Stammdaten

Titel: Towards Safe Physical Human-Robot Interaction with Capacitive Sensors for Tactile and Proximity Perception
Beschreibung:

Physical interaction between humans and robots is an essential part in Human-Robot Collaboration (HRC). Such interactions may be split into undesired collisions and intended actual interactions. Undesired contacts, in general, shall be avoided or at least obey certain force limits exerted by the robot, depending on the body region affected. During desired contacts these limits shall also be obeyed, nevertheless. In rehabilitations robotics, patients with limited tactile perception may not notice a violation of the limits leading to potential serious injuries. Limits on forces and safety measures for industrial applications are given in ISO/TS 15066:2016. Traditionally, robots obey these limits by reducing their power and speed. Unfortunately, this only allows robots to move at small pace limiting their profitability for most industrial applications. Internal force/torque sensing at the joints may be inaccurate, underestimating actual forces, as shown in previous works.

A method to combine both proximity detection and tactile perception with a single technology are capacitive sensors. They may be attached on the robot as a deformable skin. In this manner they can be used as tactile sensors to record forces exerted through physical interaction. Furthermore, they may operate as proximity sensors to detect humans close to the robot up to distances of 400 mm. In this manner robots may operate faster and brake (or stop) in case of human intrusion.

The actual exerted forces are a complex interplay of various parameters such as mechanical properties, inertia, power and pose of the robot. To consider all these parameters simulation models may be generated. Such simulations of typical interaction scenarios can vastly reduce the number of experiments needed. With additional real time capability such simulations can also act as digital twins.

Schlagworte: Human-Robot Collaboration, Human-Robot Interaction, Safety in Robotics, Capacitive Sensors
Typ: Angemeldeter Vortrag
Homepage: https://www.proxelsandtaxels.org
Veranstaltung: 4th Workshop on Proximity Perception in Robotics, IROS 2021 (Prague (virtually))
Datum: 27.09.2021
Vortragsstatus: stattgefunden (online)

Beteiligte

Zuordnung

Organisation Adresse
Fakultät für Technische Wissenschaften
 
Institut für Intelligente Systemtechnologien
Universitätsstraße 65-67
9020 Klagenfurt am Wörthersee
Österreich
   hubert.zangl@aau.at
http://www.uni-klu.ac.at/tewi/ict/sst/index.html
zur Organisation
Universitätsstraße 65-67
AT - 9020  Klagenfurt am Wörthersee

Kategorisierung

Sachgebiete
  • 202036 - Sensorik
Forschungscluster Kein Forschungscluster ausgewählt
Vortragsfokus
  • Science to Science (Qualitätsindikator: n.a.)
Klassifikationsraster der zugeordneten Organisationseinheiten:
TeilnehmerInnenkreis
  • Überwiegend international
Publiziert?
  • Nein
Arbeitsgruppen Keine Arbeitsgruppe ausgewählt

Kooperationen

Keine Partnerorganisation ausgewählt