Supplementary Material To: A Safety Benchmark for Capacitive Proximity Sensing in Robotics – Safe Transient Contact Compliant to ISO TS 15066: Power and Force Limiting

Citation Author(s):
Klagenfurt University
Chemnitz University of Technology
Karlsruhe Institute of Technology
Klagenfurt University
Joanneum Research Robotics
Karlsruhe Institute of Technology
Joanneum Research Robotics
Chemnitz University of Technology
Karlsruhe Institute of Technology
Klagenfurt University
Submitted by:
Serkan Ergun
Last updated:
Wed, 10/21/2020 - 05:16
Data Format:
1 rating - Please login to submit your rating.


Supplementary Data for Publication  "A Safety Benchmark for Capacitive Proximity Sensing in Robotics  - Safe Transient Contact in Compliance to ISO TS 15066: Power and Force Limiting


The Supplementary Data includes

  • STL files of the test objects (already available)
  • Measurement Data of each institute for the shown test objects (are partially uploaded, list of files will be completed soon)


Disclaimer: This publication was submitted and is currently under review to be published for IEEE RA-L with ICRA 2021 option.


Fundings: The research leading to this results has received fundingfrom the ”K ̈arntner Wirtschaftsf ̈orderung Fonds” (KWF) andthe ”European Regional Development Fund” (EFRE) withinthe CapSize project 26616/30969/44253 and from the Fed-eral Ministry of Education and Research (Bundesministeriumf ̈ur Bildung und Forschung in Germany, BMWF) within theproject SINA (Verbundprojekt-Nr.: V5ARA114)



Preliminary version of the abstract:


During the co-presence of human workers and robots in the same workspace, Power and Force Limiting (PFL) according to ISO TS 15066 is required to avoid injuries during undesired contacts. Capacitive Proximity Sensors (CPSs) offer a cost-effective solution to cover the entire robot manipulator with a fast close-range perception for Human-Robot Collaboration (HRC) tasks. Compared to pure tactile or force sensing, the sensors allow to increase the operating speed of cobots while still maintaining PFL. The robot reacts before impacts occur and does not suffer from occlusion, as the sensors can be placed directly on the robot’s surface to cover the full area and close the perception gap between tactile detection and mid-range distance sensors. However, since capacitive coupling to obstacles varies with distance, shape, and material properties,the projection from capacitance to actual distances is a general problem. A reference benchmark is required for the evaluation to ensure safe PFL. We derive the requirements described in ISO TS 15066 for PFL of critical body regions and propose a method for determining the pre-detection operation speed to comply with PFL. We propose an universal benchmark test procedure for fellow researchers to evaluate their capacitive sensors for safe PFL. With researchers from four different institutes presenting three different concepts of capacitive sensors and their applicability on different robots, the benchmark test procedure is evaluated, and the general applicability is demonstrated.



**** Test objects *****

File list:



This *.zip directory contains *.stl files with the test objects used in the publication


These files are ready to pre-processed with slicing software to be 3d printed.


Bigger objects such as the face and the sphere are splitted into multiple parts, which can be then glued together. When covering your test object with copper foil, make sure that the adhesive layer is also conductive.


In order to mount the test objects to your robot you  need a plastic tube (or cylindrical rod) with 12 mm outer diameter, which can be bought at your local hardware store or be 3d printed as well.

An additional adapter is needed to mount the plastic tube  to your robot's TCP.



**** Sensing directivity measurements****

The measurement results for the sensing directvitiy can be accessed by

File list:




(further measurement results will be uploaded soon)



Each file is structured as follows:

Columns 1 to 3 represent the Cartesian Coordinates of the lowest point of the test object with respect to the center of the electrode

Column 4 shows the average of measurements conducted in 10 ms calibrated by the baseline value

Column 5 shows the standard deviation of 500 measurements at each point

Column 6 and 7 represent basline value and baseline standard deviation (by the same means as for Column 5-6) respectively.

    NOTES: AAU recorded the baseline value at 40 cm (this is why that value is constant for the whole column)

Column 8 indicates if that point was detected by means explained in Section IV - A of the paper


 **** Spatial resoltuion measurements****

This section will be uploaded soon.


If you need further information, do not hesitate to contact us at:

Serkan Ergun

Institute for Smart System Technologies

Sensors and Actuators Group

Universitätsstrasse 65-67

A - 9020 Klagenfurt