“Generally speaking, modeling the developmental trajectory for almost any intelligent task that a 2-year-old child can perform on a robot is a potential research topic for us.”
– explains the website of the Developmental Robotics Lab at Iowa State University, US. As an example to help clarify the (somewhat ambiguous) statement above, see the lab’s work on teaching robots to ring the doorbell.
“Buttons are ubiquitous in human-inhabited environments. These simple 1-dof widgets are used to control many essential devices and mechanisms. A robot that cannot press buttons would not be able to use anything controlled by a button and thus would not be very useful.”
Professor Alex Stoytchev, and colleagues have published several papers on the subject. e.g. ‘Learning to Press Doorbell Buttons’ (in: Proceedings of the 9th IEEE International Conference on Development and Learning (ICDL), Ann Arbor, Michigan, August 18-21, pp. 132-139, 2010.)
Also see: (later research) ‘Learning to Detect Doorbell Buttons and Broken Ones on Portable Device by Haptic Exploration In An Unsupervised Way and Real-time’
Plus: A plethora of doorbell-pressing robot videos are provided here.