> Now this looks like a fairly AGI-friendly approach to controlling> animated > characters ... unfortunately it's closed-source and> proprietary though... Looks like a fun system but its purpose is to fake people into thinking that dumb game AIs are actually smarter by encapsulating some control algorithms behind a simple interface. I'm not sure how that is AGI-friendly exactly except that it similarly could make an AGI system appear to be smarter than it is by producing a more natural looking animation. That could help sell product or make extra-wizzy demos (perhaps misleadingly so), but I'm curious how you think it could be useful for AGI development itself. A complex many-channel detailed body model with rich kinaesthetic feedback, on the other hand, could be interesting for approaches wanting to leverage complex world interaction for the acquisition of low-level (sub)symbols. I'm not really a fan of virtual embodiment in general but this seems like one area where it could actually be a benefit because the technology for getting such detailed body-feedback from robots just isn't there as far as I can tell (skin and proprioception with thousands of channels, for example). Getting a nice level of detail for that would require massive supercomputers for the physical simulations, but maybe in ten years the price will come down. For people who don't see the perception/actuation tasks solved by animals as important to AGI, I guess the whole issue of sensors and body control is not even interesting.
------------------------------------------- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244&id_secret=101455710-f059c4 Powered by Listbox: http://www.listbox.com
