Evolutionary Minimum Cost Trajectory Planning for Industrial Robots

A method for computing minimum cost trajectory planning for industrial robot manipulators is presented. The objective function minimizes the cost function with the constraints being joint positions, velocities, jerks and torques by considering dynamic equations of motion. A clamped cubic spline is used to represent the trajectory. This is a non-linear constrained optimization problem. The problem is solved by evolutionary techniques such as Elitist Non-dominated Sorting Genetic Algorithm (NSGA-II) and Differential Evolution (DE) methods. Numerical application for a 6 link revolute robotic manipulator (PUMA 560) is illustrated. The results obtained from the Proposed techniques (NSGA-II and DE) are compared with the conventional technique Sequential Quadratic Programming (SQP)[3] for different values of weighting coefficients. The proposed NSGA-II and DE converge quickly and they show their superior nature by giving better results than the conventional optimization method SQP [3].