Jonah T.

Reflection

I was in the second year BlueStamp program and I thoroughly enjoyed it. Not only did I take on a more advanced and challenging project, but I thought about how it would work in the real world. I got much better at coding, c++ specifically, and now have strong tools in mechanical engineering, electrical engineering, and computer science. I also had to pivot and change my project, and did so successfully. Although this was slightly worrisome at times it undoubtedly imparted even greater experience to me, furthering preparing for my later life as an engineer.

For my final milestone I have  completed the Robotic Ball Retrieval System and gotten it to take the ball and move it into a home base. Now if it were implemented on a tennis court it would have the capability to take a tennis ball and move it to a specific place, replacing the ball boy. The Pixy Cam now looks only for the color of a tennis ball, and upon tracking and clamping it starts looking for orange. It then sees the orange of the box and takes the ball to it, eventually releasing the ball. Because this was my second year program I took on not only a more advanced and challenging project, but also a financially viable one. This project could replace four out of the six ball boys needed at tennis courts. As it would not be fast enough to work at the net, but would work in the back of the courts. It would therefore save tennis tournaments lots of money because they would not be paying ball boys. They also would gain advertisement space. The saved money and ad revenue also makes the robot easier to sell for a higher price. To see exact figures click on the “Link to PowerPoint” below. I had two great challenges in building this project. First, I had to switch projects because the camera was not strong enough to build the Artificially Intelligent Ball Catching Trash Can I originally wanted to build. I had to pivot to the Robotic Ball Retrieval System in less than a week before I had to be done. I ended up doing everything well and getting the project to work, but I was pressed for time. Second, whenever the lighting would drastically change, either due to the time of day or some other factor I would have to reassign the signatures of the Pixy Cam. This caused me to have to go on the Pixy Cam app and take new pictures of the ball and the box whenever the lighting changed.

For my second milestone I have completed the larger part of my Robotic Ball Retrieval System. The robot is now able to go to a ball and secure it enough to transport the ball. The robot spins until it sees the ball in the middle 20 x-values (4.6875 ecegrees) on its 2D screen. It then goes straight towards the ball. However, if the ball is ever to the left of that middle range the right wheel speeds up, and if it is to the right of the range the left wheel speeds up. This makes the robot always go towards the ball and never miss the ball to the left or to the right for if the robot strays or the ball moves the wheels correct for that adjustment. I had trouble assembling the clamp and creating the code for this milestone. Although the clamp works well and was completed in much less time than I anticipated, I had limited time and resources to make it. I had to search within the BlueStamp lab for resources I could use to make the clamp. I figured out how to do it with a servo a few bolts and a piece of bendable metal. I had to do a lot of drilling and fine-tuning of the mechanism, but I am very pleased with the final result. The robot had a lot of trouble seeing and going accurately to the ball, and so I had to add the aforementioned left-right correction mechanism. I also had to adjust how fast the robot spun and the range in which it decided to go towards the ball and stop spinning so that it could see the ball before spinning past it. For my next and final milestone I will get the robot to take the colored ball to another colored place. This is important because it makes the robot a practical invention. If the robot can do that it could be implemented on tennis courts to replace a ball boy. It would see a tennis ball and take it to a specific place, all whilst out of the way of the players. This would leave the court clean after the match and save large courts money my replacing the ball boy, who gets paid by the court. After my next milestone the robot will really be a financially viable invention.