Particle Swarm Optimization Applied to Finite Thrust Transfers between Two Circular Non-coplanar Orbits

Open Access
- Author:
- Flanagan, Peter Thomas
- Area of Honors:
- Aerospace Engineering
- Degree:
- Bachelor of Science
- Document Type:
- Thesis
- Thesis Supervisors:
- Robert Graham Melton, Thesis Supervisor
Robert Graham Melton, Thesis Honors Advisor
David Bradley Spencer, Faculty Reader
Dr. George A Lesieutre, Faculty Reader - Keywords:
- Particle
Swarm
Optimization
finite
thrust
non-coplanar
circular
orbits
swarm
intelligence
scheme
space
trajectories - Abstract:
- As a swarm intelligence scheme, the Particle Swarm Optimization (PSO) technique is a stochastic population-based method, representing an intuitive methodology for global optimization and has been successfully applied to several fields of research. Through mimicking the unpredictable motion of bird flocks in search of food, PSO uses the mechanism of information sharing that affects the overall behavior of a swarm to converge to the optimal values of the unknown parameters for the problem under consideration. For this research, PSO was used to optimize the finite thrust transfers of a spacecraft between two circular orbits that are not coplanar. The transfer trajectory consists of two thrusting arcs separated by a coasting arc. For better performance, the plane change was incorporated in the second thrusting maneuver. The dynamics of the system depend of the twelve coefficients from three cubic polynomials used to represent the in-plane and out-of-plane thrust pointing angles as well as the three time intervals corresponding to the three arcs of trajectory. Using MATLAB, the PSO algorithm will determine these fifteen parameters as the solution converges to the global optimal solution, minimizing the objective function, which corresponds to minimizing propellant consumption. The algorithm consists of eight functions, using ode45 to numerically integrate the state equations for each thrusting arc. Several tests were conducted on the PSO algorithm to analyze the convergence to the global minimum including varying the swarm parameters and the ratio of outer to inner radii values, β. Sometimes, the algorithm converged on a local minimum as the solution. Further research will attempt to correct the issue of local convergence, in hopes of consistently obtaining the global minimum.