I recently became aware of an effort by ISO (International Organization for Standardization) to define a standard for domestic service robots -- more specifically, ISO-13482 "Safety requirements for non-medical personal care robots." I must confess having mixed feelings about this development. On one hand, it is exciting that the personal robotics revolution is near-enough at hand to warrant the definition of a standard -- there are many standards for industrial robots (eg. ISO-10218 and ISO-9409), but none for domestic personal robots. On the other hand, I'm a bit concerned that a somewhat-binding international standard is being developed prematurely and in a rather closed-door fashion -- issues upon which I will elaborate below. Thankfully, there will be plenty of discussion at IROS 2010 (Taipei, Taiwan in mid-October) at the "Workshop on Standardization for Service Robots." Lack of resources will likely preclude my attendance, so perhaps someone can fill us in after the fact...?
Most of my knowledge about ISO-13482 is second- or third-hand, as I have not yet figured out how to receive a gratis copy of the preliminary standard. Presumably, personal care robots will fall into four different classes based on the level of physical interaction with users, and thus will possess different safety requirements. Here are my own (largely unsubstantiated) interpretations of the classes:
It is abundantly clear to people working on domestic robots (like myself) that we do not yet know what design elements are necessary or sufficient for safe operation (hardware, software, and everything else). This space is under exploration by industry and academia alike. For example, I am the only researcher I know of that has had a large autonomous mobile manipulator (EL-E) operating in their own home -- I certainly do not understand what it would take to make our robot intrinsically safe, so I doubt that others lacking such experiences could accurately identify the necessary elements. [True, this argument is based on several tacit assumptions that do not take into account my own ignorance.]
To provide a hypothetical example, say that the standard required ultrasonic or infrared ranging elements around the base of the robot. This would add undue cost and complexity to a robot employing a frame-rate 3D imager such as the most recent low-cost depth cameras.
I was shocked to find out that ISO-13482 was already in a later-stage of development, and yet none of the myriad personal robotics researchers I know were aware of its existence -- and more importantly, that they were not consulted about its creation. There would seem to be a lot of untapped experience and expertise not being considered.
It is also a bit disconcerting that I'm unable to locate gratis copies of the proposed standard online. Considering that other ISO standards (eg ISO-9409) retail for ~$50 USD, I'm less than optimistic about locating a gratis version and posting it here on Hizook. Perhaps this is just the typical manner in which these standards are developed, but I'm still disappointed... It seems so antithetical to recent trends in openness.
Let's curb the negative undertones... I'm still largely in favor of this development. It seems to imply that there is definite progress and interest in personal robots. As long as the standard is fluid (updated to reflect advances in research and application), it could be a useful reference to all robot builders (hobbyists, researchers, and industry) -- it would be nice to have design guidelines already vetted in household environments.
I mentioned the IROS 2010 (Taipei, Taiwan in mid-October) "Workshop on Standardization for Service Robots." I'm curious about what will be discussed, though I will likely be unable to attend. Below is a copy of the call for papers in its entirety (for archival and reference purposes).
IROS 2010 Workshop on Standardization for Service Robots: Current Status and Future Directions
Date: October 18th (Monday), 2010
Place: Taipei International Convention Center, Taipei, Taiwan
The full day workshop aims to introduce the current status on service robot standardization activity and to encourage discussion and mutual understanding on elements for realizing effective service robots in real world. Invited sessions with representative professionals who are actively working on standardization of service robots and related fields are organized, as well as general sessions consisting of submissions of related interests. Panel discussion on future direction of establishing effective standards for service robots will also be held.
Motivation and Objectives
Recently, numbers of activities have started to establish standards for service robots in various international organizations such as IEEE, ISO, ITU-T, OGC and OMG. Following the first standard for robotic components in 2008 (Robotic Technology Component 1.0, OMG), the world's first standard aimed specifically for service robots has just been published in February 2010 (Robotic Localization Service 1.0, OMG).
With the rapid progress in robotics and IT technology, the robot systems are fast becoming larger and more complicated, Moreover, as recent notions such as structured or smart environment suggests, researchers are now heading toward spatial expansion of the idea of robots; that is, robots are no longer limited within a single body but the total system integrated over environmental facilities, multiple cooperative robots and external information resources. At the same time, it is getting more and more difficult to build robots that can effectively operate in real environments. Thus, both for starting up a new industry for service robots and for the ease to focus on specific research fields such as HRI, interconnectability and reusability of various elements that make a complete robotic system are essential. For a long time, service robots had been thought to be in 'research' stage. But it is now the time to head toward industrialization and standardization.
The aim of this workshop is to introduce the ongoing activities for standardization and to seek future elements and directions for service robots. In addition to current works on robotics, trends in related areas that are considered to be essential for service robots to serve in our daily environments such as ubiquitous network, sensor network or geographical information systems will be introduced.
The topics of interest include, but are not limited to:
- Standardization on service robotics and related fields
- Interoperability with related systems
- Middleware and platform for service robots
- Design pattern for service robots
- Robotic service description language
- Robotic service performance measurement
- Proposal of new items for standardization
Manuscripts must be in English, max. 6 pages in the standard IROS format, converted to PDF.
Submit by sending the pdf to: nishio _ ieee.org (replace '_' (underscore) with atmark)
Jul. 01, 2010: Paper submission deadline
Jul. 20, 2010: Notification
Jul. 27, 2010: Camera ready deadline
Oct. 18, 2010: Workshop
The workshop will be of particular interest to robotic engineers and researchers that work in the general areas of service robots. It will also be an opportunity to engineers and researchers working in the field of sensor network, ubiquitous network or geographic information system to learn the state of the art in service robotics and how these systems are to be used in various robotic services, and at the same time, to utilize robotic technology in their fields. Standardization is especially important for the emerge of the new service robot industry and therefore this workshop can attract the interest of the general service robot community.
* Shuichi Nishio
Intelligent Robotics and Communication Laboratories
Advanced Telecommunications Research Institute International (ATR)
email: nishio _ ieee.org
* Young-Jo Cho
Principal Member of Engineering Staff
Robot/Cognitive System Research Department
Electronics and Telecommunication Research Institute (ETRI)
email: youngjo _ etri.re.kr
* Miwako Doi
Humancentric Laboratory, Corporate Research & Development Center
email: miwako.doi _ toshiba.co.jp
* Wonpil Yu
Robot/Cognitive System Research Department
Electronics and Telecommunications Research Institute (ETRI)
email: ywp _ etri.re.kr
Note: please replace '_' (underscore) with atmark in email addresses
For questions/submissions, please contact: Shuichi Nishio [nishio _ ieee.org; replace '_' (underscore) with atmark]