Guidance, navigation, and control for satellite formation flying using differential drag

S. Chocron, D. Choukrouny

Research output: Contribution to conferencePaperpeer-review

Abstract

This work is concerned with the development of relative navigation algorithms for formation ying of small satellites via differential drag only along low Earth orbits and their integration within a guidance navigation and attitude control and determination architecture (GNC/ADC). The design includes realistic features such as high variability in the air density, attitude control for ballistic coefficient modulation, attitude determination via rate gyroscopes and vector measurements, and several relative navigation filters using relative position sensing and various air density models. This GNC/ADC algorithm is tested in a high integrity simulation environment. It enables closure of the range between two nanosatellites from an initial 7.5 km down to 10 m within eight orbits at an altitude similar to the International Space Station. Degradations in the performances, as compared to an ideal guidance scheme, are due to the potential high variability of the air density, the relative navigation errors, and the attitude control errors, in order of dominance. The best relative navigation filter appears to be a robust H∞ filter. A comparison of that filter with several Kalman filters matched to various air density models shows quicker convergence, lesser sensitivity to jumps in the air density, and a similar steady-state accuracy, albeit with a noisier behavior.

Original languageEnglish
Pages835-858
Number of pages24
StatePublished - 1 Jan 2018
Event58th Israel Annual Conference on Aerospace Sciences, IACAS 2018 - Tel-Aviv and Haifa, Israel
Duration: 14 Mar 201815 Mar 2018

Conference

Conference58th Israel Annual Conference on Aerospace Sciences, IACAS 2018
Country/TerritoryIsrael
CityTel-Aviv and Haifa
Period14/03/1815/03/18

Fingerprint

Dive into the research topics of 'Guidance, navigation, and control for satellite formation flying using differential drag'. Together they form a unique fingerprint.

Cite this