Evolutionary Robotics, Artificial Life, Adaptive Systems, Ethical Implications
My research is inspired by a vision of self-sufficient robot collectives that adapt to tasks and environments that were incompletely known at design time. In particular, I focus on the use of evolutionary algorithms for the on-line development and adaptation of robot morphologies and controllers to achieve this vision. This combines the use of evolution for optimisation (as seen in most evolutionary robotics work) and for adaptation to the environment (as studied in artificial life and, in fact, biology).
An important aspect of my research considers the ethical implications of autonomously adapting machines: how can we make sure that they do what we want them to, and do not do what we don't want?