Our Project | Smart Farming

Precision. Intelligence. Sustainability.

The Future of Precision Agriculture – Machine Learning for Autonomous Decision Mechanisms
(EU-Projekt Acronym: SMART FARMING)
2023-1-DE01-KA220-HED-000166720

SMART FARMING is an EU-funded research and innovation project that aims to make agricultural production processes more resource-efficient and environmentally sustainable with the help of artificial intelligence (AI). The focus is on developing and testing a prototype software system for autonomous decision support based on agronomic drone data and location-specific environmental parameters.

Using winter wheat as a reference crop, the project is investigating how AI-supported systems can contribute to optimising irrigation, fertilisation and plant protection. The focus is on reducing the use of inputs and improving yield stability.

Project Number: 2023-1-DE01-KA220-HED-000166720

Project Duration: 01/11/2023 – 31/10/2026

Project Objectives:

Our goal is to develop an innovative autonomous decision-making mechanism based on artificial intelligence and machine learning to achieve significant advancements in precision agriculture. The project aims to increase crop yields and improve the utilization of underused or unused fertile land, thereby contributing to long-term food security.

Project Results:

The project outcomes include the development of data collection and processing software for decision-making, the creation of an autonomous decision-making tool, a user-friendly mobile application, a handbook titled The Future of Precision Agriculture, the organization of 15 seminars, multiplier events and a conference, the publication of scientific articles, the establishment of a project website, and the production of video content.

Project Partners:

prEUnec GmbH, Germany (Coordinator)
Vondiconsulting Unternehmensberatung Vondrak KG, Austria (Partner)
Universitatea Aurel Vlaicu, Romania (Partner)
University of Cukurova, Türkiye (Partner)
RobyCode UG, Germany (Partner)

Output	Description
Software I	Drone data analysis software (e.g., plant health, soil quality)
Software II	Autonomous decision-making tool for precise agricultural actions (e.g., irrigation, fertilisation)
Mobile App	User-friendly interface for field data display & drone communication
Project Website	Central hub for all project updates, materials & outcomes
Book	The Future of Precision Agriculture
Seminar Materials	Educational content for students & professionals
15 Seminars	Delivered online and on-site across partner countries
4 Scientific Articles	Topics: sensor tech, AI, yield modelling, sustainability
2 Project Videos	Two 10-minute videos on smart farming & decision-making
3 Multiplier Events + 2 Final Conference	For stakeholder engagement, knowledge sharing, public outreach

Experimental Field at Cukurova University (Adana, Türkiye)

In the framework of the SMART Farming project, a large-scale experimental field was established in cooperation with the Eastern Mediterranean Agricultural Research Center in Adana. The selected site covers approximately one hectare and meets strict criteria for precision agriculture and drone-based data collection.

Key Features of the Experimental Field:

Crop Rotation & Soil Preparation:
To prepare ideal soil conditions and remove excess nitrogen (N) from previous cultivation, silage corn was planted and harvested in October 2024. The field was then fertilised and sown with wheat as the target crop.
Field Design:
A structured layout of 54 experimental plots (27 early- and 27 late-crop) was implemented, each measuring 6×10 m with 3 m spacing and buffer zones. Different levels of nitrogen (0, 10, 20 g/m²) and irrigation (0%, 50%, 100%) were applied to evaluate crop response.
Drone Integration:

Multispectral drones were used to monitor vegetation development. Over 2700 drone images in 10 spectral bands were collected through 5 flight operations, allowing real-time analysis of crop growth, water stress, and nutrient status.
Challenges & Learnings:
Despite challenges such as weather instability, drought, weed infestation, and technical constraints in irrigation infrastructure, the experimental setup remained intact. Manual weeding and fine-tuned irrigation systems were applied to ensure data quality.