Intelligent agents that learn from their past experiences

Research Project, 2020 – 2025

Embodied Artificial Intelligence is a multidisciplinary area that requires the cooperation of different fields such as computer science, engineering, robotics and dynamical systems. This PhD thesis will develop a novel learning algorithm to allow high-level intelligence, such as problem-solving and reasoning to be applied to real-world physical systems, e.g. robots. Mainly, this work will be focused on investigating learning methods on the semantic aspects of intelligence to develop general purpose solutions for robotic applications.

The goal of this PhD thesis is to develop compact and flexible model representations to allow robots the transference of their past experiences to current situations. These compact models should be human-readable to provide adequate feedback to the person interacting with the intelligent agent, e.g. a robot. This feedback should provide information to users, in an efficient form, about possible hazardous situations and potential errors, by predicting ahead-of-time the actions performed by either a human or a robot to infer their potential consequences. Thus, the communication between the users and the robots should be meaningful and bi-directional.