Differential evolution/particle swarm optimizer for constrained slew maneuvers

摘要

This work examines a strategy for combining two heuristic methods, Differential Evolution (DE) and Particle Swarm Optimization (PSO) to better address the problem of stagnation when solving time-optimal slew-maneuver problems with path constraints. Several approaches are considered, including the use of only PSO or DE, and use of PSO with DE applied to some iterations. Applying a newer form of DE, in which random fractions of the vector-difference components are employed, stagnation is not evident, but the computational time is greater. The combined PSO/DE method handles the stagnation very well, with a 40% reduction in computation time compared with just DE.