Well, the combination of unreliable motors and limited sensors was driving me crazy with the maze. In retrospect, I’d add an accelerometer to the robot next time, so that I could be sure about position when turning. I’m using a simple left-hand rule for the maze, working on the principle that if the maze is simply connected (technical term that, meaning that all the walls are connected together or to the outer boundary), if you keep your left hand on a wall at all times, you can keep going and eventually you’ll get out the exit. Based on this, I can make do with just sensors on the left and front of the robot. But the “sticky-outy” bit in the middle of the maze was causing me grief, because it wasn’t so easy to determine when I’d cleared it, and the robot had a tendency to end up in the open(-ish) space in front of it going round and round in circles.
I toyed with the idea of adding an I2C-based accelerometer during the week, but I’m not going to have any time to work on the robot next week anyway, so that was never a realistic option. So in the end, I’ve ended up hard-coding the left hand turn, tweaking speed parameters on the left and right motors until it turned with the appropriate circumference. And it appears to work (thank heavens!). Multiple tests, a few slight touches on the wall, but can definitely solve the maze 😀
And the added excitement for today is that I’ve collected our team t-shirts from the wonderful Sue at Heatherhill Farm Embroidery. (As seen in the image at the top of the post.) Now I just need to get Glitterator glittering to match the shirts!