From newshub.ccs.yorku.ca!ists!torn!utgpu!news-server.csri.toronto.edu!rpi!think.com!news!moravec Tue Jun 23 13:20:53 EDT 1992
Article 6278 of comp.ai.philosophy:
Path: newshub.ccs.yorku.ca!ists!torn!utgpu!news-server.csri.toronto.edu!rpi!think.com!news!moravec
>From: moravec@turing.think.com (Hans Moravec)
Newsgroups: comp.ai.philosophy
Subject: Re: Frogs vs. Robots (was Re: Transducers)
Date: 17 Jun 92 03:03:22
Organization: Thinking Machines Corporation, Cambridge MA, USA
Lines: 18
Message-ID: <MORAVEC.92Jun17030322@turing.think.com>
References: <BILL.92Jun10174436@ca3.nsma.arizona.edu> <60806@aurs01.UUCP>
	<1992Jun11.182144.12157@mp.cs.niu.edu> <60810@aurs01.UUCP>
	<YAMAUCHI.92Jun17004033@heron.cs.rochester.edu>
NNTP-Posting-Host: turing.think.com
In-reply-to: yamauchi@cs.rochester.edu's message of 17 Jun 92 04:40:33 GMT


yamauchi@cs.rochester.edu (Brian Yamauchi) writes:

> The same is not true about any mobile robot system currently in
> existence -- even in an artificial environment with all of the
> necessary resources (i.e. a building with recharger outlets).  At
> least, I haven't heard of any mobile robot which can operate without
> human intervention for a period of a month, much less multiple
> years...

Denning sentries have patrolled for a couple of months at a time
(recharging themselves by day) without intervention in a half
dozen test locations.  The sites had to be prepared with navigational
aids, and the routes preprogrammed.
Cybermotion has a couple of robots on similar tours in Glaxo warehouses
in North Carolina.
Overall behavior is more bug-like than frog-lke, though.



