Genetic Algorithms in Search, Optimization and Machine Learning
Genetic Algorithms in Search, Optimization and Machine Learning
A fast and elitist multiobjective genetic algorithm: NSGA-II
IEEE Transactions on Evolutionary Computation
Robotics and Computer-Integrated Manufacturing
| Hi-index | 0.00 |
Force balancing is a very important issue in mechanism design and has only recently been introduced to the designing step of robotic mechanisms. In creating the best robot design, the statical balancing plays a vital role because it reduces the required motor power. To get a simple and more-effective control system, elimination or significant reduction of the gravity load at a powered joint is an important one. With utilization of these objectives an optimization problem is formulated. The average force on the gripper in the working area is taken as an objective function. The design variables are lengths of the links, angles between them and stiffness of springs. This paper describes the use of conventional and evolutionary optimization techniques such as Newton's method (NM), conjugate gradient method (CGM), Genetic Algorithm (GA), Elitist Non-dominated Sorting Genetic Algorithm (NSGA-II) and differential evolution (DE) to solve the above problem. An industrial robot with 6-degree-of-freedom (6-DOF) (APR 20) is considered as a numerical example. The robot has a spring balancing system that has to be optimized. The existing optimization model is improved by adding two new variables. Also, a comprehensive user-friendly general-purpose software package has been developed using VC++ to obtain the optimal parameters using the proposed DE algorithm. The methods used in this article can be applied to find out solutions for a wide range of similar problems without further simplifications.