Hello all. I've been operating quadcopters (and a hexacopter!) for a couple of years now for my company, and lately I've encountered a new challenge that such vehicles are not capable of performing.
I have arrays of photovoltaics that have serial numbers on the underside that I need to locate. The process of doing so manually would involve physically detaching and lifting each module, a process that would likely take many hours, if not days, depending on the site.
I am conceptualizing a wheeled or tracked drone that could traverse under the modules and take pictures of the serial numbers, which would allow me to gather that information without any additional hassle. As someone who has never built a drone I expect there will be a number of challenges involved, but I'd like to ask advice from this community as to how feasible the idea is from the outset.
Key points that the vehicle would need to conform to;
- Low profile - The tightest spaces I would need to operate in would be 3", on average, with some variation depending on the roof type and auxiliary equipment placement.
- Traction - Capable of keeping traction on surfaces possibly in excess of 45 degrees, though the average would be considerably less than that. Metal and Shingled roofs would be the most common.
- Suspension - A lesser subpoint compared to Traction, but some roofs might have pipes, ridges, etc. that need to be traversed over. If the drone is miraculously only an inch in height, having the suspension to get over such obstacles would be ideal.
- Camera - At the most basic, I would need a camera that can get high enough resolution imagery to make out the serial numbers at close proximity. Having the ability to see the image output from a controller would be ideal, and being able to link it to a phone (for monitoring applications) would be even better.
- Battery - I am unfamiliar with the battery life of most ground-based drones, but if possible a battery life of up to half an hour or longer would be ideal.
Now, this seems like a tall order to me, being utterly new to this field with minimal research. I don't have any drawings or proof of concepts, but I'm hoping that I can get some advice on the various parts I'd need to look into to make something like this mildly feasible, if it even is.
My ultimate desire would be to see something tracked, most likely, something vaguely serpentine. Most of the time I expect I would be able to travel in a straight line from one end of the array to the next, then do a 180 turn to the next row.
Anyways, hopefully this all makes sense and is not completely ridiculous. Thanks in advance!