12,550 miles strong
July 7th, 2016
Happy Wednesday, Lily crew! We’ve seen a few requests on our social media channels for more specific technical details around the bugs we’re tackling at HQ. In the interest of these requests, let’s dive into a hardware issue we solved a few months back.
First, some background. As your Lily Camera prepares for flight, it will look for a satellite 12,550 miles away from itself to catch a GPS signal. GPS is sensitive. It doesn’t like noise. Especially not electromagnetic noise. But when you combine GPS and a camera in a small, confined hardware space, you get just that. Waves and waves of it.
We began to see problems in the Lily Camera’s ability to understand its exact location in the world. This affected flight performance. We realized that we needed to find a way to reduce electromagnetic interference (EMI) to ensure that the Lily Camera’s internal antennae gets a clean GPS signal. If we could do this successfully, we knew that Lily would be able to safely and accurately navigate the skies. The solution we chose to use was shielding.
Essentially, sheets of copper are strategically placed within the device to absorb any noise coming from the camera or the codec (the processor that handles video and audio). The shields are also designed to block any electromagnetic interference emitted by any other electronics located nearby. With these safeguards in place, your Lily Camera receives a stable GPS signal.
And now, a more recent example of flight control behavior we’re in the midst of perfecting. A couple weeks back, you heard about the OptiTrack System we use to isolate and test beta Lily Camera features one at a time. Rigs like the one picture above are used to test pitch and yaw control. (Picture the motion of a seesaw. That’s pitch. Picture a dog’s tail wagging back and forth. That’s yaw.)
Lily Test Engineer Carlos Alvear created this rig with a 3D printer and a laser cutter. It’s built strong enough so that the Lily Camera won’t fly away mid-test., and high enough off the ground so that the unit avoids ground effect. Why test pitch and yaw in-house like this? Precision and control.
We hope that these two examples give you a better idea of the sorts of issues we are solving. If you would like to hear more of these types of in-depth technical reports, let us know and we will make sure to include them more often in our updates.