A What?

There is a problem that many people have the luxury of not facing or knowing about. Patient noncompliance is when one refuse their medications and/or do not take them as prescribed. This problem can lead to devastating consequences to ones health. With a growing elderly population, some of the issues that they face are opening the bottle, taking the prescribed dose of medication, and remembering to take the pills at the correct time of day. If only there was a way to automate this process. A device that could dispense pills at timed intervals would be incredibly useful in a wide variety of applications, I call it an automated pill dispenser.

Project Origins

It was the summer of 2016 and a college friend of mine had scouted me to be on her senior design team. Later we picked up another member making us a group of three. In the following weeks we met to brainstorm ideas. We wanted a project that could be completed within our time constraints, budget, and manpower. Our first couple of ideas failed miserably. Then as all college students do, I called my mother for ideas. At the time my mother was taking care of my grandmother who was living with my family, back home. My grandma was diagnosed with Alzheimer’s disease and had a complicated set of medications that needed to be administered at regular intervals. My mother’s first suggestion was a robot that could monitor and provide continuous care. Now that would be amazing and impressive, but not within the scope of senior design. My grandma could not remember if and when she had taken her medication, let alone which medication she needed to take. If I could create a device able to dispense pills for her and remind her when to take them the problem would be solved! So we set out on an ambitious task of creating a device capable of dispensing medication in hopes of helping my family out.

Project Definition

Given infinite time we could solve all of the worlds problems with apes randomly putting parts together. As educated apes, we set out to make a device that would complete these criteria:

1. Dispense a single pill when requested
2. Work across a wide range of pill shapes and sizes, it had to be a modular solution and not one size fits all
3. Have a low, if not zero, counting error

Easy enough right? Well, approximately eight months later and we were able accomplished our goal.

The (Boring) Details

It was a long process that started with 3D printed prototypes and ended with a bucket of 3D printed prototypes and extra parts. If you would like to read our Final Report that was created for Senor Design you can download it at the here. It contains a LOT of information about the design process and the revisions the project went through.

The Automated Pill Dispenser

At long last! Behold it in all of its glory.

To the left are all of the control buttons. They are big, easy to use arcade style buttons. We designed the pill dispenser to work with two different types of pills. The top was a fiber capsule style pill. On the bottom we used a small sugar pill, similar to many birth control medications. These two pills allowed us to capture a wide range of different pill dimensions without designing for hundreds of different pill types. The disk insert allows us to swap which types of pills the machine can dispense. Designing and printing new disks is a straightforward process.

Here is a video of it working.

Some of the major challenges we faced were getting the light sensor to reliably detect pills and getting only one pill dispensed at a time. The first problem would be easily solved with an industrial light sensor, buying a part from the internet made to detect object passing through. Our time constraint did not allow for this option, but future revisions will. Getting only one pill to be dispensed at a time was a challenge. We tried using a plastic brush, but that created a lot of jams. What fixed the most problems was aligning the pills before they got to the spinning disk. This keep them from stacking. With some geometry changes to the disks and the basin the problem was mostly averted.

Conclusion

This project was a great learning experience. Over the course of a few months my team was able to design and redesign a prototype until a working model was achieved. As I continue working, I have goals to improving reliability, functionality, and user friendliness in hopes of it becoming a functional piece of medical equipment. I would love for this project to help as many people as possible.