Multiple Dynamic Rendezvous – A Benchmark Problem for Hybrid Optimal Control
- Paper number
IAC-07-C1.I.06
- Author
Dr. Paul Williams, Australia
- Year
2007
- Abstract
Hybrid optimal control problems are very important for planning future astrodynamics missions that require visits to multiple planets/asteroids, as well as a variety of other uses. The central characteristic of a hybrid optimal control problem as opposed to a conventional optimal control problem is that they involve a combination of discrete and continuous variables. In the aforementioned asteroid planning problem, the particular asteroids to be visited must be picked and the order in which to visit them determined, in conjunction with the optimal means for doing so. The optimal trajectory is found as the sequence of asteroids that minimizes the fuel expenditure, or ratio of fuel expended to time taken. The sheer number of asteroids makes any direct search impractical. Furthermore, reliance on ephemeris data complicates the procedure. In this paper, a benchmark problem that captures the core characteristics of the asteroid planning problem is posed. This problem considers two powered vehicles moving under acceleration constraints in three dimensions. The vehicles are allowed to maneuver only within a prescribed rectangular prism, and are subject to minimum (collision avoidance) and maximum (communications range) separation constraints. The goal of the optimization problem is to maneuver the two vehicles to rendezvous with a certain number of moving targets out of a larger collection of targets. The precise targets to be visited are not specified. The vehicles must return to their respective starting points. In this paper, an algorithm that combines genetic algorithms with nonlinear programming is proposed to solve this problem. The results will serve as the basis for comparisons of future algorithms for solving these types of problems, and will be very useful to the astrodynamics community.
- Abstract document
- Manuscript document
IAC-07-C1.I.06.pdf (🔒 authorized access only).
To get the manuscript, please contact IAF Secretariat.