Junction times in singular optimal control

摘要

Finding junction times on partially singular optimal control trajectories is in general difficult analytically and especially so when the optimal control is explicitly dependent on both the state and adjoint variables. This paper presents a new approach for linear problems whereby variations in the final state, measured from a reference optimal solution, can be used to find corresponding junction times. The approach is shown to be effective when applied to a problem which has an optimal solution involving a single bang/singular junction for certain choices of end state.