Optimal Formation Reconfiguration Using Genetic Algorithms

Abstract

Spacecraft formation flying has received an enormous amount of attention over the past few years, researcher have begun to consider the advantages of spacecraft formation compared to large, complex, single purpose spacecrafts, the use of spacecraft formation has the potential to expand functionality, distribute risk and reduce cost and also it involves tremendous challenges ranging from spacecraft formation keeping to reconfiguration or configuration. The spacecrafts in orbit may undergo differential disturbances from gravitational perturbation due to Earth's oblateness, atmospheric drag, and solar radiation pressure and so on, consequently these effects have to be taken into account in order to implement formation reconfiguration. Whereas the consumption of fuel on board is one of the key formation reconfiguration problems, it is important to have a formation with saving fuel for orbit maintenance, whilst collision avoidance is also a significant constraint condition on formation reconfiguration. In order to deal with these challenges, many methods have been considered. With a view to tackle these problems, the present study is carried out. The nonlinear equations of relative motion on formation flying are investigated and the same geometry and mass of spacecrafts are assumed in this present paper. In this model the disturbances coming from the non spherical Earth is more important than others and then considered. In the present study, optimal controllers based on genetic algorithms are developed for spacecraft formation reconfiguration including the constraints of minimum fuel, avoiding collision and final configuration. Genetic algorithms (GA) are especially powerful techniques for the research of optimal control, GA are such methods that may be used to solve search and optimization problems. They are based on the genetic evolution process of biological organisms, over many generations, natural populations evolved according to the principles of natural selection, by mimicking this process, GA are able to evolve solutions to real world problems. The numerical results were be analyzed that demonstrated the good performance of the control strategy proposed for reconfiguration. The method proposed may be viable for future space mission.