1: \begin{abstract}

2: This paper seeks insight into stabilization mechanisms for periodic

3: walking gaits in 3D bipedal robots.

4: Based on this insight, a control strategy based on virtual

5: constraints, which imposes coordination between joints rather than a

6: temporal evolution, will be proposed for achieving asymptotic convergence toward a periodic motion. For planar bipeds with one degree of underactuation, it is known that a

7: vertical displacement of the center of mass---with downward velocity at the step transition---induces stability of a walking gait. This

8: paper concerns the \textit{qualitative extension} of this type of property to 3D walking {\color{black} with two degrees of underactuation}.

9: It is shown that a condition on the position of the center of mass in the horizontal plane at the transition between steps induces synchronization between the motions in the sagittal and frontal planes. A combination of the conditions for self-synchronization and vertical oscillations leads to stable gaits. The algorithm for self-stabilization of 3D walking gaits is first developed for a simplified model of a walking robot (an inverted pendulum with variable length legs), and then it is extended to a complex model of the humanoid robot Romeo using

10: the notion of Hybrid Zero Dynamics. Simulations of the model of the robot illustrate the efficacy of the method and its robustness.

11: \end{abstract}

12: