This paper applies interval optimization to the fixed-time multiple impulse rendezvous and docking problem. Current methods for solving this type of optimization problem include for example genetic algorithms and gradient based optimization. Unlike these methods, interval methods can guarantee that the globally best solution is found for a given parameterization of the input. The state transition matrix approach for the linearized CW-equations is used to avoid interval integration. Thruster pulse amplitudes are optimized by an interval branch and bound algorithm, which systematically eliminates parts of the control input space that do not satisfy the final time state constraints. Interval analysis is shown to be a useful tool in both sensitivity analysis and nonlinear optimization of the rendezvous and docking problem.
展开▼
