An Energy-Based Control Architecture for Shared Autonomy

In robotic applications where the autonomy is shared between the human and the robot, the autonomous behavior of the robotic system is determined considering mainly the task to be executed and the data collected from the environment using, e.g., formal methods and machine learning techniques. Nevert...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on robotics 2022-12, Vol.38 (6), p.3917-3935
Main Authors: Benzi, Federico, Ferraguti, Federica, Riggio, Giuseppe, Secchi, Cristian
Format: Article
Language:English
Subjects:
Autonomy
Behavioral sciences
Collaboration
Control systems
Couplings
Formal method
Human computer interaction
Human-centered robotics
Machine learning
Multi-robot systems
Multiple robots
multirobot systems
Optimization
optimization and optimal control
physical human–robot interaction
Robot control
Robotics
Task analysis
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In robotic applications where the autonomy is shared between the human and the robot, the autonomous behavior of the robotic system is determined considering mainly the task to be executed and the data collected from the environment using, e.g., formal methods and machine learning techniques. Nevertheless, it is important to correctly translate high-level decision into low-level control inputs in order to avoid an unstable behavior due to a naive implementation of the autonomy. In this article, we propose an energy-based architecture for shared autonomy that allows to reproduce as closely as possible the desired behavior, while ensuring a robust stability of the robotic system. The proposed architecture is experimentally validated in two application scenarios: shared control of a multirobot system and variable admittance control in human robot collaboration
ISSN:1552-3098
1941-0468
DOI:10.1109/TRO.2022.3180885