Energy minimization of redundant manipulators using generalized pattern search

Optimal motion planning is very important in the operation of robot manipulators. Its main target is the generation of a trajectory from start to goal that satisfies objectives, such as minimizing path traveling distance or time interval, lowest energy consumption or obstacle avoidance and satisfying the robot's kinematics and dynamics. Due to increasing demand on fuel consumption and the uprising political crisis, there is a need to optimize the energy consumption for stable and durable operation. In this paper, an optimal trajectory is designed for a three degree-of-freedom planar redundant manipulator under the criterion of minimum energy consumption. Each joint trajectory is assumed as a fourth order polynomial and the proposed algorithms are required to optimize the coefficients in such a way the actuator torque will be minimized. Two optimization algorithms are presented here, the Genetic Algorithm (GA) and the Generalized Pattern Search (GPS). A simulation study to identify the optimal performance parameters is carried out using MATLAB. Comparison between the proposed two algorithms is carried out using the energy consumption per cycle, E, for validation.