• This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn more.

CPE Senior Design Project 2016

travelerdawg

Multi-rotor Joker
#1
Hello everyone,

I am a senior Computer Engineer at the University of Alabama in Huntsville. I have been approved by my professor to present a project to the class and university for consideration.

This project involves developing a camera base with the capabilities of tracking a race drone through a portion of a race course. The idea is to build multiple bases to provide producers the means to have another perspective of the drones as they traverse the race course. The reason I am calling this a camera base is that it should be able to accommodate multiple cameras up to a certain size and weight (GoPro, Xiaomi, RunCam, Cannon Vixia, Sony Handycam).

My end goal is to increase the enjoyment for spectators of drone racing. We have the FPV view and the broad angle of the camera on the sidelines. However, the camera on the sidelines has a hard time tracking the drones as it is human operated and the shots I would like to get would put a camera operator in too much danger to be practical.

What am I looking for from the community? I have attempted to contact flitetest directly and have not been successful. I have attempted to contact MultiGP and have not been successful. I am hoping that by posting on the forums that I can start a dialogue with somebody at flitetest and with the flitetest community in order to receive advice and ideas on what spectators of the sport would like to see. Also, sponsors would be greatly appreciated and handled through the university.

Thank you for your time and I hope that I have interested some of you in supporting this senior design project. I have already received support from my fellow classmates in the way of requests to join my team and I hope you will do the same. If there is interest from the community, I will keep the community updated with regular status updates.

Sincerely,
Travelerdawg
 
Last edited:
#2
Sounds like a great idea. I would love to see the final result of your project and I'd be willing to help (I have graduated from CalPoly with my Computer Engineer degree). Unfortunately, I am not sure how I can help you contact MultiGP or FliteTest.

From my point of view, I think you need to use something other than a GoPro or RunCam as I think you will need to use a camera that can zoom in closer that those camera are able to (the others might work but I do not know them well enough). I believe you need a camera and system that can keep the racer in frame and filling at least a 1/4 of picture line (or 1/16 of the picture by area) though the closer you can zoom in the better. My reason is, a good tight zoom level makes it easier for the viewer to track the racer and help give them a sense of the speed and skill being displayed by the racer. I think if you are zoomed too far out, the racer is too small to easily see and understand what the racer is doing.

At that level of zoom, I think your biggest trick will be developing a system with the accuracy, speed, and low enough latency of detecting the position and speed of the racer good enough to keep the racer in frame. But, if you are able to pull it off, it will be a great tool for live broadcasts of drone racing.

Good luck.
 

travelerdawg

Multi-rotor Joker
#3
I really appreciate your response and input. I believe that you are right about using cameras that have zoom capability. I believe we can use a raspberry pi 3 or similar computer for image processing and signal tracking. The idea is to have the platform keep track of the multicopter and perhaps interface with the camera zoom once tracking has been accomplished. I personally own a GoPro and a Versacopter that we can use for testing purposes.

If anyone else has more ideas, please reply! :applause:
 

travelerdawg

Multi-rotor Joker
#7
Nobody seems really interested in this thread, but I will still give a quick update. We have managed to get the x,y,z coordinates from the UWB modules using source code from Decawave. This has allowed us to determine the angle for pan and tilt that the servos need to move to in relation to the base station and the tag that is attached to the drone. We have 3d printed enclosures that the hardware is going into and will be demoing the product next week. Everything is starting to come together and it looks like this senior design project will be a success! I will discuss with my mentors whether or not I can post pictures and a video on here if there is any further interest in the project. Thanks!
 

travelerdawg

Multi-rotor Joker
#10
Sorry, yes, I will get back to this thread. Maybe I can cobble something together this weekend. I am not sure what I can disclose but I think I can post the presentation slides if y'all would like to see that. I also have a short video somewhere of the final product in action. It isn't NEARLY as impressive as the original idea but it does track and has the potential to advance with another senior design team.