Robot Collision Avoidance Using an Environment Model for Capacitive Sensors

Alwin Hoffmann, Andreas Schierl, Andreas Angerer, Matthias Stüben, Michael Vistein, and Wolfgang Reif

In recent years, the topic of safe human robot cooperation has become important, and with it the requirement to reliably detect humans in the work space. Capacitive sensors mounted to the robot structure can be used to measure the presence of conductive objects. This allows to detect humans, however static obstacles in the workspace also influence sensor measurements. Thus we propose to record an environment model containing the expected sensor values for relevant robot poses. Using this model, distance estimation and real-time reaction can be performed even in the presence of large metallic objects in the robot workspace. A demonstration of our approach has been shown at the Hannover Messe 2015.
published 26.05.2015 Planning, Control, and Sensing for Safe Human-Robot Interaction, 2015 IEEE International Conference on Robotics and Automation, Seattle, USA, May 26, 2015


For questions regarding the publication, please contact!