Etho-robotics: What kind of behaviour can we learn from the animals?

Abstract

The trends of robotics changes as service robots gain more ground and become part of our daily lives. Service or social robots have to function in less controlled and more dynamically changing environments than industrial robots, and the users of social robots are generally less technologically literate than the users of industrial robots, therefore social robots have to be able to interact naturally with humans and to fit in the human social environment. Social robotics faces new challenges that require an interdisciplinary approach. In contrast to previous approaches where the communication and behaviour of social robots were based on human-human interactions, ethorobotics offer a new direction. Today robotics is not advanced enough to reach the physical and cognitive capabilities of humans thus human-animal interaction can serve as a better model for designing the behaviour of social robots. Human-dog relationship is a good example for this paradigm as dogs have similar roles as social robots will have in the future. Dogs acquired social cognitive skills during domestication that enhances the interspecific relationship with humans and helps their interactions. Etho-robotics research uses ethological principles and methods to derive complex behavioural models which can be transcribed to mathematical form and implemented into robots. Human-robot interaction studies can be conducted to evaluate and refine the implemented models. Etho-robotics research was already used to create behavioural models for attachment and for multiple aspects of human-robot interactions. The application of Fuzzy Rule Interpolation methods fits well the conceptually “spare rule-based” structure of the existing descriptive verbal ethological models, as in case of the descriptive verbal ethological models the “completeness” of the rule-base is not required. The main benefit of the FRI method adaptation in ethological model implementation is the fact, that it has a simple rule-based knowledge representation format. Because of this, even after numerical optimization of the model, the rules are still “human readable”, and helps the formal validation of the model by the ethological experts. On the other side due to the FRI base, the model has still low computational demand and fits directly the requirements of the embedded implementations.