Bipedal humanoid walking behavior is a very hot topic in robotics research. It is a challenging topic for humanoid robots as various factors ranging from mechanic design of the robot to the system dynamics will affect how well the robot walks. In the other research, the mechanical structure of the robot can be easily changed in order to fit for the application and the research purpose. However, as the requirement in the standard platform league of Roboup, the NAO robot hardware cannot be modified and change. The improvement of the NAO robot walking can only be done by improving the embedded software.

In this thesis research, the objective is to develop a walking module for the NAO Robot. The proposed walking module must be Omni-directional. Every walking motions are fluent and...[ Read more ]

Bipedal humanoid walking behavior is a very hot topic in robotics research. It is a challenging topic for humanoid robots as various factors ranging from mechanic design of the robot to the system dynamics will affect how well the robot walks. In the other research, the mechanical structure of the robot can be easily changed in order to fit for the application and the research purpose. However, as the requirement in the standard platform league of Roboup, the NAO robot hardware cannot be modified and change. The improvement of the NAO robot walking can only be done by improving the embedded software.

In this thesis research, the objective is to develop a walking module for the NAO Robot. The proposed walking module must be Omni-directional. Every walking motions are fluent and without stopping between each transactions. It must be stable and maintain balance when walking and walk faster than the original behavior. Even there exist external disturbances, such as pushes, the proposed walking module can still keep the NAO robot stable.

The proposed algorithm of the walking module is using several of mathematics models like Dubins curve, Bezier curve and experiments in order to improve the walking efficiency. Also, forward and inverse kinematics has been adopted to improve the stability and precision of the robot moving. The testing result showed that there is over 100% improvement in the walking speed and greatly improve the stability of the robot to resist external disturbances.