SPENCER · Social situation-aware perception and action for cognitive robots

SPENCER will break new ground for cognitive systems in populated environments. While there is an increasing focus on making robots more socially aware, related approaches are still limited in their capacity to perceive, model and learn human social behavior and respond with appropriate actions in real-time from mobile platforms. SPENCER will systematically address these problems and advance the fields of perception of individual humans and groups, normative human behavior learning and modeling, socially-aware mapping, and socially-aware task, motion and interaction planning under real-world conditions. In particular, by addressing these problems simultaneously in a multi-disciplinary project team, we will exploit synergies which will enable us to design cognitive systems that reach new levels of autonomy, efficiency, robustness, and safety in populated environments, optimizing the trade-off between objective and subjective performance measures. The project is motivated by an end-user market-pull and a technology-push. End-users seek innovative solutions to guide people and efficient mobile information provision. Robotics research meanwhile, has become suitably advanced to start considering humans as more than objects but as people with relationships, social rules and culturally diverse backgrounds. To this end, SPENCER will combine robotics research with social signal processing and socio-psychological user studies that will guide the learning process of socially normative robot behaviors into the relevant directions. The SPENCER consortium includes a large European airline as end-user that will deploy the SPENCER robotic demonstrator for the purpose of smart flow management of transfer passengers at the Amsterdam Schiphol Airport. This deployment has a large exploitation potential and is an excellent benchmark of the research developed in SPENCER given the demanding nature of airports as highly populated real-world environments.