Adaptive Robotic Ecologies

In this talk I will provide an overview of my research experience, from software engineering and human-robot interaction for ubiquitous robotic systems, to my most recent focus on cognitive robotic ecologies in the EU project RUBICON. The project has developed learning solutions which yield cheaper, adaptive and efficient coordination of robotic ecologies, i.e. systems of heterogeneous robotic devices working together in applications such as ambient assisted living and in-hospital transport systems. The RUBICON approach builds upon a unique combination of methods from cognitive robotics, machine learning, plan-based control, and wireless sensor networks. I will illustrate the innovations advanced by RUBICON on each of these fronts before describing how the resulting techniques have been integrated and applied in a range of use cases. The resulting systems are able to provide useful services and pro-actively assist the users in their activities. RUBICON learns through an incremental and progressive approach driven by the feedback received from its own activities and from the user, while also self-organizing the manner in which it uses available sensors, actuators and other resources in the process. I will summarise the lessons learned by adopting such an approach and will outline promising directions for future work.