Optimizing NAO humanoid walking using ABC algorithm

Humanoid locomotion of biped robots is an interesting research area due to its similarity to human walking. Furthermore, wide application of humanoid locomotion especially in robocup competitions made it more interesting for several researchers in the fields of electronics, mechanics and computer science, although it has its own complexity. This paper proposes a novel approach that optimizes the walking speed and stability for NAO biped robots. This method has been implemented on simulated NAO robot in Robocup 3D simulation environment (rcsssever3d). We have used Artificially Bee Colony (ABC) algorithm in order to achieve full stability and considerable fast walking speed without any external disturbances. Experimental results show that using this approach also, significantly decreased optimization time in order to get optimal values for parameters in our fitness function. At end a comparison has been made between the described optimization solution and GA optimization, while same walking method and foot trajectory formula has been used.

Keywords:

Zero Moment Point(ZMP) , Biped Robot walking , NAO , Artificial Bee Colony(ABC) , Genetic Algorithm(GA)