This paper proposes a new selfish gene algorithm called the Replaces and Never Penalizes Selfish Gene Algorithm (RNPSGA). This new variant of selfish gene algorithm replaces the alleles of the less fit individual by the alleles of the fittest rather than penalizing them. The intensification of the search is then increased. The proposed algorithm is tested under some famous benchmark functions and compared to the standard selfish gene algorithm. We analyzes also the quality of convergence, the accuracy, the stability and the processing time of the proposed algorithm. We design by Solidwork a new virtual model of the humanoid robot hanging on the bar. The model is controlled using Simscape/Matlab. The proposed algorithm is then applied to the designed humanoid robot. The objective is to realize the gymnastic movements on the bar. An intelligent LQR controller is proposed to stabilize the swing-up of the robot.
CITATION STYLE
Tighzert, L., & Mendil, B. (2016). Realization of Gymnastic Movements on the Bar by Humanoid Robot Using a New Selfish Gene Algorithm. In Lecture Notes in Networks and Systems (Vol. 1, pp. 49–65). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-319-33410-3_4
Mendeley helps you to discover research relevant for your work.