An Autonomous Seeder for Maize Crop

Abstract

This paper presents a solution for sowing maize using an autonomous robot. The robotic seeder minimizes human involvement in the sowing process and hence reduces errors, time consumption and increases efficiency. An adjustable feature has been introduced which can be used to adjust the width of the seeder by considering the width of seedbeds in the fields. The robot is equipped with a 2D laser scanner and a seed sowing mechanism. The sensor provides environmental information in the form of distance data which is desired for controlling the movement of the seeder. The information is further used to extract seedbeds using Random Sample Consensus (RANSAC) with line fitting. For autonomous navigation, differential drive and PID control have been implemented. The sowing mechanism allows accurate seed sowing in proper alignment. It drops seeds continuously on equidistant positions during movement of the seeder. A simulated field using V-Rep has been created for testing the functionality of the seeder in simulation. The results from simulation present trajectories of the autonomous seeder and seeds sowing pattern. A hardware prototype is also presented showing the proposed seeder.