Line following

Line Following
This is not the 2017 course

Time limit

5 minutes

Control method


Aim of the challenge

Your robot will autonomously follow a black line approximately 15mm wide on a white background. The 2014 course layout is as shown at the top of the screen – click here or on the image for a larger version. This year’s course will be different but we recommend using the design for practice. The black line does not go to the very edge of the perimeter but instead allows a robot A4 in size to navigate safely around.

Note for 2015 competitors: the course will not have vast amounts of colour as per last year – we’ve learned from our mistakes!

Ranking and points

Robots will be allocated 5 minutes to complete the course as many times as they can up to a maximum of 5 circuits. Each circuit of the track will be timed and the shortest time taken will count for ranking purposes.

Points will be awarded to the top finishers like so:

  • 1st: 40 points
  • 2nd: 32 points
  • 3rd : 25 points
  • 4th : 18 points
  • 5th : 15 points
  • 6th : 12 points
  • 7th : 10 points
  • 8th : 8 points
  • 9th : 6 points
  • 10th : 4 points
  • 11th : 2 points
  • 12th : 1 point

Additional points

  • 10 points will be awarded for each completed circuit of the course.


  • The first rescue is free of points deduction.
  • Subsequent rescues incur a deduction of 3 points per rescue.
  • Up to and including 5 total rescues are permitted, after which you must abandon the challenge.
  • Points will be totalled at the end of the challenge and it is not possible to go below zero points.

25 thoughts on “Line following

    1. Hi. We won’t be releasing details of the shape or length of the course – it’s designed to be a surprise until the event. Suffice to say, long enough to be challenging, short enough not to be ridiculous!

  1. Hi. just a thought, any idea what the course is being layout on, will it be PVC tape on paper/hardboard or a printed like last year. and if possible to have a sample at the next jam or Pi & Pints to test sensors

    1. Good question Brian. It’ll hopefully be printed like last year. However, we weren’t sure if the reflective properties of the material were conducive. Last year, we had to go over the black line with black electrical tape for people to pick it up, but I wasn’t sure if that was a big problem or just one team that had that problem. Can you remember?

  2. Have just seen this thread. It was me that had major problems and requested that the line was taped over. I remember other teams having similar issues but can’t remember who sorry. Only thought would be to check last year’s results and see if the completion rate was significantly higher once the course had been taped, that would indicate whether taping made a difference. It did improve it for me, but still wasn’t 100% successful.

        1. That’s the weird thing… the course we had done last time _was_ a Cannybots course! Letting people try one out might be handy, though!

  3. Ok, I’ve just spent the last 24 hours going batty with my line followers thinking I’d broken them, as they were working fine when we last tested, when we realised that actually it’s that the printed line has faded too much – even though I’ve gone over it in black marker. Line sensors are picking up black phone case and black lego bricks fine, but not the printed line. We’re off to get some electrical tape, hoping that that will work. But if you’re going to have a printed line for the competition, we’ll need to get different line sensors. (We’re currently using TCRT5000.)

  4. Mike. We’ve always tested using a line printed on A0 paper (inkjet) and the Ryantek sensor array with no problems. The sensor also had no problems with the cannybots course last year (we won!) and works on electrical tape too. Not sure if we are doing anything different to others, but Harry’s code is on GitHub via

  5. Shame – RyanTek currently out of stock of that sensor board. Interesting that it’s a 3 sensor board. I used similar 3 sensor setup (not RyanTek) which worked beautifully on electrical tape and printed (laser and inkjet) at home whilst testing but then failed miserably on the CannyBots course last time round.

    1. That particular sensor will be back in stock in about a month – he’s literally just sold out and Chinese New Year prevents quick replacement of stock.

  6. Hi Mike, – I’m sure it would be hear and I’m pretty sure you tweeted a month or so ago to say that the course is secret – am I remembering rightly? Also, what is the course made from? I’ve tried paper which catches and gets chewed up too quickly and we’ve tried white foam board from hobby craft (currently on offer) which is brilliant but very very slippy… My guess is the surface will be a major factor how fast it can go.


    1. This is the 2017 competition rules, so I hope you’re not a competitor!
      As long as the robot re-acquires the line, there’s no problem.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.