Shop OBEX P1 Docs P2 Docs Learn Events
Robot locomotion/obstical advoidance logic. input wanted — Parallax Forums

Robot locomotion/obstical advoidance logic. input wanted

DgswanerDgswaner Posts: 795
edited 2008-01-07 22:39 in Robotics
I'm looking for input on robot control logic... not code just the process in how it functions. I'm using the propeller but it really doesn't matter what platform your using, I have working code with the propeller and one ping. I'm not getting 100% usable results so I want to look into different processes and see if I can find the best solution for what I'm doing. If you want to contribute, I'm asking for any configuration and for any sensor or any quantity of sensors. Feel free to add to or take away from what I show. Ultimately I'm not getting usable results because of the nature of the PING Ultrasonics. I'm running into my soft fabric covered couch and when it's heading toward a wall at an angle it doesn't see anything until it's right next to the wall. which is normal in this application.


pan ping servo left to right then to center.
---if the center is clearest path go straight.
---if objects on left side are closer than right
move ping servo to far left.
turn right until distance is > 30
center ping servo
--- else if objects on right side are closer than left
move ping servo to far right.
turn right until distance is > 30
center ping servo

once moving forward if distance is less than 40 cut speed to 1/2
once distance is less than 20 stop and repeat pan process.

or from the ping around bs2 code.

pan servo to 5 different locations. then back to center
turn to the direction of the clearest location.


Thanks for the input

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
"A complex design is the sign of an inferior designer." - Jamie Hyneman, Myth Buster

DGSwaner

Post Edited (Dgswaner) : 1/7/2008 9:44:39 PM GMT

Comments

  • LilDiLilDi Posts: 229
    edited 2008-01-07 21:30
    Your experiencing typical limitations of ultrasonic range sensors. Soft fabrics and sharp angles to the object are a problem with these type of sensors. I have had better luck with IR range sensors when it comes to soft fabric or sharp angles.

    Ping Document:
    www.parallax.com/Portals/0/Downloads/docs/prod/acc/28015-PING-v1.4.pdf
  • DgswanerDgswaner Posts: 795
    edited 2008-01-07 21:42
    I questioned putting that part in this thread I was worried about it getting side tracked, I do realize that this is normal behavior for a PING and in this thread I am merely looking for other thought process for obstical avoidance. so I can decide on which sensors to add based on what people have had the best luck with. I normally have IRDP sensors on all of my bots but I'm waiting on some out of stock parts from mouser! thanks for the input tho.

    I've had several small bots that work with only IRPD and they won't hit anything. but with my current bot it's 17" in diameter and over 2' tall. it has enough torque to move a small couch and moves quite fast compared to a boe bot. I need about 6-8" to stop. with IRPD's range 2-16" +/- that really cuts down the reaction time.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    "A complex design is the sign of an inferior designer." - Jamie Hyneman, Myth Buster

    DGSwaner

    Post Edited (Dgswaner) : 1/7/2008 10:00:06 PM GMT
  • LilDiLilDi Posts: 229
    edited 2008-01-07 22:15
    Have you looked at the Sharp line of IR rangers?

    http://www.acroname.com/robotics/info/articles/sharp/sharp.html#e8
  • DgswanerDgswaner Posts: 795
    edited 2008-01-07 22:39
    wow those are really cool I could average the distance from the IRsensor and the PING (when it's with in IR range) and get the best of ultrasonic and IR world.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    "A complex design is the sign of an inferior designer." - Jamie Hyneman, Myth Buster

    DGSwaner
Sign In or Register to comment.