Physical and perceived safety in human-robot collaboration

Autor: Marco Maiocchi, Andrea Maria Zanchettin, Paolo Rocco
Rok vydání: 2020
Předmět:
DOI: 10.5281/zenodo.4781434
Popis: This paper presents a complete motion control system for industrial manipulators, designed to maximize the robot productivity and meet the safety requirements of the human worker during human-robot collaboration. The real-time solution of a constrained optimization problem allows to control the robot’s motion in accordance with a pre-programmed trajectory, satisfying both the kinematic contraints and those imposed by the adopted collision avoidance strategy. A human safety perception system is also introduced, which aims to increase the human worker’s psychological safety and prevent potentially dangerous movements. A Mixed Reality software application, which provides the human worker with a digital visualization of the robot motion intention, and a real-time motion prediction algorithm, which computes the future poses of the robot, complete the system. Finally, a control stategy based on the human field of view is presented.
https://youtu.be/wLFhPEckOYk
Databáze: OpenAIRE