When it comes to physical collaboration between humans and robots, robots currently have a shortcoming: their ability to observe and adapt to human dynamics is limited. This leads to inefficient collaboration and unergonomic interaction. In this work, we combine a dynamic phase state system (PSS) based on a network of stable heteroclinic channels (SHC) with Compliant Movement Primitives (CMP). The combination of PSS and CMP enables intuitive human interaction in tasks where humans and robots physically cooperate. The capabilities of the control system were demonstrated in simulations involving helping a humanoid bipedal robot Talos stand up from a squat position by pulling on its hands.
Phase State System for Generating Interactive Behaviors for Humanoid Robots
Mechan. Machine Science
International Conference on Robotics in Alpe-Adria Danube Region ; 2022 ; Klagenfurt, Austria June 08, 2022 - June 10, 2022
2022-04-23
8 pages
Article/Chapter (Book)
Electronic Resource
English
Statistical Learning for Humanoid Robots
British Library Online Contents | 2002
|A Human-like Upper-limb Motion Planner: Generating naturalistic movements for humanoid robots
BASE | 2021
|