Transform cones detecting model into ROS network

[gmin7]

Description:

• Convert the current darknet model/framework to a ROS compatible

Acceptance Criteria:

• Installed the correct packages for YOLOv3
• Successfully start up the SLAM navigation repo (IGVC course) with the darknet nodes simultaneously (they don't have to be communicating, just simultaneously compiling
• The weights we trained with YOLOv3 are incorporated into the nodes
• Create a flow diagram describing how the inference node and the camera subscriber are implemented (this will involve describing how you break down the inference model into smaller chunks and who subscribes/publishes to what topics)

Notes:

• Documenting what is an is not possible in the comments section

[AnnieJDZ]

https://github.com/leggedrobotics/darknet_ros
YOLO ROS: Real-Time Object Detection for ROS, can use our own detection objects