| Topic: |
Science > Physics |
| User: |
"" |
| Date: |
05 Oct 2005 01:12:18 AM |
| Object: |
Robot Equiped to Navigate Building's Interior |
Is it physically possible a robot could be created to navigate a
building, including hallways, doorways, and staircases? ..and also
navigate near the presence of people?
.
|
|
| User: "CWatters" |
|
| Title: Re: Robot Equiped to Navigate Building's Interior |
05 Oct 2005 01:57:04 AM |
|
|
<mwmiller314@gmail.com> wrote in message
news:1128492738.813514.167510@z14g2000cwz.googlegroups.com...
Is it physically possible a robot could be created to navigate a
building, including hallways, doorways, and staircases? ..and also
navigate near the presence of people?
Yes but it might be fooled if it doesn't know the combination to the lock on
the lab.
.
|
|
|
|
| User: "Sam Wormley" |
|
| Title: Re: Robot Equiped to Navigate Building's Interior |
05 Oct 2005 04:12:00 PM |
|
|
wrote:
Is it physically possible a robot could be created to navigate a
building, including hallways, doorways, and staircases? ..and also
navigate near the presence of people?
Don't forget to consider the navigational capabilities of
cruise missiles.
.
|
|
|
|
| User: "Sam Wormley" |
|
| Title: Re: Robot Equiped to Navigate Building's Interior |
05 Oct 2005 01:18:33 AM |
|
|
wrote:
Is it physically possible a robot could be created to navigate a
building, including hallways, doorways, and staircases? ..and also
navigate near the presence of people?
Robots obey the same physics laws as people.
.
|
|
|
| User: "" |
|
| Title: Re: Robot Equiped to Navigate Building's Interior |
05 Oct 2005 02:44:31 AM |
|
|
well clearly a person can navigate a building, but that doesn't mean a
human crafted machine will do the same
.
|
|
|
|
|
| User: "Ian Parker" |
|
| Title: Re: Robot Equiped to Navigate Building's Interior |
05 Oct 2005 05:28:47 AM |
|
|
Robots obey the same physics laws as people.
Yes as Sam says they obey the same physical laws, they do not think or
sense in the same way. For example it is not intuitively obvious
whether robots should have visual systems like we have or whether they
should have a laser or sonar system which would give a solid map.
To navigate a robot will need a brain which has a solid map. Given a
solid map navigation is fairly straightforward, the robot simply finds
routes of connecting free space. The "Greedy" algorithm will do this
quite readily. The computer finds the shortest route and optinizes
using a variational principle (Should this be sci.math?).
Another interesting question is the question of balance in bipedal
robots. Dynamic balance is a fairly straightforward eigenvalue problem,
and it seems amazing to me that so little progress has been made up to
now.
Buildings have to have expensive modifications to accvomodate the
disabled. Would it not be far better to have a "wheelchair" which could
climb stairs?
In the civil field I feel that radar/sonar are the ways to provide a
map. In the military field where stealth is often required, the method
of triangulation that we use is needed. I have discussed this in
correspondence.
See the Title= the nonsense of manned spaceflight" Word to be found =
"Iraq"
.
|
|
|
| User: "" |
|
| Title: Re: Robot Equiped to Navigate Building's Interior |
05 Oct 2005 05:40:26 AM |
|
|
A human's advantage is that s/he has years and years of training and
experience of a nerual net which no robot can come close to having at
this point in time. And it can be created by pleasurable, unskilled
labor, which explains why there are 6E9 human robots prancing the Earth
right now (cheap, abundant fossil fuel helped with this also, a
situation to end soon). So you could create a robot to do what you
say, though it would take a lot of memory trained for a long time.
Perhaps you might want to do that only in environments which humans
can't handle.
We work cattle Thursday.
- Don
.
|
|
|
|
| User: "" |
|
| Title: Re: Robot Equiped to Navigate Building's Interior |
05 Oct 2005 03:50:48 PM |
|
|
Can you direct me to any examples of robots acheiving this navigation
using a 'solid map'?
.
|
|
|
|
| User: "" |
|
| Title: Re: Robot Equiped to Navigate Building's Interior |
05 Oct 2005 03:51:00 PM |
|
|
Can you direct me to any examples of robots acheiving this navigation
using a 'solid map'?
.
|
|
|
|
| User: "" |
|
| Title: Re: Robot Equiped to Navigate Building's Interior |
05 Oct 2005 03:51:05 PM |
|
|
Can you direct me to any examples of robots acheiving this navigation
using a 'solid map'?
.
|
|
|
| User: "Ian Parker" |
|
| Title: Re: Robot Equiped to Navigate Building's Interior |
06 Oct 2005 04:40:45 AM |
|
|
Don't forget to consider the navigational capabilities of cruise missiles.
A cruise missile uses GPS, other methods of navigation have been tried
but everything is gravitating to GPS. The insurgents in Iraq have
nothing which will seriously interfere with satellite transmissions, so
there is no point in looking elsewhere. Earlier versions did indeed
have a solid map and flew close to the ground following contours.
GPS by the way will place ordinance in a 2m circle.
In discussing technology we should really be looking for ways forward.
If I were tasked with building such a robot, if I were asked for (say)
a trolly you could load up in the kitchen and which could then move
though into various dining locations, a solid map would be the only way
forward. Solid maps and VR "worlds" are the obvious way of introducing
robots into a military situation.
Could I also point out that the Pentagon, at intervals, has a race for
automated cars to Las Vegas. So far no car has met the required
standard - Not to say it is impossible. I am simply stating the route
development will have to take.
.
|
|
|
|
|
|
|
Related Articles |
|
|
