Omnidirectional ZMP-based walking for a humanoid robot

• Autor(a) principal: Marcos Ricardo Omena de Albuquerque Maximo
• Data de Publicação: 2015
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Biblioteca Digital de Teses e Dissertações do ITA
• Texto Completo: http://www.bd.bibl.ita.br/tde_busca/arquivo.php?codArquivo=3242
• topic: Robôs humanóides; Dinâmica de robôs; Controle de robôs; Locomoção por pernas; Inteligência artificial; Robótica; Controle

Humanoid walking is considered one of the hardest problems in Robotics. Current state-of-the-art humanoid robots are able to achieve high speeds on flat ground. However, they still exhibit agility, dexterity, robustness, flexibility and energy efficiency far below a typical human does. In this thesis, our main goal is to develop an omnidirectional walking engine for a humanoid robot. We follow an approach based on the Zero Moment Point (ZMP) concept, which provides an useful criterion for biped stability. To avoid dealing directly with the complex dynamics of a high degrees of freedom humanoid robot, we used the 3D Linear Inverted Pendulum Model (3D-LIPM) to approximate the robot dynamics. The resulting equations allowed us to find a suitable center of mass (CoM) trajectory to maintain the robot balance analytically by solving a boundary value problem. Furthermore, we employed strategies to improve the walking robustness: we make the robot move its arms in order to compensate the yaw moment induced by the legs and we developed a feedback controller that uses the torso angular velocities to stabilize the walk. Taking advantage of the methods developed for walking, we also developed a kicking motion. Finally, experiments were done to validate the methods developed in this work.