Introduction and development of intelligent systems for precision framing activities in greenhouse environment

Abstract

The main target of this research activity has been to develop an autonomous electrical vehicle able to perform precision farming tasks inside greenhouses. The vehicle can increase operations safety level for operators and quality of the chemicals distribution, allowing a lower environmental pollution and a better greenhouse product quality. Farmers thus obtain a return on their investment by using such a robot and technologies thanks to a waste reduction of phytosanitaries and fertilizers with a consequent cost lowering. Moreover, due to constant reduction of electronic parts cost, like computer, sensors and power control system and to the presence on the market of new low cost devices for precision farming, the development of a dedicated autonomous vehicle is not as expensive as in the past. In this thesis new low-cost solution will be introduced and detailed. Meanwhile, the robot has been equipped with an intelligent vision system in order to perform tomatoes detection in greenhouse and automatic guidance: the algorithm allows the tomato recognition and, as future development, once it will be integrated with the robot, it will be really useful to perform precision farming activities: fruit classification, harvesting, local chemicals treatment etc. During a first step, the activity has been focused on the development and test of the electrical vehicle able to autonomously navigate along the greenhouse's rows. After the vehicle had been fully tested, great effort has been carried out on the development of the vision algorithm and sensor integration through the development of a versatile and high-modular software framework. Encouraging results will be shown and interesting perspective can rise from this analysis.