What robots are

As a researcher interested in human-technology interface, I have always been intrigued by robotics. My work in the field has been limited to several projects here and there: seminars/workshops organization about it, the writing of a research grant about human-robot interaction for game design (a project I conducted two years ago in France) and tutoring design students at ENSCI on a project about humanoid robots. If we include networked objects/blogjects in the robot field (which is not obvious for everyone), Julian and I had our share of work with workshops, talks and few prototypes at the near future laboratory.

Robolift, the new conference Lift has launched with French partners, was a good opportunity to go deeper in the field of robotics. What follows is a series of thoughts and notes I've taken when preparing the first editorial discussions.

"I can't define a robot, but I know one when I see one" Joseph Engelberger

The first thing that struck me as fascinating when it comes to robots is simply the definition of what "counts" as a robot for the people I've met (entrepreneurs in the field, researchers, designers, users, etc.) For some, an artifact is a robot only if it moves around (with wheels or foot) and if it has some powerful sensor (such as a camera). For other, it's the shape that matters: anthropomorphic and zoomorphic objects are much more "robotic" than spheres.

Being fascinated by how people define what is a robot and what kind of technical objects count as a robot, I often look for such material on the World Wide Web. See for example this collection of definitions and get back here [yes this is a kind of hypertext reading moment]. The proposed descriptions of what is a robot are all focused on different aspects and it's curious to see the kinds of groups you can form based on them. Based on the quotes from this website, I grouped them in clusters which show some characteristics:

What do these clusters tell us anyway? Of course, this view is limited and we cannot generalize from it but there are some interesting elements in there:

  • The most important parameters to define a robot revolve around its goal, its mode of operation, its physical behavior, its shape and technical characteristics.
  • Technical aspects seems less important than goal and mode of operation
  • Defining a robot by stating what it isn't is also relevant
  • While there seems to be a consensus on the appearance (humanoid!), goals and modes of operations are pretty diverse.
  • Certain aspects are not considered here: non-humanoid robots, software bots, block-shaped, etc.

It's interesting to contrast these elements with results from a research study about the shapes of robot to come (which is a topic we will address at the upcoming Robolift conference).

I found this on BotJunkie few months ago, I ran across this interesting diagram extracted from a 2008 study from the Swiss Federal Institute of Technology in Lausanne:

"Researchers surveyed 240 people at a home and living exhibition in Geneva about their feelings on robots in their lives, and came up with some interesting data, including the above graph which shows pretty explicitly that having domestic robots that look like humans (or even “creatures”) is not a good idea, and is liable to make people uncomfortable."

Some of the results I find interesting here:

"Results shows that for the participants in our survey a robot looks like a machine, be it big or small. In spite of the apparent popularity of Japanese robots such as the Sony Aibo, the Furby and Asimo, other categories (creature, human, and animal) gathered only a small percentage. (...) The preferred appearance is therefore very clearly a small machine-like robot"

These results are consistent with another study by Arras and Cerqui. These authors looked at whether people would prefer a robot with a humanoid appearance. They found that "47% gave a negative answer, 19% said yes, and 35% were undecided"".

Why do I blog this? Preparing the upcoming robolift conference and accumulating references for potential projects about human-robot interactions.