In this thesis, a theory for optimal path planning of general three dimensional mechanical systems is developed to minimize operation time of the systems that are driven to pass a specified path.
In the optimization problem formulation, the operation time is taken as cost function to be minimized, system kinematic equations and equation of motion with Lagrange Multipliers as state equations, time required to pass each interval of the path as design variables, and maximum actuator capacities and time histories of actuator torques as design constraints.
The time history constraints are imposed to consider retarded responses of actual actuators.
An adjoint variable technique is utilized to perform the first order design sensitivity analysis of the functions in the problem and the system is optimized through the gradient algorithm.
A computer program based on the theory is developed and tested with an example of a six degrees oof freedom industrial robot. It is found that the suggested method is stable and minimization of ooperation time is achieved.
