Cicerobot, your next museum guide
In a brief article, ANSA reports about a robot which will guide visitors in the Agrigento museum in Sicily. Cicerobot is 1.5-meter high and is equipped with wheels, a keyboard, a monitor, video camera and sensors. As many objects in this archaeological museum are very old and precious, a special emphasis was given to navigation. So far, Cicerobot only talks Italian, but it should be soon fluent in other languages. And as it is able to collect data from different sensor locations, it also plays the role of a security guard. Finally, it will also offer virtual tours of the museum via Internet.
Here is the introduction from ANSA.
A flaming-red robot will soon be guiding tourists round a Sicilian archaeological collection. The 1.5-metre-high robot, named Cicerobot by his creators, is kitted out with wheels, a keyboard, a monitor, video camera and sensors, enabling him to steer visitors safely through the rooms of Agrigento's Regional Archaeological Museum.
"Cicerobot is able to plan out tours in accordance with the needs of individual visitors," explained Antonio Chella, head of the Palermo University robotics laboratory that invented the mechanism.
Here is a picture of the robot (left) and a 3D view of its conceptual space (CS) (right). (Credit: University of Palermo)
And there is a 3D representation of the robot environment. (Credit: University of Palermo)
In another short article, the Guardian also reports about Cicerobot.
Harris Dindo, part of the science team at Palermo University that developed the robot, said: "It uses the technique of latent semantic analysis, which means it can answer many of the questions tourists throw at it and have intelligent interaction with them."
"Its sensors allow it to guide tourists around the museum, avoiding obstacles, but the director of the museum was naturally quite concerned at first because the environment is very delicate and there are many display cases full of fragile things," Dr Dindo said. "We had to get the navigation right."
For more information about Cicerobot, it's time to turn to the Proceedings of the Symposium on Next Generation Approaches to Machine Consciousness (PDF format, 127 pages, 5.69 MB), published after the AISB’05 conference about éSocial Intelligence and Interaction in Animals, Robots and Agents" held at the University of Hertfordshire, Hatfield, UK, in April 2005.
This document includes a paper called "Planning by imagination in Cicerobot, a robot for museum tours" (pages 51-60), from which the illustrations in this post have been extracted.
Here is the introduction of this paper.
The current generation of autonomous robots has showed impressive performances in mechanics and control of movements, see for instance the ASIMO robot by Honda or the QRIO by Sony. However, these state-of the-art robots present only limited capabilities to perceive, reason and act in a new and unstructured environment.
We claim that a new generation of autonomous robots, effectively able to perceive and act in unstructured environments and to interact with people, should be aware of their external and inner perceptions, should be able to pay attention to the relevant entities in their environment, to image, predict and to effectively plan their actions. In a word, they should include some form of artificial consciousness.
This is why the designers of this robot gave a particular attention to the perception of its environment. Here is for example an image acquired by the robot camera during its tours at the Archaeological Museum. (Credit: University of Palermo)
And there is the 3D robot inner representation of the "Telamone" museum hall. (Credit: University of Palermo).
And now, let's jump to some of the conclusions about the architecture of this robot and its future.
The architecture is organized in three computational areas. The subconceptual area is concerned with the processing of data coming from the robot sensors. In the linguistic area representation and processing are based on a semantic network formalism. This area is essentially the long-term memory of the robot. The conceptual area is intermediate between the subconceptual and the linguistic areas. Here, data is organized in geometric and "gestaltic" structures in terms of conceptual spaces.
[But] the generation of the flow of consciousness requires that the robot should store in the conceptual space at time t all the information of the conceptual spaces at previous times, starting from the beginning of the robot life. This is a hard requirement to be satisfied because of the physical limitations of the robot memory. Some mechanism that lets the robot to summarize its own past experiences should be investigated.
So will robot guides completely replace human ones anytime soon? Probably not. But it seems sure that more and more museums will experiment with robots.
Sources: ANSA, March 21, 2006; Barbara McMahon, The Guardian, March 22, 2006; and various web sites
You'll find related stories by following the links below.