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 / 844
843  |  845
Subject: 
Re: CyberMaster memory limit?
Newsgroups: 
lugnet.robotics
Date: 
Sat, 28 Nov 1998 22:33:31 GMT
Original-From: 
Laurentino Martins <lau@mail.telepac.pt[stopspam]>
Viewed: 
2271 times
  
At 20:16 28-11-1998 Saturday , you wrote:
In message <4.1.19981128170141.00976610@mail.telepac.pt>, Laurentino
Martins <lau@mail.telepac.pt> writes
I'm going to use the Tacho counters in the threads to get the approximate
position and direction where the rover is facing (the difference between • the 2
Tachos should give us a direction), which I think is feasible with some
finetunning.
Then, at each impact the computer will map a point in the screen and with • some
luck map a room. :-)
Is anybody working on a project like this?

This use of software to determine direction is what I was alluding to in
an earlier post when the others were trying to get their mobile to
travel in a straight line using magnets and differentials, etc. On the
Cybermaster CDrom there is a test routine which calibrates the turning
360 degrees , so there is some experimenting still to do with the
Tachos.

There's nothing unknown about the Tachos. If you need help using them, gimme a
call. :-)

I have been experimenting with explorations using the grabber tool
fitted on the tank unit model. I was surprised how powerful it was and
that the batteries lasted quite a long while.

Yeah, I've also started with that bot. It's very well designed!
And I've never exchanged any batteries! :-)

What would you suggest as the best design for exploration then?

The best vehicle would be one with a lot of traction to the ground to avoid any
slippage. That was my first idea, but after playing a bit with the wheel base
models they teach in the CD, I've thought that would be much easier to me if I
stick with the base tank configuration.
I know it's worst for precision to use threads, but for my first project using
the CyberMaster I though that would be OK.
Then, I needed a way to detect impacts ahead of the real impact. If you let you
vehicle touch the object then there's the chance it to rotate with the violence
of the impact.
I designed two flexible bug-like antennas. They are in the shape of an open V
with independent rotation (although they never cross). Behind each one I have
put one touch sensor so I can sense impacts on the left or right without
affecting the CyberMaster unit.
Then I created a program that senses impacts and voids them by backing up a bit,
rotate a random value to the opposite side of the impact (between 5º and 45º),
and proceed forward. I also used another task to watch sensors during maneuvers.
If when maneuvering to avoid an obstacle you touch another object in the
opposite side, them you must be in a tight alley (my room have a few of them).
It then reverses (between 5 to 20 centimeters) rotating a random value and
proceeds forward.
With this simple design I can explore my whole room and the simple trick to get
out of difficult situations is the random rotations. It works best than I
expected!
In the future, I'm going to put the 3rd. touch sensor in a bumper in the back of
the unit since it might impact on some object when reversing and I would never
know.
I haven't done that already because I'm a bit out of ideas how to make flexible
a bumper for the whole rear (the CyberMaster is my only LEGO). Any ideas?

Laurentino Martins



Message has 1 Reply:
  Cybermaster Sensors
 
In message <4.1.19981128213514....lepac.pt>, Laurentino Martins <lau@mail.telepac.pt> writes (...) In the book "Mobile Robots" by A.M.Flynn it suggests three bump sensors which are activated by circular plastic skirt made out of acrylic plastic. It (...) (26 years ago, 30-Nov-98, to lugnet.robotics)

Message is in Reply To:
  Re: CyberMaster memory limit?
 
Your program generates a .RCX file of 440 bytes. I've created one with 452 that works OK, but had to stop around that number because after more 2 or 3 lines of code the last task simply disappeared (my program had 1 main plus 3 tasks). I guess the (...) (26 years ago, 28-Nov-98, to lugnet.robotics)

20 Messages in This Thread:









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

This Message and its Replies on One Page:
Nested:  All | Brief | Compact | Dots
Linear:  All | Brief | Compact
    

Custom Search

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