Sumo Robot

As part of our course work for technology in 5th year, our class was entered into the Robotics Ireland Sumo robot competition. In this competition, each pair had to create a sumo robot, with its aim to push the other robot of the table. The robots started with their backs facing each other, it was crucial the robot could turn 180 as fast as possible. The robot used a proximity sensor, to notice the opposing robots and drive towards them. The difference between who won and who didn't was the ultimate design.

Aims

With the First Lego league under our belt, we knew what it took to create a functioning robot. We knew carefully designing and planning was key to a successful robot. We started by using CAD, to create various prototypes for the sumo robot. These prototypes varied in width, height, and overall shape. The key features we needed were.

• Small wheels to increase torque

• A sharp ramp at the front of the sumo, to lift the opponent

• Sturdy Frame

• Materiel with high friction to wrap around the wheels.

With these criteria in mind, and the multiple prototypes, we chose the final design.

Description

The above CAD is the robot we decided on. It took approximately one to two months to make. At the front of the robot we sharpened the acrylic to act as a ramp to lift the opponent to give us more leverage, the proximity sensor positioned above is the eyes of the robot, once it stops something its drives. On top was the battery pack along with the push button and toggle switch. The toggle switch was used to turn on the robot, once the push switch was pushed following this the robot would function immediately. The push button reacts faster once the robot is on. Evens a matter of tenths of seconds could be the difference a win and a loss. In side the frame contained the Genie E18 board, this was the brain of the robot. On either side the motors powered the wheels. To create as much friction as possible between the wheels and the table, we glued rubber strips around the wheels. Under the frame contained two LDRs, if these LDRs noticed the white of the perimeter of the table the robot knew to turn in an effort to try and avoid being pushed off. All parts of the robot were cut out using the CNC machine, then glued together. Once the frame was assembled, the electronics were inserted.

Problems

While the manufacturing aspect of the robot wasn't too challenging. It was the programming aspect that proved to be the most difficult and time consuming. Some days the robot could work perfectly other days it would struggle. There was constant need to tweak the code, to ensure the robot would run as reliably as possible. With all projects there was problems, the ramp wasn't sharp enough so we often were the ones who were being lifted (later fixed with eraser blades being glued to the ramp), as the LDRs are so close to the ground LEDs were required to improve LDR visibility. As mentioned above problems with the board, such as low battery level altering performance and overheating off the mother-chip.

Result

The competition was in may, and in the days leading up the robot was still unreliable. Thankfully on the day the robot, performed quite well winning all the heats. However it was knocked out in the quarter finals. Overall we placed 5th.

Conclusion

When compared to the first Lego league robot, we had the experience to know to plan the robot before we actually started manufacturing it. In the end this saved us a lot of time in manufacture modifications. The lesson learned, to always allow for more time than you think for the software, because while manufacture might go to plan 80% of the time, the software never does.