Document describing the structure, implementation and utilization of neural networks for tracking game objects on the field in real time. We used the Caffe library from Berkeley Vision, the OpenCV library, and the DIGITS software from NVIDIA to create the neural network.
Zebravision 3.0 is a culmination of libraries and several applications that are designed to use a Logitech C920 camera mounted to a robot to detect recycling bins in the 2015 FRC game: Recycle Rush. It is designed to run in real time using the OpenCV library onboard a NVIDIA Jetson TK1 DevKit mounted to the robot.
Want to track and follow an object but don't have strong programmers? We provide a Raspberry Pi image that lets you train a target,and then it sends steering commands to the cRIO telling you where in the image the target is.
This paper explores vision processing technique for the 2013 Ultimate Ascent game, and discusses the math behind constructing an autonomous multivariable shooter controller.
Everyone is wrong! - Chris is me [more] 
