ITU

Automating the Incremental Evolution of Controllers for Physical Robots

Research output: Journal Article or Conference Article in JournalJournal articleResearchpeer-review

Standard

Automating the Incremental Evolution of Controllers for Physical Robots. / Faina, Andres; Jacobsen, Lars Toft; Risi, Sebastian.

In: Artificial Life, Vol. 23, No. 2, 2017, p. 142-168.

Research output: Journal Article or Conference Article in JournalJournal articleResearchpeer-review

Harvard

APA

Vancouver

Author

Bibtex

@article{12778aa001a546eab98417eecfa4fcb4,
title = "Automating the Incremental Evolution of Controllers for Physical Robots",
abstract = "Evolutionary robotics is challenged with some key problems that must be solved, or at least mitigated extensively, before it can fulfill some of its promises to deliver highly autonomous and adaptive robots. The reality gap and the ability to transfer phenotypes from simulation to reality constitute one such problem. Another lies in the embodiment of the evolutionary processes, which links to the first, but focuses on how evolution can act on real agents and occurindependently from simulation, that is, going from being, as Eiben et al. put it, “the evolution of things, rather than just the evolution of digital objects.…” The work presented here investigates how fully autonomous evolution of robot controllers can be realized in hardware, using an industrial robot and a marker-based computer vision system. In particular, this article presents an approach to automate the reconfiguration of the test environment and shows that it is possible, for the first time, to incrementally evolve a neural robotcontroller for different obstacle avoidance tasks with no human intervention. Importantly, the system offers a high level of robustness and precision that could potentially open up the range of problems amenable to embodied evolution.",
author = "Andres Faina and Jacobsen, {Lars Toft} and Sebastian Risi",
year = "2017",
doi = "10.1162/ARTL_a_00226",
language = "English",
volume = "23",
pages = "142--168",
journal = "Artificial Life",
issn = "1064-5462",
publisher = "M I T Press",
number = "2",

}

RIS

TY - JOUR

T1 - Automating the Incremental Evolution of Controllers for Physical Robots

AU - Faina, Andres

AU - Jacobsen, Lars Toft

AU - Risi, Sebastian

PY - 2017

Y1 - 2017

N2 - Evolutionary robotics is challenged with some key problems that must be solved, or at least mitigated extensively, before it can fulfill some of its promises to deliver highly autonomous and adaptive robots. The reality gap and the ability to transfer phenotypes from simulation to reality constitute one such problem. Another lies in the embodiment of the evolutionary processes, which links to the first, but focuses on how evolution can act on real agents and occurindependently from simulation, that is, going from being, as Eiben et al. put it, “the evolution of things, rather than just the evolution of digital objects.…” The work presented here investigates how fully autonomous evolution of robot controllers can be realized in hardware, using an industrial robot and a marker-based computer vision system. In particular, this article presents an approach to automate the reconfiguration of the test environment and shows that it is possible, for the first time, to incrementally evolve a neural robotcontroller for different obstacle avoidance tasks with no human intervention. Importantly, the system offers a high level of robustness and precision that could potentially open up the range of problems amenable to embodied evolution.

AB - Evolutionary robotics is challenged with some key problems that must be solved, or at least mitigated extensively, before it can fulfill some of its promises to deliver highly autonomous and adaptive robots. The reality gap and the ability to transfer phenotypes from simulation to reality constitute one such problem. Another lies in the embodiment of the evolutionary processes, which links to the first, but focuses on how evolution can act on real agents and occurindependently from simulation, that is, going from being, as Eiben et al. put it, “the evolution of things, rather than just the evolution of digital objects.…” The work presented here investigates how fully autonomous evolution of robot controllers can be realized in hardware, using an industrial robot and a marker-based computer vision system. In particular, this article presents an approach to automate the reconfiguration of the test environment and shows that it is possible, for the first time, to incrementally evolve a neural robotcontroller for different obstacle avoidance tasks with no human intervention. Importantly, the system offers a high level of robustness and precision that could potentially open up the range of problems amenable to embodied evolution.

U2 - 10.1162/ARTL_a_00226

DO - 10.1162/ARTL_a_00226

M3 - Journal article

VL - 23

SP - 142

EP - 168

JO - Artificial Life

JF - Artificial Life

SN - 1064-5462

IS - 2

ER -

ID: 81927409