Intelligent trajectory planning and control of a humanoid robot using a new elitism-based Selfish Gene Algorithm

Abstract

The current development in science and civilization consists of searching for solutions to enhance our life, security, economy, finance and health while protecting environment. In recent years, we have witnessed the arrival of complex machines with structures similar to humans known as humanoid robots. The combination of these technologies and optimization technics may result in robust, safe, reliable, and flexible machines that can substitute for humans in multiple difficult tasks. In order to contribute to this topic, we propose two new evolutionary algorithms based on the selfish gene theory and elitism strategies. Therefore, permanent elitism-based selfish gene algorithm (peSGA) and nonpermanent elitism based selfish gene algorithm (neSGA) are proposed. In order to validate and to evaluate the performance peSGA and neSGA, a numerical experiment is performed using IEEE CEC 2014 functions. The obtained results show that the proposed algorithms are very competitive. Furthermore, evolutionary optimization of a walking robot is formulated. The proposed algorithms are applied to the generation and control of the optimal motion of a humanoid robot.