Perception (Internal representation of the world built from real-time sensory information) : Development of innovative methods and algorithms able to generate and interpret the information-rich representations of complex, dynamic, uncertain environments required for the next generation of autonomous systems

Reactive Decision Making (Reacting to perceptions, linking 'sensing' through 'thinking' to 'action’) : Innovations in the areas of decision-making under uncertainty, multi-platform co-operation, and the application of machine learning techniques to produce adaptable decision-making strategies that increase substantially the range of operation and robustness of autonomous systems

Architectures (Describing how algorithmic components should be combined and how control will be achieved) : Work to allow autonomous system architectures to be reasoned about and to determine whether, and under what conditions, they have a range of desirable properties, including heterogeneity, scalability, robustness, openness, flexibility, and reconfigurability. A measure of success will be the ease with which developments from other themes can be integrated and tested within the developing architectural concepts and models