An efficient Hybrid global optimizer for accurate modeling of manipulator dynamics in minimum-time trajectory-planning problems

Abstract

This research focuses on examining the minimum time trajectory planning problems for a robot. With the model of manipulator dynamics, we use a two-segment proportional-integration model that includes the squared jerk and trajectory integration to provide an overall execution time (including integration time) and the squared jerk segment as an approximation for the acceleration derivative. The augmented Lagrange multiplier (ALM) technique, which adds Lagrange multipliers, is used to help maximise the functional goal. The new hybrid global optimizer that has been created for each of the initial processes in the manipulator dynamics prevents a localised effective value from forming because a decade-long history of past effective values is saved and reused during the iterative analysis process, keeping the previous decade's effective value constant. Simulation findings show that the suggested method works for a three-DOF planar robot, and a three-degree-of-freedom (3-DOF) robot will allow for time-optimal trajectory planning.