Robotics I-CAT - Online Article

Introduction

Progress in various technologies - hardware, software and artificial intelligence - brings the vision of personal robotics closer. While much work is done on the technology many open questions needs to be answered about the social impact of having a personal robot in a domestic environment. Philips Research is investigating technical and social aspects of user-interface robots in an 'Ambient Intelligence' environment, in which technology is seamlessly integrated in your life.

Project Goal

Our goal is to stimulate Human-Robot Interaction research by building a research community through supporting a common hardware and software platform.

iCat

iCat is a research platform for studying human-robot interaction topics. The robot is 38 cm tall and is equipped with 13 servos that control different parts of the face, such as the eyebrows, eyes, eyelids, mouth and head position. With this setup iCat can generate many different facial expressions - happy, surprise, angry, sad - that are needed to create social human-robot interaction dialogues.

A camera installed in iCat's head can be used for different computer vision capabilities, such as recognizing objects and faces. Each foot contains a microphone to record sounds, perform speech recognition and to determine the direction of the sound source. A speaker and soundcard are installed to play sounds and speech. Finally, touch sensors and multi-colour LEDs are installed in the feet and ears to sense whether the user touches the robot and to communicate further information encoded by coloured light. For instance, the operation mode of the iCat (e.g. sleeping, awake, busy, listening) can be encoded by the colour of the LEDs in the ears.

Social Intelligence

Various user studies on social intelligence aspects of user-interface robots have been performed in our test facility HomeLab at Philips Research Eindhoven, the Netherlands. HomeLab is a smart home concept with various means to observe the behavior of users. During some of our studies we investigated the perceived personality of the iCat by letting users interact with the iCat during a game setting (TicTacToe) or task setting (programming a VCR). These studies show measurable differences in effectiveness and enjoyability of the tasks the users had to perform, depending on iCat's personality.

OPPR

Programming user-interface robots like the iCat is easy with our Open Platform for Personal Robotics ('OPPR') software development environment. We have developed the Dynamic Module Library ('DML') that allows you to easily build software components that run on several PCs.

OPPR contains a Robot Animation Editor for creating believable robot animations. This graphical tool gives you precise control over every individual servo and multi-colour LED. Once you have developed animations they can be used to control the iCat using the Robot Animation Engine. This engine blends at runtime pre-scripted animations and robot behaviors (sensor-actuator loops). With this setup you can create separate animations and control algorithms for head movements, facial expressions, eye-blinking and lip synchronization. OPPR contains a ready-to-use library with animations that can be used for building human-robot dialogues. A scripting engine makes developing these dialogues easy.

OPPR can be both used by beginners and advanced programmers. Scripting and graphical tools make developing animated dialogues easy. Advanced programmers can build their own DML software components, such as vision and speech recognition components, using C++.

Concluding Remarks

When one thinks of the Ambient Intelligence, one always thinks of technological factors, regardless if he/she is engineer, businessman, doctor or journalist. So, we just did not want to be narrow-minded and limit our discussion only to those issues that are directly connected with the scope of the seminar. Besides, it is clear that developing new technologies is necessary prerequisite for creating Ambient Intelligence. The current projects and already developed prototypes show that Ambient Intelligence is not that far as it first looks like, at least concerning the technology itself. But, even if many scientists are optimistic about the progress in this field (despite many unanswered questions), it is financing these projects that may present a big obstacle for further researches.On the other hand, we shall be wrong if we claim that every technological progress is always good and desired. Therefore, we here get into the sensitive area of social acceptance.

Social issues are here broadly discussed and that due to two reasons. Firstly, without social acceptance there will be no Ambient Intelligence. Secondly, the title of the report contains the word “society”, which somehow suggests to take social issues into account when discussing the scenarios. I dare to say that the social acceptance is crucial for the AmI, and thus even more important than technological issues. Security and privacy were often mentioned in this context, and we agreed that without providing them AmI will be sentenced to fail. Nevertheless, it is not only about security and privacy. Some people do not simply like the idea of AmI itself, not only ist possbile consequences. But, speaking of AmI, we must carry in mind that this will not be a single product, but AmI comprises many features which base on same or similar technological principles.

Michael Dertouzos

Director of the MIT Laboratory cheered the AmI saying: “There’s been 40 years of people serving machines, and now it’s time to make the machines humancentric, so they’ll serve people.” If this comes true one day, it will mean Ambient Intelligence has succeeded. Objectives are set, motivation among the scientists is high, obstacles are perceived, but there is still a lot of hard work ahead. On the other hand, world has always been changing and there is no reason for it to stop changing and make progress. Although still a vision, Ambient Intelligence has good preconditions to turn into something more real.”

About the Author:

No further information.




Comments

Manvendra Rana on 2009-03-05 21:57:36 wrote,

It was a best combination of robotics and ambient intelligence