Visual Servo Control of a Mobile Robot in Agriculture Environments^#

Abstract

Abstract In this paper, a visual servo control for a mobile robot evolving in intensive farming environments consisting of rows of fruit trees is presented. The calculation of the current position of the robot is based on the perspective lines that define the navigation's path. These lines are obtained through the processing of visual images captured by the robot. The image is segmented using statistical techniques based on the analysis of the color characteristics and the spatial distribution of the descriptive points of the path. The perspective lines are determined through the Hough transform, which describe mathematically the navigation environment. Based on these lines, the relative position of the vehicle in the grove alleyway is calculated and a kinematics control algorithm is designed for the autonomous navigation along the corridor. The stability of the resulting control system is proved based on the Lyapunov theory. The experimental results in an environment of growing olive trees shows a good performance of the system design. Keywords: AgricultureImage processingMobile robotNonlinear controlPerspective lines Notes #Communicated by M. Ceccarelli.