Home

throughput.sourceforge.net

A repository for replicable experiments with human-robot systems

---

A new kind of papers?

One may think of theoretical/concept papers, proof of concept papers, and experimental papers, as steps in a research idea "lifecycle". We believe that more papers of the "experimental" kind would greatly help the research activities in robotics and the industrial exploitation of the results. This is consistent with the Good Experimental Methodology Guidelines promoted by the Euron GEM Sig. Such "experimental papers" should likely contain (at least):

• Description: A journal paper with text+figures+multimedia, according to the above GEM Guidelines (or similar)
• Data sets: Complete set(s) of data for experiment (similar to International Journal of Robotics Research "Data Papers")
• Code: Complete code identifiers and/or downloadable code (executables may be enough)
• Hardware: Hardware description or hardware identifier (if it is identifiable)

The Throughput repository aims to provide a mechanism for sharing complete experimental papers for human-robot systems. Each experimental paper consists of the 4 parts listed above, in addition to the following part specific to experiments that include human participants:

• Method: Complete description of experiment method used, for example the instructions given to the participants

That is, each experimental paper is a sufficient description of an experiment that has both a robot and a human component, for the purpose of allowing replication of the experiment. The complete list of experimental papers currently in the repository can be found here: [ExperimentalPapers]

---

Experiment requirements

• Robot requirements:

  1. A simulated robot on a commercially available simulator, or
  2. A real commercially available robot platform.

• Human experiment requirements:

  1. Sufficient data. Including participant statistics, participant instructions, experiment protocol, and
  2. Sufficient “control”. Including sufficient repetitions, clear success/failure criteria, homogeneous set of participants.

---

Contributing

If you are interested in contributing an experimental paper, please contact Fabio Bonsignorio or Martin F. Stoelen.

The repository is a continuous work in progress, and we welcome both collaboration and feedback!

---

Licensing

All software code ('code' folder in each experimental paper archive) is made available under the GNU Lesser General Public License, version 3.0 (LGPL-3.0).

All other content (including the 'data_sets' and 'hardware' folders) is made available under the Creative Commons Attribution Licence, version 3.0. The licensor permits others to copy, distribute, transmit and adapt the work. In return, licensees must give the original author credit.

---

Attribution

Please include the following acknowledgement and citation when using an experimental paper. Replace [maintainer] with the name of the individual who maintains the experimental paper, with a citation to the maintainer's work (if given in the page for the respective experimental paper).

The experimental paper was obtained from the Throughput repository \cite{Throughput}, and is maintained by [maintainer] \cite{Maintainer}.

@misc{Throughput,
author = "Fabio Bonsignorio and Martin F. Stoelen",
year = "2012",
title = "throughput.sourceforge.net: A repository for replicable experiments with human-robot systems",
url = "http://throughput.sourceforge.net/",
}

---

This work is licensed under a Creative Commons Attribution 3.0 Unported License.

---

[ExperimentalPapers]