The Playground Experiment aims at showing how a robot equipped with an intrinsic motivation system, and in particular artificial curiosity, can explore its environment autonomously and develop skills which were not pre-specified, and with an increasing complexity for an extended period of time.
Links: the homepage of the playground experiment
Participants: Pierre-Yves Oudeyer, Frédéric Kaplan, Verena V. Hafner and A. White
The Robot’s Playroom was designed to offer new learning opportunities and exploration spaces for the Sony AIBO, and to test curiosity-driven learning algorithms. Thanks to this experiment the robot can now draw, ride a bike, control switches, pick up everyday objects, watch itself in a mirror, and more …
Team: Frédéric Kaplan, Pierre-Yves Oudeyer,
Collaborations: Martino d’Esposito, and ECAL Design Students
Question: Can we train a robot like a dog?
Some techniques used for animal training might be helpful for solving human robot interaction problems in the context of entertainment robotics. We present a model for teaching complex actions to an animal-like autonomous robot based on “clicker training”, a method used efficiently by professional trainers for animals of different species. Our implementation of clicker training on an enhanced version of AIBO, Sony’s four-legged robot, suggests that this new method can be a promising technique for teaching unusual behavior and sequences of actions to a pet robot.
Participants: Frédéric Kaplan, Pierre-Yves Oudeyer, and Eniko Kubinyi and Adam Miklosi from the Ethology Department of the Eötvös University (Budapest, Hungary)
Question: What are the mechanisms needed to learn the meaning of new words in natural social contexts? How are the social regulation mechanisms involved in language learning? How can one draw the attention of a robot towards particular aspects of their environment? What are the interactions between acquisition mechanisms and language evolution?
We investigate the mechanisms that enable humans and robots to learn new words and to use them in appropriate situations. We have built a number of robotic and computational experiments studying the mechanisms of concept formation, joint attention, social coordination and language games, and articulating the roles of learning, physical and environmental biases in language acquisition. The unifying theme of all these experiments is development: we explore the hypothesis that language can only be acquired through the progressive structuring of the sensorimotor and social experience. These experiments are described in the papers below.
Interaction between an AIBO and its human trainer in the
AIBO’s First Words experiment.
Participants: Luc Steels, Masahiro Fujita [Sony DCL Tokyo], Frédéric Kaplan, Angus McIntyre, and Pierre-Yves Oudeyer
Question: Do dogs see AIBO, Sony’s four-legged robot, as a conspecific?
We are conducting a series of exploratory studies on animal robot interactions in collaboration with the ethology group of the University of Eötvös University (Hungary). The purpose of these experiments is to investigate, from an ethological point of view, how much dogs see AIBO as a conspecific. The questions adressed are: what is the influence on the dog’s reactions of movement, smell, presence or absence of eyes, sounds, etc.
Two kinds of situations are tested. In the first one, puppies and adult dogs interact freely with the robot. In the second one, we organise a situation of implicit competition in which the dog has to defend a piece of meat against the arrival of the robot. Comparative studies are done with a remote control car and a real puppy. The results are being analysed and will be published in the near future.
Hungarian Vizsla/robot competition (first pilot study, April 2000)
Mops/robot interaction (first pilot study, April 2000)
Dog attacks AIBO movie
The horrible screams that you hear at the end of the movie were made by the experimenters, who were startled to see the dog attack the AIBO.
This was the first time that the AIBO was attacked, but it was not the last. During the course of the experiment, the AIBO was sometimes knocked over, bitten and chewed. It is still in perfect working order, and shows no visible signs of damage.
Nevertheless, we strongly advise you not to try anything similar with your AIBO. AIBO is strongly built, but it contains many delicate components that could be easily damaged. Your warranty will not cover you if AIBO is damaged in this way.
No animals were hurt or mistreated in any way during the course of this experiment.
Participants: Frédéric Kaplan, and Eniko Kubinyi and A. Miklosi from the Ethology Department of the Eötvös University (Budapest, Hungary)
