A Dynamic Walking Strategy for the Nao Robot Considering Path Planning, CoM and ZMP

Abstract

This paper deals with the problem of the locomotion of robots with two legs with a design inspired by human locomotion (anthropomorphic bipedal robots). The future intention of the research group is to consolidate a robotic platform capable of interacting side by side with the human being, in particular by providing services to people. We propose a dynamic bipedal walking scheme based on a simulated model of inverted pendulum and using the Zero Moment Point (ZMP) as a control strategy. To solve the problem of closed-loop control, the strategy estimates the robot's Center of Mass (CoM) continuously. The formulation, analysis, solution and implementation is done on the Nao robot from Aldebaran Robotics. The strategy was encoded in a NAOqi module in Python, and executed in a Nao V5. The behavior of the robot during the walk demonstrated the success of the strategy. Keyword -Bipedal robot, Center of mass, Dynamic walking, Locomotion, Zero moment point I. INTRODUCTION Despite the great progress made in recent years in robot control, and in particular in the area of bipedal walking schemes, many problems remain open to solution in this area [1]