Preposition and Gesture Project
The symbol grounding Problem:
There should be several groups of basic commands/ instructions, which are used as a base live/ built in, initial memory. That the robot receives from the programmer. The basis of each group is a simple command upon which sub-commands are built as a second layer and on top of which is built a third layer.
The robot’s “learning process” is the ability to connect the different groups at different levels among themselves, and to remember the out- come creating a “learned” memory (different then the built in/ initial memory). The more tasks the robot is given the more “memory links” he has to create.
Comments (1)
Nicholas Davis said
at 3:13 pm on Sep 9, 2008
What would these basic commands look like? Something similar to the simplified robot language we talked about last year? If so, what are some of the words that would be included? How would these be grounded? Just programmed in? Or something more like the image schema language that paul cohen was working with? If the answer is the latter, then how do we make a robot have image schemas relating to how its body functions rather than based on the proposed human image schema system.
You don't have permission to comment on this page.