The objective of the weed and feed competition, as of May 2019, is to develop an unmanned, autonomous machine that will achieve the following goals:
- Autonomously maneuver two or four 150-ft rows at a time and autonomously turn at each end. The agBOT shall make two 150-ft passes.
- Autonomously observe crop plants and fertilize plants as needed.
- Identify three common weeds: giant ragweed, cocklebur, and foxtail within and between rows.
- Arrange for weed to be destroyed either chemically or mechanically as the agBOT moves through the field.
- Provide real time observation on methods of movement, plant identification, fertilization, and/or treatment of plants back to the base station.
Teams are scored based on Mechanics, Software, Innovative Solution, and Execution.
Subsystems Overview
Computer Vision
The computer vision team subsystem is responsible for designing and implementing algorithms that will enable the machine to perform vision-related tasks, not including navigation, such as localization of crops and weeds in a field. Members of this team will be working primarily with image data and will gain experience in state-of-the-art image processing techniques like convolutional neural networks.
Mechatronics
The mechatronics team deals primarily with the mechanical and electrical components of the agBOT, including mechanical mounts and interfaces, wire harness design and electronic control systems. Members of this team will gain experience in part design and fabrication, using general-purpose computers like raspberry pi to control electro-mechanical systems, among other things.
Autonomous Navigation
Members of this team are responsible for developing the autonomous navigation capacity of the machine, such that the machine knows where it is in space, where it is going, how to get there and what to do if it encounters an obstacle.
Machine Intelligence
The Machine Intelligence subsystem generates useful insights from data (excluding image data) and builds computational models to encapsulate these insights and support in-field decision-making. Members of this group will gain experience in data analysis techniques, working with databases and machine learning.