Subject:
|
Re: Ideas for a Science Fair Project
|
Newsgroups:
|
lugnet.robotics
|
Date:
|
Sat, 9 Jan 1999 18:54:48 GMT
|
Original-From:
|
Mark Williamson <markw98@ibm.SPAMLESSnet>
|
Viewed:
|
1206 times
|
| |
|
|
How about this:
The Psychology of Robots
In this experiment, the participants build robots in whatver configuration
they choose. The main program of each robot must have an Identity value.
Identities are Red, Green, Blue, Brown, Yellow. A second value, Preference,
tells the robot that it prefers Red, Green, Blue, Brown, or Yellow, maybe in
a certain order, or pattern. Some robots could be nuetral, i.e. no
preferences. These values could also be set randomly at program
initialization. Even the identity could be random, therefore, you would
have completely random, but patternized behavior. As is the natural
behavior of all species, random, but patterned according to the animal's
nature.
Now, the control algorithm tells your robot to seek out robots that have
higher preference values, and possibly avoid robots of lower preference
values. Each robot could be configured to emit a modulated value with it's
IR system, indicating it's Identity value.
Why is this at all interesting? Because it demonstrates real world dynamics
of society and behavior. Of course whether or not the judges agree, is
another question.
Regards,
Mark Williamson
--
Did you check the web site first?: http://www.crynwr.com/lego-robotics
|
|
Message has 1 Reply:
 | | Re: Ideas for a Science Fair Project
|
| Maybe this could be set up to work like those name tag things on the MIT website, using IR to indicate a series of True false variables (up to 8) and then issue a compatability rating of 1-8. other algorithms include using 4 true/false pairs (...) (26 years ago, 10-Jan-99, to lugnet.robotics)
|
2 Messages in This Thread:
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
