hello friends,
any person with a slightest interest in robotics must have heard of line follwer robot..... its the first autonomous robot which most of the robot-hobbyist build..... here i am showing my line follower journey from scrap to robot.....
Planning:
planning..... the first step to success...... well, like every other enthusiast, i wanted my line follower to be the best..... some feature which i wanted were:
1. average speed of 1m/s.
2. able to take sharp turns.
3. able to take acute angle turns.
now, i wont lie..... the final product was not able to do all these three things together. so i compromised with speed.
Shopping List:
here are the main components which i bought with their prices (prices are quite high where i live) .
2 x DC Motors 12V 300rpm 750mA full load current ............................. Rs 450/-
1 x L293D Motor Driver ............................................................................... Rs 90/-
6 x IR-Led-Photodiode Pair.......................................................................... Rs 90/-
2 x wheels 7cm dia..........................................................................................Rs 70/-
1 x castor wheel small ....................................................................................Rs 35/-
Other discretes
Stage1 : Mechanical assembly
I am not a mechanical guy, so i always choose material which i can work on easily..... my options are cardboard, wood, thermocole, duct tape, fiber string etc....... i chose simple differential drive mechanism. 2 wheels attached to separate DC motors at the back. castor wheel at the front.
Stage2 : Electronics
Circuit designing is my favorite task.... thats why i always try to make changes in circuit and see the effects. most of the line followers which i see on internet use comparators like LM324 in their sensor circuits. its a good circuit but i thought of giving the ICs lying in my room a try..... there was a ULN2803...... i gave the sensor output to it directly. to tune the sensitivity i was controlling the current through IR-led using variable resistor...... this approach worked well. other circuit was nothing very innovative so i am not describing it.
Stage3 : Programming
i tried more than 10 different approaches but the best results were given by PID undisputedly.... the tuning for various coefficients was very time consuming.... even after all my efforts the best i could get was 20cm/sec.... but on my track that was not bad at all.
FINAL SHOWDOWN
here is the video of my line follower:
Here are a few snaps of various other variations which i tried.
