dynamic robot

Most of the biped robots are controlled using pre-computed trajectories methods or methods based on the multibody dynamics models. The pre-computed trajectory based methods are simple but the system gets highly vulnerable to the external disturbances. In contrast, the dynamically based methods make the system acts faster, but they need extensive knowledge about the kinematics and dynamics of the system. This fact gave rise to the main purpose of this study, i.e., developing a controller for a biped robot to take advantages of the simplicity and computational eciency of trajectory-based methods and the robustness of the dynamic-based approach. To do so, this paper presents a twolayer hierarchical control framework for an under-actuated, planar, 5-link biped robot model. The upper layer contains a centralized dynamic-based controller and uses all system sensory data to generate stable walking. The lower layer in this structure is a decentralized trajectory-based controller network which learns to control the system based on the upper layer controller output. When the lower controller fails to control the system, upper layer controller takes action and makes the system stable. Then, when the lower layer controller gets ready, the control of the system will be handed to this layer.