Utilizing a Phase State System for Reliable Physical Assistance in Human-Humanoid Robot Collaboration

This paper explores the application of the Talos humanoid robot in supporting sit-to-stand or stand-to-sit tasks that require humans to exert significant force within a short time frame. These tasks pose challenges related to human unpredictability and the need for precise robot stability control. To tackle these challenges, we propose a novel spatial reference approach that allows the robot to dynamically respond to human forces while maintaining stability. Our approach is based on a dynamic phase state system, guided by stable heteroclinic channel networks, which enables effective collaboration between the robot and the human. By applying forces to the robot's grippers, the human can actively control the underlying dynamic system and influence the robot's motions. While the primary focus of this study is on developing supportive motions to assist humans in the sit-to-stand process, our dynamic model can be extended to encompass a wide range of other movements and supportive actions. The results of our research demonstrate the potential of our approach in achieving efficient and reliable human-robot collaboration. This study provides valuable insights and paves the way for further exploration and development in the field of assistive robotics.