Emergent Behaviours Driven by Chaotic Dynamics

We explore the emergence of behaviours in a chaos driven robot. A two neurons neural network exhibits a rich variety of dynamics when stimulated with input. We linked the phase space of the network to two behaviour activation levels. A people detector is the input to the network. The results are surprising and they contradict the expectations, which is a clear signal of the emergence of non-programmed behaviours. An analysis and discussion of the results is presented, with pointers to future development and applications.