An application of an analog computer to solve the two-point boundary-value problem for a fourth-order optimal control problem

Abstract

The application of Pontryagin's maximum principle to trajectory optimization problems results in a two-point boundary-value problem. Computationally, this problem is solved by various digital techniques which are sometimes inconvenient and costly. An analog computer can be used to solve a large class of two-point boundary-value problems if the accuracy is acceptable. In this paper, an analog computer is used in conjunction with a human operator who has a display of the phase planes of the admissible trajectories. The human operator, having a general knowledge of the behavior of the system, adjusts control law parameters until the boundary conditions of the system are satisfied. Apparently this technique has been avoided previously due to the impression that unacceptable errors would be introduced in solving the problem. This technique was applied to a time-optimal rendezvous with bounds on rocket thrust and fuel available and demonstrated that accurate analog computer solutions are possible. Solutions of the rendezvous problem were compared with an exact solution using MIMIC.