Planned intervention: On Wednesday April 3rd 05:30 UTC Zenodo will be unavailable for up to 2-10 minutes to perform a storage cluster upgrade.
Published July 20, 2020 | Version v1
Journal article Open

A Layered Control Approach to Human-Aware Task and Motion Planning for Human-Robot Collaboration

  • 1. STIIMA-CNR - Institute of Intelligent Industrial Technologies and Systems, National Research Council of Italy
  • 2. Dipartimento di Ingegneria Meccanica e Industriale, University of Brescia
  • 3. ISTC-CNR - Institute of Cognitive Sciences and Technologies, National Research Council of Italy

Description

Combining task and motion planning efficiently in human-robot collaboration (HRC) entails several challenges because of the uncertainty conveyed by the human behavior. Tasks plan execution should be continuously monitored and updated based on the actual behavior of the human and the robot to maintain productivity and safety. We propose control-based approach based on two layers, i.e., task planning and action planning. Each layer reasons at a different level of abstraction: task planning considers high-level operations without taking into account their motion properties; action planning optimizes the execution of high-level operations based on current human state and geometric reasoning. The result is a hierarchical framework where the bottom layer gives feedback to top layer about the feasibility of each task, and the top layer uses this feedback to (re)optimize the process plan. The method is applied to an industrial case study in which a robot and a human worker cooperate to assemble a mosaic.

Files

ROMAN_2020_final-2.pdf

Files (1.2 MB)

Name Size Download all
md5:0d3b34fd379985ee2987735001f5b467
1.2 MB Preview Download

Additional details

Funding

SHAREWORK – Safe and effective HumAn-Robot coopEration toWards a better cOmpetiveness on cuRrent automation lacK manufacturing processes. 820807
European Commission