We will be building a Fireproof DIY drone from a mix of 3d printed parts and machined parts. The Drone will use a semi-custom made pathfinder/path-making software developed by us to map out a burning building. After building a map of the house (will show where the trapped individuals are) the drone will determine the best path to the POI and fly to them. Once the Drone has arrived, it will deploy oxygen masks with filters so that the POI doesn't die of inhaling too much smoke. Then one of two things will happen, the drone will either stay with the POI until the firefighters can bring them to safety or the drone will escort them out of the house (situations will vary).
Will add pictures as we develop this project.
Determine how to wrap the materials on the drone and the camera
Investigating Fireproof materials: When researching fireproof materials we first looked for other fireproof drones and what they used for protection. We considered various fireproof insulators and ceramic tiles, however these materials were far too heavy for our drone. After looking at an article by Popular Science about a fireproof drone made by the Korea Advanced Institute of Science and Technology, we found that they used materials called “aramid fibers”. With this discovery in mind we did some research and discovered that aramid fibers are various forms of kevlar, nylon, and polyethylene. We also discovered that aramid fibers are the exact materials used in fireproof suits. Further research revealed another promising material, aerogel. Aerogel is a fabric that is used by NASA for heat shielding in their spaceships. After doing more research we decided that a certain aerogel (known as Pyrogel) is another material worth considering.
We then researched various statistics, such as the average temperature of a fire in a burning building, the amount of oxygen necessary to survive a certain amount of time, diagnostics of oxygen masks and equipment used by firefighters, whether people die first from heat or inhalation of toxic fumes, and insulation standards and notations. This information, along with data found in a pdf of firefighter statistics, we then used to calculate various r-values necessary for our insulators at different temperatures.
Determining how to wrap the materials on the drone and the camera: After finishing materials research, we moved to how we would use aramid fibers and Pyrogel in our design. As we were discussing how to integrate the materials if the drone, we thought of the idea to use multiple layers instead of just one layer. After adopting this method into our design we had three final options. Option one was having the first layer be aramid fibers, followed by a air cooling layer that would use thermoelectric cooling to keep everything at a stable temperature. This layer would also have thermoelectric paint to extend battery life of the drone. Then, the last layer would then be pyrogel. Some of the positives of this design were that it was energy efficient and had a %100 chance of working. However because of the extremely advanced levels of math and physics required to make the design work we didn’t choose option one. Option two would have the first layer be aramid fibers, followed by pyrogel and thermoelectric peltier coolers. The pros of this design are that it would easier to build (not triple integrals involved) and cheaper than all the other options. The only con with this design is that we might need to make major changes along the way for it work, which will make it more complicated and expensive. The third and last option would have the aramid fibers for the first layer, followed by a flexible insulator. The third layer would then use pyrogel as a thermal insulator. Finally, the fourth layer would use cryogel and a fan to keep the internal parts of the drone cool. Basically, the third option is us making a custom fireproof suit for the drone. The only con to this design is that it is the most expensive of the designs. We will only go with this design if we get the necessary funding for our project. If we don’t get enough funding, we will go with option two.
Drone: Our Drone will be based off of DJI’s Zenmuse XT 2 drone. Some of the differences however, will be the fact that the drone will enter the house and won’t be as expensive and heavy.
Camera: We considered the following different cameras for our Drone:
DJI Stereo vision guidance
-10C to 40C
12W with all sensors activated
ZED Stereo Camera
0C to 45C
Scanse meet sweep
-10C to 60C
DJI Zenmuse XT
-10C to 40C
All of these cameras look great, except that they need to be able to handle being in a fire. We will need to either make the cameras more rugged or get a camera that can handle the extreme conditions that it will be in.
Vision: We had two different options for how the camera will generate a 3d map for us. The first option is Stereoscopic vision. Stereoscopic vision is basically human binocular vision. The second method is Lidar. How Lidar works is that is uses light from a pulsed laser that can measure ranges. It does this will thousands of points that form a 3d map.
We ended up going with Lidar since it doesn’t need visual analysis.
What we will be doing next is looking more at Lidar and rugged Cameras.