SkAIGreen: AI-based active control of drones that seek indicator species for grassland biodiversity

About the project

To support agricultural practices that reduce biodiversity loss, the EU introduced the Common Agriculture Policy. One support tool comprises result-based agripayment schemes (RBAPS), in which payments are made to farmers if they show the presence of a minimal number of indicator plant species on the targeted parcel, via a manual monitoring procedure. This procedure is inefficient, expensive, difficult to implement by farmers and to verify by the control body. The natural next step to address these shortcomings is to automate species identification.

The SkAIGreen project will therefore develop AI and deep-learning based active control of one drone or a drone team to automatically seek RBAPS indicator plant species for the biodiversity of Romanian grasslands. AI and machine learning will play a key role both in species identification and in active drone control to find as many species as quickly as possible; and we will make fundamental advances in active robot control. The novel technology will provide faster, cheaper, and more verifiable monitoring results than the existing human-based procedure; and will be quantitatively validated in a case study in the Angofa region of Romania.

The project is funded under the Experimental-Demonstrative call of the Romanian Authority for Scientific Research, via UEFISCDI (project number PN-IV-P7-7.1-PED-2024-0481, contract number 124/2025). It takes place at the Automation Department of the Technical University of Cluj-Napoca, Romania, for a duration of two years and a total budget of about 150 000 EUR.

Team

At the coordinator, Technical University of Cluj-Napoca:

Lucian Busoniu, principal investigator (professor, PhD Eng), manages the project and focuses on fundamental research aspects related to learning and active control.
Catalina Dan (associate professor, PhD Eng) provides botanical expertise and focuses on species identification.
Alexandru Codrean (associate professor, PhD Eng) focuses on drone control.
Georgiana Daneliuc (technician, MSc Eng) focuses on species identification.
Davian Martinovici (technician, MSc Eng) focuses on neural networks and deep learning.

At the partner, Adept Transilvania:

Razvan Popa (PhD), partner PI, inputs use-case parameters and interfaces with end-users.
Cristian Gherghiceanu manages the case study and exploitation.
Anamaria Roman (PhD) focuses on plant biology and ecology.
Cristian Malos (PhD) focuses on GIS in the experiments.

To get in touch with the team members, email first the PI at lucian [at] busoniu [dot] net.

See also the publications resulting from the project, as well as the impact the project is having. The administrative info page contains more detailed logistical information.