Martin Norsell

Multistage trajectory optimization with radar-range constraints

The problem of finding an optimal aircraft trajectory for long-distance flights in three dimensions subject to radar detection constraints is considered. A general point-mass model previously developed is not suitable because the time discretization needs to be very fine to resolve the rigid-body dynamics resulting in very large optimization problems. Different reduced mathematical models are derived and compared to the more general performance model. Finally, a long-range mission involving two subsonic jet trainers approaching a radar target is numerically simulated. The results indicate that the time a hostile radar has an approaching aircraft under surveillance can be greatly reduced by using the proposed methodology.

On Modeling RCS of Aircraft for Flight Simulation

This paper investigates the implementation of the radar cross section (RCS) of aircraft in modeling and simulation (M&S). More specifically, it addresses the tradeoff between accuracy and computational cost introduced by spatial RCS fluctuations. High-resolution RCS matrices, generated using Physical Optics (PO), were used in an investigation of RCS matrix resolution, and an evaluation of different bilinear interpolation methods is presented. The spatial Fourier transform was used for resolution analysis. It was found that the smallest RCS interpolation error was obtained using splines. Furthermore, results showed that the distribution of the relative interpolation error in detection range was well approximated by a log-normal distribution.

Conservative RCS Models for Tactical Simulation

This paper describes a procedure for generating conservative radar cross section (RCS) models able to meet the computational requirements imposed by simulation and related applications. The key concept is to downsample calculated or measured RCS data retaining local extreme values; thus, a conservative RCS matrix is obtained. Spline approximations are used in order to obtain continuity in the RCS models. RCS models with varying resolution have been generated and analyzed, and it is shown how spatial Fourier transforms can be used when determining feasibility for certain decision making applications. Furthermore, it is found that the interpolation errors obtained from the conservative RCS models are well described by generalized extreme value theory.

Reduction of RCS samples using the cubed sphere sampling scheme

An alternative to the traditional method of sampling radar cross section data from measurements or electromagnetic code is presented and evaluated. The Cubed Sphere sampling scheme solves the probl ...