative merits maximizing information. resulting sensor-tasking optimization shown be combinatorial in nature, for which computational complexity increases an increase well sensors. Depending upon and available sensors,...

Abstract Uncertainty in predictions from a model of volcanic ash transport the atmosphere arises uncertainty both eruption source parameters and wind field. In previous contribution, we analyzed probability cloud presence using weighted samples dispersal runs reanalysis field to propagate alone. this probabilistic modeling is extended by ensemble forecast fields as well uncertain parameters. The impact on variability due unresolved scales motion physics also explored. We have therefore...

This paper presents a few novel quadrature rules to evaluate expectation integrals with respect uniform probability density function. In 1-dimensional the most widely used numerical method is Gauss-Legendre quadratures as they are exact for polynomials. As generic N-dimensional integral, tensor product of results in an undesirable exponential growth number points. The cubature proposed this can be direct alternative also designed exactly polynomials but use only small fraction addition, have...

Direction-dependent scaling, shaping, and rotation of Gaussian basis functions are introduced for maximal trend sensing with minimal parameter representations input output approximation. It is shown that shaping the radial helps in reducing total number function units required to approximate any given input-output data, while improving accuracy. Several alternate formulations enforce parameterization most general presented. A novel "directed graph" based algorithm facilitate intelligent...