THIS IS A PAST EVENT
This page is maintained for archival purposes.
EVENT SUMMARY, FOLLOW-UP
- Invited Talk, Title-TBD, Matei Ciocarlie, Google, Inc., USA [Slides]
- Invited Talk, Title-TBD, Sarah Osentoski, Bosch, USA [Slides]
- Cloud Based Localization for Mobile Robot in Outdoors, Xiaorui Zhu [Slides] [Publication] [BibTex]
- A cloud robotics architecture for an emergency management and monitoring service in a smart city environment, Gabriele Ermacora/Antonio Toma [Slides] [Publication] [BibTex]
- On Distributed Knowledge Bases for Small-Batch Assembly, Maj Stenmark [Slides] [Poster] [Publication] [BibTex]
- Life-long Learning Perception using Cloud Database Technology, Tim Niemueller [Slides] [Publication] [BibTex]
- Robot Reinforcement Learning using Crowdsourced Rewards, Christain Penaloza [Slides][Publication] [BibTex]
- Kinect@Home: Crowdsourced RGB-D data, Rasmus Göransson [Slides] [Poster] [Publication] [BibTex]
- Invited Talk, Inheriting Humanoid Systems from Remote-Brained Robotics, Masayuki Inaba, Uni. Tokyo, Japan [Slides]
- Agent driven Peer-to-Peer Cloud Robotics, Vineet Nagrath [Slides] [Poster] [Publication] [BibTex]
- C2TAM: A First Approach to a Cloud framework for Cooperative Tracking
And Mapping, Javier Civera [Slides] [Publication] [BibTex] - Invited Talk: Cooperative Tracking And Mapping, Gajamohan Mohanarajah [Slides]
- Invited Talk, Web-enabled and experience-based learning of everyday manipulation tasks, Moritz Tenorth/Michael Beetz, TU Bremen, Germany [Slides]
- A Cloud Robotics Architecture to Foster Individual Child Partnership in Medical Facilities, Joan Navarro [Slides] [Publication] [BibTex]
- Lighting Session [Hagiwara-slides]
ARCHIVED PAGE:
Welcome to the page of the IROS 2013 Workshop on Cloud Robotics: Online Knowledge Bases, Web Services, and Cloud Computing for Robots.
Date: Sunday, 3. November 2013
Time: 9am-5pm
Location: Tokyo Big Sight Conference Tower, Room 610 (on Google Maps)
Contents
Program
09.45 Invited Talk, Title-TBD, Speaker TBD13:55Agent driven Peer-to-Peer Cloud Robotics, Vineet Nagrath
| 09.00 | Opening, Markus Waibel |
| 09:15 | Invited Talk, Title-TBD, Matei Ciocarlie, Google, Inc., USA |
| 09:30 | Invited Talk, Title-TBD, Sarah Osentoski, Bosch, USA |
| 10:00 | A cloud robotics architecture for an emergency management and monitoring service in a smart city environment, Gabriele Ermacora/Antonio Toma |
| 10.15 | On Distributed Knowledge Bases for Small-Batch Assembly, Maj Stenmark |
| 10.30 | Coffee break and Posters |
| 11.30 | Life-long Learning Perception using Cloud Database Technology, Tim Niemueller |
| 11.45 | Robot Reinforcement Learning using Crowdsourced Rewards, Christain Penaloza |
| 12:00 | Kinect@Home: Crowdsourced RGB-D data, Rasmus Göransson |
| 12:15 | Demo 1: Kinect@Home, Rasmus Göransson |
| 12.30 | Lunch break |
| 13.30 | Invited Talk, Inheriting Humanoid Systems from Remote-Brained Robotics, Masayuki Inaba, Uni. Tokyo, Japan |
| 14.10 | C2TAM: A Cloud framework for Cooperative Tracking And Mapping, Javier Civera |
| 14.25 | Demo 2: Cooperative Tracking And Mapping |
| 14.40 | Demo 3: Rapyuta |
| 15.00 | Coffee break and Posters |
| 15.30 | Invited Talk, Web-enabled and experience-based learning of everyday manipulation tasks, Moritz Tenorth/Michael Beetz, TU Bremen, Germany |
| 15.55 | A Cloud Robotics Architecture to Foster Individual Child Partnership in Medical Facilities, Joan Navarro |
| 16.10 | Lighting Session/General Discussion |
| 17.00 | End |
Aims and Scope
Cloud Robotics is an emerging field of robotics rooted in cloud computing, cloud storage, and other Internet technologies centered around the benefits of converged infrastructure and shared services. Driven by the rapid progress of wireless technology and availability of data centers, recent years have seen the rapid development of software frameworks and services that allow robots to leverage the powerful computational, storage, and communications infrastructure of modern data centers - the giant server farms behind the likes of Google, Facebook, and Amazon - for robotics tasks and robot learning.
This two-part workshop will provide an overview of the state-of-the-art, provide a platform for discussion and exchange of ideas, and allow researchers to gain hands-on experience with state-of-the-art Cloud Robotics tools.
Part 1 (Talks): Invited talks and contributed paper presentations.
Part 2 (Live-demos): Hands-on session, live demonstrations, and discussions
This part will provide an opportunity to try out and experiment with a state- of-the-art Cloud Computing Platforms and services. It will also include live demonstrations of Cloud Robotics services.
Motivation
Cloud Robotics has rapidly gained momentum with initiatives by companies such as Google , Willow Garage, and Gostai as well as more than a dozen active research projects around the world [Cloud Robotics Survey, Goldberg et al.].
RoboEarth [Waibel et al., 2010] pioneered the idea of a World Wide Web for robots - a large database that allows robots to upload, store, and share their knowledge about objects, environments, and actions at different levels of abstraction. This database enables robots with potentially different hardware and software to download, use, and improve shared knowledge.
The rapid progress of wireless technology and availability of data centers is now adding a new dimension to the World Wide Web. Today robots can not only use the web as a source of shared information, but also as a powerful computational resource. The usage of the cloud as a computational resource is further strengthened as wireless communication data rates increase consistently. Additionally, widely popular robotic sensors are finding their way into more and more mobile devices, increasing the demand for cloud based robotic applications [PrimeSense]. Examples include online services for map building, computing 3D models [Kinect@Home ], online grasp planning [Kehoe1 et al., 2013] and robot assisted remote health-care [Chen et al., 2013]. Such cloud based services for robots hold the potential for lighter, cheaper, and more powerful robots.
Statement of objectives
- Provide an overview of the current status of this fast-moving field.
- Bring together researchers to exchange ideas, and to identify and discuss the field's key scientific challenges and opportunities.
- Provide a hands-on experience with cloud robotics tools for attendees.
- Learn from industry experts and promote an exchange between industry and academia.
- Live demonstrations of cloud services for robots.
Intended audience
The primary audience is researchers from robotics, computer vision, machine learning, and AI working on real-world large-scale problems in perception, manipulation, and exploitation of semantic knowledge, including:
- mobile robots (e.g., outsourcing navigation, SLAM, planning tasks to the cloud)
- robot co-workers (e.g., natural language interaction using speech-to-text services)
- knowledge representation for robotics (e.g., leveraging large online knowledge
databases) - formalization of actions, plans, and task assignment in robot teams (e.g.,
collaborative mapping)
Attendees will come away with an understanding of the current status and challenges of Cloud Robotics along with an experience of how to use Open Source tools to leverage Internet resources for their robots.
List of topics
Topics of interest include, but are not necessarily limited to:
- Cloud Computing for Robots. E.g., how can robots take advantage of the rapid increase in data transfer rates to offload tasks without hard real time requirements?
- Leveraging existing resources. E.g., How can robots use the powerful computational, storage, and communications infrastructure of modern data centers - the giant server farms behind the likes of Google, Facebook, and Amazon - for robotics tasks and robot learning. To what extent can existing IaaS, PaaS, and SaaS resources be used for robotics?
- Online Knowledge Bases. E.g., How will/should online knowledge bases grow? What are the processes leading to creation of a substantial knowledge base useful in real applications, i.e. the bootstrapping of the KB?
- Reuse of knowledge. E.g., How can a robot decide which knowledge (e.g., map or skill) to reuse in a new situation?
- Massive data for robots. E.g., How can robots exploit the vast amounts of data that general-purpose repositories (e.g., Google or Youtube) and robotic-specific ones (e.g., RoboEarth) offer?
- Web Services for robots. E.g., Can we create services for things like reasoning in an uncertain world or semantic knowledge? What existing cloud services for humans can be leveraged for robotics?
- Cloud Robotics design choices. E.g., What architectures provide the optimal trade- offs between content aggregation and caching vs. accessibility and scalability vs. response time for robotics applications? What are useful metrics and optimal trade- offs between on-board computation and the use of cloud services?
Call for papers
Submission and review process will follow the standard IROS main conference submissions. The page limit is 6 pages. Papers should be prepared in the IROS format detailed both in the workshop and main conference websites. Papers violating the paper length and formatting requirements may be rejected without review.
Extended versions of select papers may be included in an upcoming Special Issue on Cloud Robotics (currently under negotiation).
Papers should be submitted through the following site: http://cmt.research.microsoft.com/CR2013/
Live demonstrations are also welcome, simply contact the organizers.
Organizers
- Markus Waibel, ETH Zurich (main organizer)
- Ken Goldberg, UC Berkeley
- Javier Civera, Uni. Zaragoza
- Alper Aydemir, NASA JPL
- Matei Ciocarlie, Willow Garage
- Mohanarajah Gajamohan, ETH Zurich
Contact
Markus Waibel, mwaibel -at- ethz.ch