To LUGNET HomepageTo LUGNET News HomepageTo LUGNET Guide Homepage
 Help on Searching
 
Post new message to lugnet.roboticsOpen lugnet.robotics in your NNTP NewsreaderTo LUGNET News Traffic PageSign In (Members)
 Robotics / 12208
12207  |  12209
Subject: 
Thoughts & Issues w/ Vision Command
Newsgroups: 
lugnet.robotics, lugnet.robotics.rcx, lugnet.robotics.rcx.nqc
Date: 
Wed, 23 Aug 2000 07:27:24 GMT
Highlighted: 
(details)
Viewed: 
1581 times
  
I purchased Vision Command tonight from Target.  I must say I'm rather
impressed with what I've seen so far.  I bought VC mainly for its
applications with RIS so that is what this post will focus on.

The VC software allows you to film time lapse and stop motion videos but I
have not tried either of these features.

The VC software allows you to divide up what the camera sees into certain
pre-defined sections (about 20).  You can then have your program respond
differently based on what happens in each section.  For example:  If the
camera detects movement on the left side of the room it plays a sound.  If
the camera detects a specific color on the right side it takes a picture,
records a video, sends a command to the RCX, etc.  The camera can respond to
motion, color change, and light change.  It also has a microphone for
recording sound during videos.

The commands you can send to the RCX seem limited to the outputs and sounds.

I have successfully built an RC car controller like device.  There is a
vertical lever for moving the camera up and down and a horizontal lever for
moving the camera left and right. (I used the powered camera based outlined
in the VC documentation).

What I HAVEN'T figured out is how to get the camera to take a picture or
record a video based on the RCX sensors (for example take a picture when the
touch sensor is pressed).  Sending IR messages to the tower won't do much
good as it doesn't appear as if the VC software can respond to the tower.

There doesn't appear to be anyway to get the VC software to stop or start
programs already downloaded on the RCX

Here's my hypothetical:

Lets say I want to create a robot that searches for a red ball.  I want to
write the navigation routine in NQC (because it is more powerful - i.e
getting out of corners, etc.) until the camera finds the ball.  Then I want
the VC software to take over (it is now more powerful because it can respond
to sight and probably has an unobstructed path to the ball) navigating
directly to the ball and picking it up .  How do I tell the RCX that the VC
is now in control?  The NQC program will never know the camera found the
ball and continue to run its navigation routine.  Now there is the potential
to have both VC and NQC sending commands to the outputs.

It would be nice if the VC software could use something similar to the
SendMesage(0-255) command.  Then I could use NQC for navigation while
waiting for commands from VC.

Hopefully these are limitations because I have overlooked features already
in the VC software or I am just not creative enough at this hour (it's
almost 3:30am here).

I'd like to know what others think about this.

--

http://members.home.net/rjschave



Message has 1 Reply:
  Re: Thoughts & Issues w/ Vision Command
 
(...) What about the picture quality? Could the Camera used as a cheap alternative instead of a digital camera? Only for web pictures of Lego creations, of course ;-) Regards, Hakan (24 years ago, 23-Aug-00, to lugnet.robotics, lugnet.robotics.rcx, lugnet.robotics.rcx.nqc)

3 Messages in This Thread:

Entire Thread on One Page:
Nested:  All | Brief | Compact | Dots
Linear:  All | Brief | Compact
    

Custom Search

©2005 LUGNET. All rights reserved. - hosted by steinbruch.info GbR