Memory and mental time travel in humans and social robots.

From neuroscience, brain imaging, and the psychology of memory we are beginning to assemble an integrated theory of the brain sub-systems and pathways that allow the compression, storage and reconstruction of memories for past events and their use in contextualizing the present and reasoning about the future—mental time travel (MTT). Using computational models, embedded in humanoid robots, we are seeking to test the sufficiency of this theoretical account and to evaluate the usefulness of brain-inspired memory systems for social robots. In this contribution, we describe the use of machine learning techniques—Gaussian process latent variable models—to build a multimodal memory system for the iCub humanoid robot and summarise results of the deployment of this system for human-robot interaction. We also outline the further steps required to create a more complete robotic implementation of human-like autobiographical memory and MTT. We propose that generative memory models, such as those that form the core of our robot memory system, can provide a solution to the symbol grounding problem in embodied artificial intelligence.Funding. The preparation of this chapter was supported by funding from the EU Seventh Framework Programme as part of the projects Experimental Functional Android Assistant (EFAA, FP7-ICT-270490) and What You Say Is What You Did (WYSIWYD, FP7-ICT-612139) and by the EU H2020 Programme as part of the Human Brain Project (HBP-SGA1, 720270; HBP-SGA2, 785907). Acknowledgements. The authors are grateful to Paul Verschure, Peter Dominey, Giorgio Metta, Yiannis Demiris and the other members of the WYSIWYD and EFAA consortia; to members of the HBP EPISENSE group; and to our colleagues at the University of Sheffield who have helped us to develop memory systems for the iCub, particularly Luke Boorman, Harry Jackson and Matthew Evans. The Sheffield iCub was purchased with the support of the UK Engineering and Physical Sciences Research Council (EPSRC)