In my first year at the college, I learnt a lot of new things related to autonomous robotics. So when the time of our technical fest Kshitij came, I decided to take part in Robotix’s event Minefield. I formed a team with two really good friends of mine and fellow robotics enthusiasts, Shubh Agrawal & Aashay Bhise.
The Problem statement for minefield was to build a gesture controlled semi autonomous robot that could detect metal mines by itself and then could grip objects and place them at designated positions. The event, although relatively simpler was a great challenge unto itself. The basic robot was pretty simple to make, moving via the gesture controls of hands’ tilts. The metal detection circuit was available online. The gripper could easily be bought, or could be fabricated quite easily. But robotics enthusiasts as we were, we tried to take things differently.
Basically our bot was quite different than what was given in the DIY tutorial for the event. We attached a servo motor on our robot and used two sensors for gesture control. A digital compass on one hand decided the angle at which the servo had to point. The objective was to make the robot follow the direction of the hand. It turned out to be good and we were indeed able to achieve the same. Other than that, an accelerometer on the other hand (pun was intended 😛 ) was used to control the speed of moving forward and backward, and making turns. Other teams just copied the DIY and made a simple accelerometer control.
Then for the Gripper, we got a wooden one fabricated from a carpenter and used a plastic gear motor to grip. The control of the gripper was a LED LDR pair that rotated the motor when the hands were clutched, akin to gripping something by a real hand. That was something awesome in our bot.
And the main thing, the metal detection. Everyone else used a metal detector circuit that worked on induction. It was available for about Rs. 150 online and I won’t lie, we too ordered one and by good luck or bad, it didn’t work. We tried building it again but no luck. So then I used our plan B, we ordered a pack of really strong neodymium magnets on the internet, as we had carefully read the document and the mines had to be magnetic. So if we had a magnetic material underneath the magnets, they would lower, complete the circuit and trigger a beep. That was an awesome feature that everyone, including seniors and co-participants appreciated.
And then, the sad part. The bot was working peacefully upto 3 days before the event, with just the metal detector not in place. We thought it to be a minor fix and in the process, one of the magnetometers stopped giving values. The values started drifting after sometime and the servo went crazy. As a result gripping was nigh impossible. In short words, our event was screwed :-/
We took a consolation run and couldn’t do much. But indeed it was a great learning experience and something I’d cherish for a long time from now. My first competitive autonomous bot, complete with a lot of innovations. We have broken the chassis and harvested the parts for new robots though. Other than that, I was a junior organizer for the event too, so I enjoyed a LOT taking runs for the same.