smagt@fwi.uva.nl (Patrick van der Smagt) (05/18/91)
I am trying to model motion in a quite wide perspective, but one can narrow it down to hand-eye coordination. When I look at biological mechanisms (e.g., myself), I get the impression that motions are based on the fact that one assumes all moving objects (including onself) to have a constant velocity. Correction for the fact that things have accelerations are realised by fast sensory data processing. Another indication for this assumption I find in Lee's article [1980] in which he invesigates time-to-contact. He reports (in another article, I believe) that subjects correctly estimate the time left before an approaching object hits them, under the assumption that the velocity of the object is constant. What I am looking for are other indications for my assumption. Do biological systems only `calculate' the first positional derivative from their sensory input? Patrick van der Smagt References: =========== D. N. Lee, The optic flow field: The foundation of vision. Phil. Trans. R. Soc. Lond. B 290, pp. 169--179, 1980.
jgk@osc.COM (Joe Keane) (05/21/91)
My feeling is that acceleration is handled at a pretty low level. Suppose you want to catch a baseball. You have to decide where it's going to go, and to do this you have to account for the effect of gravity. In my experience, this is done automatically by whatever tracking hardware is built in. From an evolutionary point of view, i think you can see that an animal which didn't account for gravity would be at a disadvantage. Another application is driving. A useful skill is to be able to predict the stopping position given my current velocity and deceleration, which is roughly constant. Or, given the current velocity and acceleration of my car relative to the one in front of me, determine whether their positions will overlap at some time in the future. I find that these computations are done quickly and accurately without conscious effort. Now it's hard to say what of this is learned, as opposed to hard-wired and instinctive. Maybe we could raise some kids in zero gravity. But it seems reasonable to assume that these skills are important even to animals which don't drive cars. -- Joe Keane, amateur mathematician jgk@osc.com (...!uunet!stratus!osc!jgk)