METHOD OF CHARACTERISTICS BASED NONLINEAR FILTER: APPLICATIONS TO SPACE OBJECT TRACKING

2018 Univelt Inc. All rights reserved. This paper deals with the development of a computationally efficient nonlinear filter that implicitly propagates the state probability density function for dynamical systems with no process noise. The proposed method leverages the method of characteristics to propagate probability density values of the state probability density function along the characteristic solutions, thereby avoiding the explicit propagation of the full state probability density function at all time steps. Further, this paper proposes a convex optimization procedure to reconstruct the state probability density function from these characteristic solutions. Numerical examples of capturing the non-Gaussian nature of the uncertainty in the two body problem with uncertain initial conditions is used to illustrate the proposed methods.

METHOD OF CHARACTERISTICS BASED NONLINEAR FILTER: APPLICATIONS TO SPACE OBJECT TRACKING Conference Paper