In legged mobile robotics the most common approach is to design fully actuated legs with several degrees of freedom (DOF) in order to successfully navigate through rough terrains. However, simpler leg architectures with as few as one-DOF have been developed in the past to achieve the very same goal. The ability of these simpler legs to traverse uneven terrains is arguably limited with respect to multi-DOF designs, but in some applications the reduction of the DOF and hence, of the number of actuators, as well as the simplicity of the associated control could be a great advantage and the decisive argument. In this paper, the authors propose a novel one-DOF robotic leg that has been specially designed to achieve the greatest robustness possible with respect to the difficult terrains it has to traverse. In order to do that, a method to analyze and optimize any one-DOF robotic leg with respect to its ability to overcome obstacles is proposed here. This method is based on a simple and efficient novel technique to generate synthetic terrains combined with a simulation algorithm estimating the traversability of the particular one-DOF leg design under scrutiny. To illustrate the generality of the proposed method, it is used to design both an optimal leg with the architecture presented here for the first time and also, one with the most common one-DOF leg architecture found in the literature.

This content is only available via PDF.
You do not currently have access to this content.