Project Overview

The goal of this project was to create a submersible system that could retrieve three underwater “Treasure Objects” placed in random order on the bottom of the deep end of a swimming pool.  The device had to be able to discriminate between objects constructed of different materials and bring them to the surface in a specific sequence. The three “Treasure Objects” were outwardly identical with equal weight, configuration, and dimensions, but had a central tube structure comprised of different materials- ferrous metallic (steel), non‐ferrous metallic (aluminum), and non‐metallic (polycarbonate).

Design Process


When the team first sat down to begin brainstorming, all seven of us had very different ideas for how this underwater robot would operate. After numerous brainstorming sessions and a lot of back-and-forth discussion, we decided to build a bot that would traverse the pool at the surface of the water and descend only when it reached the end where the treasure lied by using a combination of bilge pumps and motors. Our frame would be constructed of PVC and lined with pool noodles to provide buoyancy. We decided to program an Arduino that would light an LED when it detected metal allowing us to determine which was the non-metallic treasure object. To discriminate between the ferrous and non-ferrous metal, we would attach a magnet to the bot.


Once we decided what our robot would look like and how it would operate, it was time to start building. The most challenging component of this project was our limited access to the pool. We were only able to test in the pool one time throughout the duration of the project making it very difficult to identify flaws in our design. Due to our inability to constantly test the robot, we relied heavily on our calculations and intuition to make design decisions.

We started by building our PVC frame. Our decision to use PVC was quickly validated as it is waterproof, cheap, and easy to work with. We decided upon a rectangular frame with a protruding bar in which the treasure would latch to and be brought to the surface and a protruding tube for metal detection. After designing the frame, the next step was to decide where the motors and pool noodles would be placed to maximize efficiency. Without open access to the pool this seemed like an extremely daunting task. To test buoyancy, we submerged the bot into sinks and bathtubs until we found the perfect noodle placement; however, it was impossible to test downward motion in these spaces. We also added a pool float to the top of the bot that was inflated while it traversed the pool but deflated as soon as it was time for the bot to descend.

Initially, we believed that bilge pumps would be the best way to move the robot. However, after preliminary testing in the pool, we discovered we were wrong and decided to only use motors and propellors. Based on our calculations and our observations from our initial testing, we decided to place two motors with propellers that we designed using SolidWorks and 3D printed at the back of the robot to facilitate forward motion across the pool. We also decided to place two motors in the center of the robot to facilitate downward motion to retrieve the treasure objects.

SolidWorks drawings of frame


Now that the frame design was finalized, it was time to focus on the electrical components. We decided to attach the controller to the motors with wires encompassed by a hose to insure that water did not interfere. We then attached a metal detector to the protruding PVC tube and connected it to an Arduino that would light an LED when metal was detected. We would encompass all of these components in a waterproof container secured towards the front of the frame.

Teammates putting wire through the hose

Final Arduino board

Final Test

When test day came around we felt confident that our robot would successfully complete the challenge within the allotted hour. We were quickly proven wrong. As soon as we put BURTEE, our robot, into the pool and turned on the motors, one of them did not turn on. We quickly removed the robot and tried to see what was wrong. At first it was unclear where the issue was coming from; however, we soon realized that water had somehow entered the container holding all of our electrical components. We tried to quickly remove all the water and seal all of the spots through which water could potentially enter. After patching it, we put it back into the water and one of the motors still would not turn on causing BURTEE to rotate instead of move forward. In fear of running out of time I presented the only solution I saw fit- driving it down the pool backwards. Although reluctant at first, eventually everyone agreed and we were able to get it to the deep end of the pool in the absence of one of our motors.

Troubleshooting on presentation day

While the lack of a motor was sufficient to get the robot across the pool, we needed power from all four motors to get the robot to the treasure at the bottom of the pool. Luckily, we had a spare motor with us and decided to solder it to the device poolside. While this was not ideal, it was our only option. Thankfully, all four motors now worked and we were able to descend. While we thought all of our issues were now behind us, we were once again wrong. As soon as BURTEE began to descend towards the first treasure object, the LED lit. As it was not close enough to any of the objects to detect metal, I knew that more water must have gotten into the container, causing the sensor to pick up a signal. At this point, with only 20 minutes left, there was no way to fix this. We now had to rely on our magnet and intuition to distinguish between the materials. In the end, we ended up retrieving 2/3 of the treasure objects in the correct order. While the loss of our metal detector made it impossible to correctly distinguish between materials, we were able to retrieve all three objects which, in the presence of all of our obstacle, I view as a massive success.

The team on presentation day


Working on this project was an incredible experience that taught me so many valuable skills. I had the pleasure of working on a team with people with diverse ideas and skill sets yet a common goal to complete this challenge. This allowed me to learn so much from each of my team members and also how to work cooperatively with anybody. As Program Secretary, I was in charge of planning all of our meetings and documenting our progress. This responsibility taught me the importance of organization when working on a challenging project with many different components and people involved. The many challenges we faced at the pool taught me how to be reactive and make quick decisions to ensure success. While the absence of these hurdles would have made test day much easier, I feel as if those challenges and the way my team came together to fix them single-handedly made me a better engineer.

With BURTEE after testing