Optimal control approximations for trainable manipulators

A theoretical procedure is developed for testing the quality of the approximation from the optimal solution of a nonlinear optimal control problem. It is based on the monotonicity between the changes of the Hamiltonian and the value functions proposed by Rekasius, and may provide a procedure for selecting effective controls for non-linear systems. The approach has been applied to the approximately optimal control of a trainable manipulator with seven degrees of freedom, where the controller is used for motion coordination and optimal execution of object-handling tasks.