More on "easy" problems vs. hard

[cmdr_zoom]

"Pfft, how hard can that be, any child can do that" - words that would haunt roboticists through the late 20th and 21st centuries. (The usual answer involves five years, a big grant, and a lot of researchers sitting in corners twitching.)

Consider "easy" tasks like "pick up the red ball, without dropping or crushing it, and throw it to this other

Comments

[seawasp]

Indeed. Anyone working in automated metrology and robotics understands this. We have an incredible amount of dedicated hardware and firmware in our heads whose TOTAL job is to analyze spatial data and convey us information about the world, and another incredible amount of data and processing capability dedicated to understanding what the world MEANS.

Easy challenge: pull the "cut lever" on each train car. Took us 5 years and millions of dollars. Resulted in multiple patents.