Comparison of dispersion calculation methods for sounding rockets

Abstract

Sounding rocket's flight safety is a major issue when planning its flight campaign. While flight procedures for conventional sounding rockets with solid motors are well-known and tested, non-conventional designs that have begun test flights have not been assessed adequately. These new designs include liquid or hybrid motors and a parallel staging. With an additional parameter in pre-launch procedures in addition to elevation and azimuth – propellant mass – a new dimension is introduced to calculations, which increases computation time. This motivates to analyze alternatives to a large-sample Monte Carlo analysis which is the current state-of-the-art. This paper presents a short literature review of historical and current methods for calculating a landing point dispersion for sounding rockets. It follows with a description of two methods that should provide faster dispersion calculations without compromising on safety - at a cost of the dispersion estimate being larger. The first method is the root sum square, which is already present in Federal Aviation Administration regulations, and the other is to calculate dispersion using Monte Carlo with fewer iterations, and scaling the solution up to make up for an additional uncertainty. Methodologies proposed to compare any pair of dispersions mathematically are discussed. The paper describes the design of numerical testing of proposed methods, which is based on a simulation model of the ILR-33 AMBER 2K rocket. Simulation results allowed to verify dispersion calculation methods for a sounding rocket with a non-conventional configuration, and show that given an adequate margin, the Monte Carlo method can be replaced by the two aforementioned methods to save significant computing time.