Taming robots with clicker training : A solution for teaching complex behaviors Fr´ ed´ eric Kaplan (1), Pierre-Yves Oudeyer (1), Enik¨o Kubinyi (2) and ´ Ad´ am Mikl´osi (2) (1) Sony CSL Paris, 6 rue Amyot 75005 Paris, France. (2) Department of Ethology, E¨ otv¨osUniversity,G¨od,Hungary kaplan@csl.sony.fr, oudeyer@csl.sony.fr, kubinyie@freemail.hu, miklosa@ludens.elte.hu Abstract. In this paper we want to propose the idea that 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. After describing our implementation of clicker training on an enhanced version of AIBO, Sony’s four-legged robot, we argue that this new method can be a promising technique for teaching unusual behavior and sequences of actions to a pet robot. 1 Introduction The recent years have been characterised by the ex- pansion of animal-like entertainment robots (Kusa- hara, 2000; Druin and Hendler, 2000; Kaplan, 2001). The AIBO, commercialized by Sony in 1999, was the first product of this new generation of ro- bots (Fujita and Kitano, 1998). Its main originality is to be both an autonomous robot and a digital creature. As an autonomous robot, the AIBO is designed to move and behave in unknown environ- ments. But as a digital creature, it is not meant to perfom service tasks for its owner. It will not do something ”useful” and for this very reason, it may actually be a companion with whom it is pleasant to interact (Kaplan, 2000a). Interactions with current entertainment robots is still very restrained. These robots act au- tonomously without paying much attention to their owner. One of the challenges and pleasures in keep- ing a real pet, like a dog, is that the owner has to ”tame” it. A dog owner is proud when he has the impression that his pet changes its own behavior according to his teaching. We believe this is also a way for an interesting relationship to emerge be- tween an entertainment robot and its owner (Daut- enhahn develops a similar argument in (Dauten- hahn, 1999))). For this reason, a growing number of research groups are currently focusing on teach- ing techniques for autonomous robots (Billard et al., 1998; Roy, 1999; Kaplan, 2000b; Fujita et al., 2001). This paper focuses on a method for teaching ac- tions to an animal-like entertainment robot. Of course, the simplest way would be to allow the owner to program directly new actions for the ro- bot. But for the purpose of entertainment robotics it would be much more interesting if this teaching would take place only through interactions, as it does with real pets. We believe that, for this matter, a collaboration between robot builders and ethologists can be inter- esting. Exchanges between ethologists and robotic engineers have several times proven to be fruitful in the past (see (Steels and Brooks, 1994), (Arkin, 1998) and (Webb, 2000) for instance). But apart from the exception of Blumberg’s team at MIT Me- dia Lab (Blumberg et al., 1996; Blumberg, 1997; Yoon et al., 2000), the field of animal training has