The aim of the research project SAAPIN (Autonomous System for Precision and Integrated Agriculture) is to develop an autonomous system able to navigate through maize field and perform sustainable weeding operations.
The autonomous system is comprised of an autonomous mobile platform base on a modified small tractor, which is going to navigate autonomously and the weeding farming implement, joined to the tractor platform, which is responsible to perform the weeding operations.
A small tractor is selected and adapted to perform autonomous navigation, by integrating sensors to get necessary information to navigate (environment: vision, GPS, odometry, LIDAR, IMU) and by performing mechanical adaptations (direction and transmission). Control system uses a global planner algorithm to cover the whole extension and a reactive trajectory planner based on computer vision localization when navigating inside the crop rows.
A simulation environment is developed to evaluate efficiency of different navigation strategies to manage the movement of the whole system through maize fields, as well as to analyse the performance of the system and the sensors which fits better to the goal.
The weeding operation is mechanical, to be done by vertical axis rotary brushes. An intelligent system based on vision perception system identifies the position of crop and weeds, and commands the movements of brushes. Adequate inner illumination is also provided (Cloudy Day Illumination).