University of Washington

5G Connected Search and Rescue Drones

A masters capstone “launch” project

Launch Projects

As the capstone to the masters degree in Technology Innovation at the University of Washington, we are formed into multidisciplinary teams and tasked with creating an innovative new technology directly applicable to a real world problem and advancing social good. Having spent the previous year refining our skills with project based courses in design thinking, machine learning, fabrication, business, and more, the capstone is an end-to-end product launch.

Real-world Innovation

One of the ways that the program is different is in it’s understanding that solutions are more than just about the technology, but also knowing how to work in the broader environment and world. Launch teams work closely with industry sponsors with direct interests in the success of the project and students learn to work under constraints, with budgets, and with outside stakeholders. In return, students benefit from guidance and mentoring from expert teams from some of the biggest tech companies.

Network Connectivity and Drones

I chose to work on the T-Mobile network connected drone project because of its unique open-ended nature. Rather than having a specific product in mind, T-Mobile wanted to give us the freedom to explore ways that we could leverage their new 5G network and drones to showcase the transformative nature of the new technology.

An Opportunity

As an avid hiker and mountaineer, I’m frequently in areas where the things that keep you safe are your preparation, training, equipment, and group. But, accidents can happen even when you do everything right and there are many who are not as careful. Thankfully, dedicated teams of search and rescue volunteers and professions operate in all areas of the country to help locate and bring aid to those who need it where local police departments don’t have the resources or expertise to help.

Brainstorming new use-cases and technologies for drones got me thinking of how I would always hear about searches for missing hikers being called off due to dangerous conditions such as avalanche or nightfall. Helicopters were sometimes used, but even helicopter searches got called off due to high winds. But, why, in situations where having people on the ground or in the air put them at risk were we not using drones?

Design

Iteration

Rapid prototyping is all about iteration. Below is a brief summary of some of the work I did designing and fabricating an enclosure for our sensor system.

Version 1.0

Before we could start testing our drone, we needed a basic housing for our camera and thermal sensor and a way to attach them to the drone. We additionally needed some way to change the angle of the cameras to point either down or straight ahead.

I sketched some designs, took careful measurements of our sensors with digital calipers, modeled the parts in Fusion 360, designed the housing and mount, printed them in PLA, and had everything assembled and ready for my team to start testing in a week.

Version 2.0

With the basic version tested and working well, it was time to make a slightly more refined version for us to take out into the field for user testing.

Tolerances were increased to make it easier to work inside, the attachments were made more solid to prevent it opening with vibration, a bevel was put around sensors to protect them from getting bumped, and I even had some time to add our logo.

Version 3.0

After a successful first round of user testing, it was time to iterate based on the findings. One of the major findings in terms of hardware was that we needed to upgrade our sensor systems to be higher fidelity from the rough low-res prototype ones. We thoroughly explored the commercial options, but rather than use our entire budget on a single thermal/rgb camera, we decided to design our own using the same high quality parts used in commercial implementations.

Version 3.0 incorporates a professional quality FLIR thermal sensor and 4K RGB camera with interchangeable lenses. I designed the housing is designed to cleanly mount to a commercial 3-axis motorized gimbal and to keep our much more expensive sensor system safe with tight seals, strain relief, and impact resistant structure.

Refinement

Custom Power Delivery Board

As the project nears the close, our focus is on refining the design and creating a reliable and professional finished product. One of the aspects that I identified should be improved was out power management system, which in the process of adding new features and upgrading systems had turned into a bit of a “rats-nest” of cables and connectors. The power delivery board presented above would offer over-voltage, over current, thermal, and reverse polarity protections while allowing our end-user to use off-the-shelf cables to connect everything together, greatly reducing the chance that they would not need to bring the unit in for repairs.

And much more…

Even with the program coming to a close, the progress doesn’t stop. We are constantly refining the drone as we get ready for our big presentation. Expect much more to follow as I get caught up with the latest developments.

Cleaner Wiring

I have always had a high attention for detail and try to do every job with pride and to the best of my abilities, even the less glamorous ones. While the motto of “if it’s worth doing it’s worth doing right” is a good enough reason on it’s own, never is that more apparent than in electronics where clean and precise cables are the difference between a working product and a mess.

2-Way Voice Communication

We are exploring methods for enabling two way communication with rescuees on the ground to allow Search and Rescue teams to relay vital information and ask important questions.

Above is a parabolic amplification array I designed for the team that works with a common lavalier microphone. The design is parametric, so the height and diameter of the dish can be changed while automatically maintaining the correct microphone location and dish shape for optimal amplification.

BUsiness Case

Because technology doesn’t exist in a vacuum, an important part of the project is to examine the business case for the product, the IP considerations, and the ethical concerns that could arise from its use in the real world.

Because of my longer experience working outside of academics, I found myself well prepared to explore the real-world what-ifs and design plans to optimize the positives and minimize the risks.