A vision system based crop rows for agricultural mobile robot

Abstract

In a machine vision-based autonomous navigation system for agricultural field mobile robot, obtaining guidance information from crop row structure is the key in achieving accurate control of the robot. This paper presents a new method for robust recognition of plant rows based on the Hough transform. It used a camera to find a path from structured agricultural fields to automatically navigate a mobile robot following crop rows. Firstly, the camera calibration was applied to obtain the relationship between the image coordinates and the world coordinates. Secondly, pattern recognition and image processing were used to obtain quasi navigation baseline. And lastly, the real navigation line was extracted from quasi navigation baseline via Hough transform. Experimental results indicate that this method has a simple robust algorithm, low-level requirements for software and hardware, and ultimately can meet the requirement for agricultural robot's works in the field.