Will Using Raw Sensor Values Improve your EV3 Line Follower's Performance?

Поділитися
Вставка
  • Опубліковано 23 гру 2024

КОМЕНТАРІ • 18

  • @chunkytakodachi8105
    @chunkytakodachi8105 4 роки тому +2

    Are you still at MIT

  • @DUDtope
    @DUDtope 6 років тому +1

    Could you help me on how to calculate the speed of the ev3 and then represent this in my data log. I would like to see the speed time graph for the ev3

    • @Builderdude35
      @Builderdude35  6 років тому

      I have a tutorial coming in September that will teach you how to measure the motor's rotation speed, which you can use to find the robot's driving speed. The datalog is another issue and I am not sure that I can really help you with that.

  • @jayostuchanda3340
    @jayostuchanda3340 7 років тому +1

    can you do line following on those race car designs such as plethora2 and terminal velocity? I was asking since those have rack and pinion steering mechanism. And what about an automated calibration where the robot finds black and white/color next to black and calibrates?

    • @jayostuchanda3340
      @jayostuchanda3340 7 років тому +1

      i was thinking of giving the robot command move around mat until senses white or black, calibrate to minimum if black and calibrate to maximum if white and then find the other color. could also be like this place robot near black line, carry out line squaring and then calibrate to black go forward a couple of degrees till it finds edge f black and color after it and calibrate. and also do you have a third party block where you can control 3 or 4 ev3 motors? thanks.

    • @Builderdude35
      @Builderdude35  7 років тому +1

      It looks like you have some great ideas here. Yes, you can certainly do line following with a robot that has car-style steering; you would use my custom steering blocks, but use the reflected light intensity as the steering target.
      There's no block for controlling 3+ motors at once, but you can certainly make your own MyBlock.
      Good luck! You've got some great ideas and a clear direction. I recommend that you jump in and get started with experimenting.

  • @kalocsaizoltan6425
    @kalocsaizoltan6425 7 років тому +2

    Your videos helped me so much! You should make a video for WRO open cathegory too.

    • @Builderdude35
      @Builderdude35  7 років тому +1

      Thanks for the suggestion! I haven't made a WRO video in a while, but we will see what the future holds.

  • @toanphamquang625
    @toanphamquang625 5 років тому +1

    will putting your color sensors closer to the wheel makes your PID line follower less effective

    • @Builderdude35
      @Builderdude35  5 років тому

      That depends on the line you are following and how wide it is. You want the space between two sensors to be enough such that they are on either side of the line, both halfway on the edge. For FLL, which has a libe width of one inch or so, this means one stud of separation. Any more or less and the line follower might not work at all.

  • @dulacgaming
    @dulacgaming 5 років тому

    i have a question: have you tried merging two DIFFERENT techniques for line following into one program?

  • @w0ttheh3ll
    @w0ttheh3ll 6 років тому

    I was hoping to actually get some real sensor values, like some kind of reflection measurements of the red, green and blue channels in color mode. that would've been super useful. could've done complex proportional line following while detecting color markers at the same time without the huge delays from switching sensor modes.

    • @w0ttheh3ll
      @w0ttheh3ll 6 років тому

      it turns out that a custom sensor block that does exactly that (getting the RGB values) is available for download from www.mindcuber.com/mindcub3r/mindcub3r.html#ColorSensorRGBBlock

  • @spaceybread
    @spaceybread 7 років тому +1

    Great Video
    -SB

  • @arhamjain7306
    @arhamjain7306 6 років тому

    you have not told how does it works