OPTIMAL PLACEMENT OF PATH FOLLOWING ROBOT TASK USING GENETIC ALGORITHMS

On this paper, the objective is to introduce an optimization algorithm in order to determine the near optimal location of a path following task for a 6 D.O.F. non redundant manipulator, so that its end-effector can follow a given 3D curve and orientation, taking into account the maximization of the robot velocity performance. The optimization is conducted using as an objective function the minimum value of the Manipulator Velocity Ratio (MVR) along the path, producing the largest possible end-effector velocity while maintaining minimum joint speed values. The optimization of the objective function is based on Genetic Algorithms. The minimum of the MVR is determined by a technique which approximates the given path using cubic interpolation for the position and squad interpolation based on quaternions for the orientation. A large number of experiments show the effectiveness of the proposed approach and the most indicative of the results are presented and discussed at the end of the paper.