Robots are designed to help people perform tasks more accurately, but the best way to perform a task may not always be the best way to learn it. I will share research on robotic systems that encourage people to explore varied movement strategies, including real-time assistance that adapts to performance and nudges users toward underexplored solutions. These results suggest that robots can support learning by shaping not only how much assistance a person receives, but also the variety of experiences they encounter. I will conclude by discussing an open challenge: how should we adapt task conditions and schedule different task variations—particularly for dynamic activities such as manipulating tools—to promote long-term learning?
Katie Fitzsimons | Mechanical Engineering