An ISO Standard for Personal Care (Service) Robots

ISO Robotics Standards

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...?

What I know about ISO-13482

 

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: 

  • Personal Mobility: Devices such as robot exoskeletons
  • Physical Assistant: May include fixturing and hygiene robots
  • Personal Assistant: Autonomous mobile manipulators like the PR2, EL-E, and Cody
  • Robotic Devices: General class that would include robot vacuum cleaners and the like

 


Premature Standardization

 

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.

 

Closed-Door Development Process

 

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.

 

Personally, Still a Favorable Development

 

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.

 

IROS 2010 Workshop on Standardization for Service Robots

 

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
Time: 9:00-18:00
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.

Topics:

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

Submission Guideline:

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)

Important Dates:

Jul. 01, 2010:     Paper submission deadline
Jul. 20, 2010:     Notification
Jul. 27, 2010:     Camera ready deadline
Oct. 18, 2010:     Workshop

Intended audience:

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.

Organizers:

* Shuichi Nishio
Senior Researcher
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
Chief Fellow
Humancentric Laboratory, Corporate Research & Development Center
Toshiba Corporation
email: miwako.doi _ toshiba.co.jp
* Wonpil Yu
Senior Researcher
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


Contact:

For questions/submissions, please contact: Shuichi Nishio [nishio _ ieee.org; replace '_' (underscore) with atmark]

 

Comments

There were an estimated 3,540,000 service robots in use in 2006, compared with an estimated 950,000 industrial robots.there are only a few limited models, though science fiction writers and other speculators have suggested that they could become more common in the future.

@ Dynamic Walking,

Would you happen to have a source for those numbers?  It wouldn't surprise me if there were more service robots (ie. Roomba vaccuum cleaners) that industrial robots in the world. According to the classifications mentioned above, these would presumably fall under the "robotic devices" category.  I would wager that there are very few robots in the more advanced categories.

—Travis Deyle

I seem to have made a slight oversight by including only ANSI and ISO.  The United States Occupational Health and Safety Administration (OSHA) also has a series of guidelines that are probably applicable.  They also enumerate these standards:

  • American National Standards Institute (ANSI) R15.06-1999, Industrial Robots and Robot Systems - Safety Requirements. Provides requirements for industrial robot manufacture, remanufacture and rebuild; robot system integration/installation; and methods of safeguarding to enhance the safety of personnel associated with the use of robots and robot systems. This second review further limits the potential requirements for any retrofit of existing systems, revises the description of control reliable circuitry, and reorganizes several clauses to enhance understanding.
  • American National Standards Institute (ANSI) TR R15.106-2006, Technical Report on Teaching Multiple Robots. Robotics Industries Association (RIA). Provides additional safety information relative to teaching (programming) multiple industrial robots in a common safeguarded space in an industrial setting. It supplements the ANSI/RIA R15.06-1999 robot safety standard.
  • American National Standards Institute (ANSI) B11.TR3-2000, Risk Assessment and Risk Reduction - A Guide to Estimate, Evaluate and Reduce Risks Associated with Machine Tools. Provides a means to identify hazards associated with a particular machine or system when used as intended, and provides a procedure to estimate, evaluate, and reduce the risks of harm to individuals associated with these hazards under the various conditions of use of that machine or system.
  • International Organization for Standardization (ISO) TC 184, Industrial automation systems and integration
  • International Organization for Standardization ISO 10218-1:2006, Robots for industrial environments - Safety requirements - Part 1: Robot. Robotics Industries Association (RIA). Specifies requirements and guidelines for the inherent safe design, protective measures, and information for use of industrial robots. It describes basic hazards associated with robots, and provides requirements to eliminate or adequately reduce the risks associated with these hazards.
  • Canadian Standards Association (CSA) Z434-03, Industrial Robots and Robot Systems. Applies to the manufacture, remanufacture, rebuild, installation, safeguarding, maintenance and repair, testing and start-up, and personnel training requirements for industrial robots and robot systems.
  • American Welding Society (AWS) D16.1M/D16.1, Specification For Robotic Arc Welding Safety. Identifies hazards involved in maintaining, operating, integrating, and setting up arc welding robot systems.


—Travis Deyle

From October 2011 in Automation Magazine:

The draft of the new ANSI RIA R15.06 Robot System Safety Standard was introduced at the 2011 National Robot Safety Conference in Knoxville, Tenn., last month. The new standard is an adoption of the ISO standards for robots and their system implementation (ISO 10218-1 and -2).

The draft ANSI RIA standard is being developed co-operatively with CSA, so that the final product will include the ISO, CSA and ANSI requirements, all in one package.

"The new draft adds user requirements to the ISO standards, as the ISO standards only address suppliers," said Roberta Nelson Shea, director of safety consulting services for Applied Manufacturing Technologies. "It is exciting to see that robotics is expanding to industries outside the traditional automotive applications."

—Travis Deyle

Post new comment

The content of this field is kept private and will not be shown publicly.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Allowed HTML tags: <a> <em> <strong> <cite> <code> <ul> <ol> <li> <dl> <dt> <dd> <p>
  • HTML tags will be transformed to conform to HTML standards.

More information about formatting options