الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 141 |  |  |
| | | from | 141 | | |
Abstract
In recent Years, Robot arms are essential tools in industries due to its accuracy through high-speed manufacturing. One of the most challenging problems in industrial robots is solving inverse kinematics. Inverse Kinematic Problem concerns with finding the values of angles which are related to the desired Cartesian location. With the development of Soft-computing-based methods, it’s become easier to solve the inverse kinematic problem in higher speed with sufficient solutions rather than using traditional methods like numerical, geometric and algebraic.In this study, different soft-computing based methods (Artificial Neural Network, Adaptive Neuro-Fuzzy Inference System & Genetic Algorithms) are applied to the problem of inverse kinematics. Moreover, with the help of the proposed method called minimized error function (MEF), both ANN and ANFIS are able to outperform other methods. Also, a new ABC variant is presented in this study for solving the inverse kinematics problem.The performance of artificial bee colony (ABC) has been encouraging enough, however ABC algorithm lacks good compromise between exploration and exploitation. Therefore, an ABC-based algorithm; namely knowledge-based artificial bee colony (K-ABC) is proposed and it is proved to be able to converge quickly and explore the most promising area of the intended search space. To be sure that the K-ABC algorithm is trustworthy and reliable, it is tested against the most recent well-known benchmarks CEC’2017. On top of that, the K-ABC wins against their counterpart on some complex constrained engineering problems. The effect of the limit parameter is investigated as well.The experimental results are made with an inverse kinematic problem with a 5 degrees of freedom (DOF) robot arm for the different five soft-computing methods (Adaptive Neuro-fuzzy Inference System (ANFIS), Artificial Neural Network (ANN), Genetic Algorithm (GA), Deferential Evolution (DE), Particle Swarm Optimization (PSO), Artificial Bee Colony (ABC), and Knowledge Based Artificial Bee Colony (K-ABC)). The final results show that K-ABC has a mean squared error of 60% less than any other optimization algorithm for the inverse kinematic solution.