Shop OBEX P1 Docs P2 Docs Learn Events
Which would you prefer — Parallax Forums

Which would you prefer

hdn0380hdn0380 Posts: 6
edited 2006-12-12 01:50 in Robotics
Boe Bot tread or the crawler kit. Why?
I will be deploying to Iraq soon and I plan on taking my boe bot so I can work on it in between betting blown up. Just curious to know which would be better over there. I don't plan on taking it outside a controlled environment, but who knows.

Comments

  • Steve JoblinSteve Joblin Posts: 784
    edited 2006-11-23 15:48
    I think the tread kit will serve you better... you can create "terain" challenges for added fun whereas the crawler may have more limitations on the surface it is operating on.

    On a much more important note is making sure you know that you have the full support from EVERYONE that particiaptes in these forums for protecting our country! It does not matter if we think the US forces "should" be in Iraq or not, that is not the point... the fact is that we support and give our greatest thanks to those that serve to protect us and our way of life... Our thoughts and prayers will be with you on your mission and look forward to your safe return!
  • hdn0380hdn0380 Posts: 6
    edited 2006-12-11 03:29
    Thanks for the suggestion. I decide on the crawler kit...I'm tired of seeing tread, plus the legs make it look cooler, I think. I added a ping sensor to help it get around but now I think I may have a slight problem. Because the legs cause the boe-bot to move up and down as it's walking, I think it throws the ping sensor off and it constantly stops and scans the area every few steps.
    I'm using the example code "roaming with ping" that came off this site. Does anybody have any ideas on how to compensate for the up and down motion? Maybe I'll get some time to figure it out in Iraq.
    DSCN1791.JPG
    2048 x 1536 - 729K
  • Steve JoblinSteve Joblin Posts: 784
    edited 2006-12-11 18:31
    What happens if you put a "stop" on moving the legs while you take a reading? take a step, stop, take PING))) reading, take a step...

    does that solve the problem? If so, then you have determined that it is an issue with the bot moving while the PING))) is trying to take a reading..
  • Mike GreenMike Green Posts: 23,101
    edited 2006-12-11 18:44
    On the other hand, the BOE-BOT doesn't move very fast compared to the time it takes for the PING to take a reading. You're unlikely to get much of an error in a 50ms maximum reading timeframe. Echos from up close take much less time. I'd suggest ignoring the fact that the BOE-BOT is moving (at least with respect to the PING measurements).
  • Steve JoblinSteve Joblin Posts: 784
    edited 2006-12-11 20:36
    Are you taking multiple readings with the PING and averaging the values?
  • ZootZoot Posts: 2,227
    edited 2006-12-12 01:50
    I've dealt with the issue of moving, bouncing 'bots and possibly inconsistent Sonar readings as follows:

    - run the 'bot around taking readings
    - *if* a reading appears to be below your "escape/obstacle" threshold, THEN
    - stop and take a reading while you are standing still.
    - if you indeed have an obstacle while at rest, then avoid it, otherwise keep going.

    Also, I often use two different thresholds -- one a "tuned" threshold to check for possible obstacles while "underway" and a separate threshold for checking obstacles while "still"

    I agree with Mike that the time is not the issue, however bouncing around on different surfaces can make Sonar readings slide around quite a bit (in my experience, anyway). I've tried averaging, but the problem there is you don't necessarily get a quick enough reaction time -- it could take a number of cycles to bring your average low enough to trigger an escape routine. In a slow-moving Boe-bot that may not be such an issue, though.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    When the going gets weird, the weird turn pro. -- HST
Sign In or Register to comment.