A Biologically Founded Design and Control of a Humanoid Biped

During the last decade many advancements in the fields of industrial and service robotics have produced robots that are well integrated inside the industry; they can operate faster and with higher precision in comparison to human beings. Nevertheless if we take a look at the kinematic structure of these systems, it is clear that the actual machines are limited in the mobility and in the number of tasks that they can perform. This is more evident if we intend to apply those robots in an unstructured environment like home. First at all the robot should be able to move around avoiding obstacles, climbing stairs, opening doors. These movements should also be performed with a certain level of compliance for the safety of the human beings that are in the environment. Secondly, the robots should be able to use tools and other machines designed for human use, and based of the human manipulation and kinematic abilities. A possible solution for mobility, that is well applied in mobile robotics, is the choice of a wheeled traction system. This usually is a simple manner to move on flat floors, and is efficient from the energetic point of view (during the movement the center of mass acts on a straight line). However it presents important limitations, for example it is not possible for such a robot to overcome obstacles bigger than the wheels dimensions. Those limitations can be overcome if the robot is equipped with legs, that normally act by increasing the robot's DOF(Degrees of Freedom). Many studies were conducted on legged robot in order to improve the efficiency and stability during walking. A pioneering contribution was done (Takanishi et al, 2004) at the laboratories of Waseda University (Tokyo). Several other modern robots are designed to walk and behave like humans (Hashimoto et al, 2002)( 3) but until now the efficiency of the human gait is still far from being reached. In this sense, the work of McGeer (McGeer, 1990) can be considered exemplar. His passive dynamic walker made a stable gait without close position control, considering the walking motion as a natural oscillation of a double pendulum; and this is actually how humans seem to walk (Gottlieb et al, 1996) (Kiriazov, 1991). His results inspired many other works, such as the stability analysis (Garcia et al, 1998) and the physical implementation ( Wisse et al, 2001) (Kuo, 1999)(Collins et al, 2001) of several prototypes.