AUTONOMOUS SPEED SPRAYER GUIDANCE USING MACHINE VISION AND FUZZY LOGIC

Abstract

A Fuzzy Logic Controller (FLC) was developed for autonomous operation of a speed sprayer in an orchard. The operation of the FLC was graphically simulated under the real condition of the orchard. Machine vision was used to determine vehicle heading and four ultrasonic sensors were used to detect obstacles during the operation. Operation time duration of the hydraulic cylinders was inferred as output of the FLC. The results of simulation showed that the speed sprayer could be operated autonomously by the FLC that used sensor inputs from machine vision and ultrasonic sensors. After the graphic simulation, autonomous speed sprayer operation in a real orchard was conducted using the developed FLC. Orchard image analysis and signals of ultrasonic sensors were processed in real time. The speed sprayer was modified to be steered by two hydraulic cylinders. The results of field test showed that the speed sprayer could be operated by the FLC. The ultrasonic sensors didn’t contribute to the improvement of guidance performance; however, they assisted the speed sprayer to avoid trees or obstacles in emergent situations.