“…Finally, in the 'how data is input' area, all of the above work uses videos or motion capture of humans performing tasks (Bentivegna, 2002), (Brand, 1997), (Jenkins, 2000), (Kang, 1993), (Kuniyoshi, 1994), (Pangburn, 1994), (Paul, 1996), (Pollard, 2002), (Schaal, 1997), (Siskind, 2000) or teleoperation using sensors hooked to a simulated world (Bailey, 1998), (Bentivegna, 2002), (Ehrenmann, 2002), (Ogata, 1994), (Tung, 1995).…”