Human-Humanoid collaboration: the next industrial revolution?
When: November 15th, 2017
Where: HUMANOIDS 2017, The Rep Theatre, Birmingham (UK), Room TBD
In the last years there has been a proliferation of applications for collaborative robots in manufacturing and industry, from bimanual cooperation in assembly with cobots (i.e., industrial manipulators for collaboration) to physical assistance with exoskeletons. These applications have driven research in many fundamental topics for collaboration, such as shared task allocation, synchronization and coordination, control of contacts and physical interaction, role estimation and adaptive role allocation during collaboration, learning by demonstrations, safe control, etc. All the advances in the state of the art in these topics contribute to the success of the “so called” Industry 4.0, whose elite platforms are essentially cobots and exoskeletons.
At the same time, the research community in humanoids has made tremendous progresses in the whole-body control area. Now we can see humanoid robots, such as iCub, Talos or Atlas, that are torque-controlled and able to perform complex highly-dynamic tasks in presence of unforeseen physical interactions with humans or their environment. The DRC showed that humanoids can manipulate objects and tools, walk and even drive cars. In the European project Comanoid the humanoid HRP4 is performing manipulation task in complex industrial environment. Will the control be the same when they collaborate with humans? And what are the challenges that we face if we want them to integrate cobots and exoskeletons as collaborative robots for industry? Is the prior experience with such platforms sufficient for realizing collaborative applications or are there new problems that are typical of the humanoid platforms? Besides the obvious control problems, what will be the reaction and adaption of humans when collaborating with humanoids?
These questions and many others are addressed in some European projects, such as An.Dy or SecondHands, where humanoids are envisioned as future collaborators, i.e., capable of providing assistance and helping human workers. To realize such applications, the humanoids need to control precisely the physical interaction with the humans, and be “aware” of the human status. The main requirements are valid for exoskeletons and cobots, but humanoids make the equation much more complex, for two main reasons.
First, controlling the physical interaction with humans is a secondary problem that adds to controlling already their own balance. Second, the anthropomorphic appearance, combined with interacting on tasks designed for humans, can facilitate the human collaborators, but at the same time the behavior of human collaborators may vary a lot, exhibiting reactions that we may have never not observes in the case of interaction with an industrial manipulator.
This workshop aims at drawing the state-of-the-art for collaborative humanoid robots, and discussing the current challenges that the research community faces for realizing the future collaborative humanoids for industry.
The workshop will consists of invited talks (30 min) and presentations from selected contributions after a call for papers/posters. The workshop will close with a panel discussion.
Topics of interest
- Safe control of physical interaction
- Whole-body control of physical interaction with humans
- Human-in-the-loop control
- Human movement and dynamics estimation
- Anticipation and prediction
- Tele-operation of humanoid robots
- Human augmentation
- Wearable sensors
- Muscolo-skeletal models and ergonomics
- Collaborative and ergonomics control
- Applications for humanoid robots in the future
- Graham Deacon (Ocado Technologies)
- Arash Ajoudani (Italian Institute of Technology, Italy)
- Abderrhamane Kheddar (CNRS & LIRMM, France)
- Francesco Nori (Google Deepmind, UK & IIT, Italy)
- Dongheui Lee (DLR & TUM, Germany)
- Jan Peters (TU Darmstadt, Germany)
- Chris Atkenson (CMU, USA)
- Sethu Vijaykumar (University of Edinburgh, UK)
|8:30 – 9:00||Introduction|
|9:00 – 10:00||Talks|
|9:00 – 9:30||Graham Deacon (Ocado Technologies)|
|9:30 – 10:00||Poster spotlight presentations|
|10:00 – 10:30||Coffee Break|
|10:30 – 12:30||Talks|
|10:30 – 11:00||Abderrhamane Kheddar (LIRMM, France)|
|11:00 – 11:30||Francesco Nori (Google Deepmind, UK & IIT, Italy)|
|11:30 – 12:00||Dongheui Lee (DLR & TUM, Germany)|
|12:00 – 12:30||Selected contribution 1|
|12:30 – 13:30||Lunch Break|
|13:30 – 15:00||Talks|
|13:30 – 14:00||Chris Atkenson (CMU, USA)|
|14:00 – 14:30||Sethu Vijaykumar (University of Edinburgh, UK)|
|14:30 – 15:00||Selected contribution 2|
|15:00 – 15:30||Coffee Break|
|15:30 – 17:30||Talks|
|15:30 – 16:00||Jan Peters (TU Darmstadt, Germany)|
|16:00 – 16:30||Arash Ajoudani (IIT, Italy)|
|16:30 – 17:00||Selected contribution 3|
|17:00 – 17:30||Wrap-up and panel discussion|
Call for contributions
We welcome prospective participants to submit extended abstracts and short papers (up to 2 pages) to be presented as posters.
Manuscripts should use the IEEE double column template.
Papers will be selected based on relevance to the workshop, originality, technical contribution and clarity.
We will accept papers from different topics, including (and not limited to) control, learning, HRI, robot design.
The best contributed papers will be selected for oral presentation at the workshop.
Papers should be submitted via email to firstname.lastname@example.org with the tag [humanoids_workshop]
Submission deadline for papers:
21 October, 2017 extended 5 November, 2017
Notification of acceptance:
31 October, 2017 extended 7 November, 2017
The workshop is supported by the European Project An.Dy.