• cooperative human-robot dancing
• two-wheeled humanoid robot
• robot control
• physical human-robot interaction (pHRI)

Human-Robot Interaction is an interdisciplinary field of study that aims at developping robots able to safely interact with humans in different collaborative tasks. This work is part of the Cooperative Human Robot Dance that involves a real physical interaction between the two parties. Taking inspiration from the Waltzer, a "leader- follower" modality dancing, considering to work with Two-Wheeled Humanoid Robots and assuming that the interaction between the two parts takes place through the union of the hands, the novelty of the work presented here consists in modeling the dance with the robot as a pursuit of two target points by the robot's mobile base. These two target points correspond to the robot’s hands moved into space by the operator who holds them to guide him. The aim of the work is to provide control laws in forward linear velocity and angular rate that allow the mobile base to chase the two kinematic reference target points (the robot's hands) moving in space. To tackle the problem, a new control approach namely Collaborative Bi-Manual Pairing Control is designed. It's validated in Matlab simulations and compared with control approaches derived from the state of the art. It's finally tested, using Robot Operating System, on AlterEgo, the anthropomorphic Two-Wheeled Humanoid Robot born in 2017 from the joint efforts of the SoftBots Lab of the Italian Institute of Technology (IIT) of Genoa and the Research Center “E. Piaggio” of the University of Pisa. The experimental results show that, among the controls provided, the Collaborative Bi-Manual Pairing Control is effective and functional in accomplishing the goal and moreover it increases the robot’s ability to collaborate and interact, without resisting, tothe operator who moves it around.