Description of Activities

The Artificial Intelligence Department was established in collaboration with Bosch Group as Hungary's first Industrial Department in AI. We aim to integrate research, innovation, and education, fostering a modern research base and knowledge center. The department offers internships, scholarships, and doctoral programs in self-driving vehicles, industrial automation, machine vision, neural networks mapping human brain function, evolutionary technology, fuzzy systems to computational intelligence. Bosch and ELTE are collaborating on Level 4 and 5 self-driving systems, enabling machine intelligence to manage vehicles safely without human supervision and intervention. We research Cognitive Robotics and Human-Robot Interactions, focusing on emotion recognition, gesture recognition, robot locomotion, and cobots. Further members of the research group are working on improving field robot and autonomous UAV functions by integrating sensor fusion across diverse environments. They're collaborating with ELTE's Faculty of Science to model birds’ collective thermal soaring behaviours to develop realistic simulations for training AI, improve UAV efficiency, and reduce operational costs. By leveraging social cues—both passive and active—we aim to enable more effective collective flight. Our partnership with the Department of Meteorology involves integrating sensors on drones and UAVs to enhance their performance in dynamic environmental conditions.

Research Interests

Artificial Intelligence; Computational Intelligence; Machine Learning; Robotics

Research Methodology

Evolutionary computation; Fuzzy systems; Deep Learning; Spiking neural networks; Memetic algorithms; Swarm intelligence; Human-robot interaction; Human-robot collaboration; Transfer learning, Sim2real; Field robots; Sensor fusion

Research Staff

Associate Professors: János Botzheim (Head of Department, botzheim@inf.elte.hu); Dr. László Gulyás (lgulyas@inf.elte.hu), Zoltán Istenes (istenes@inf.elte.hu); Senior Researcher: Ellák Somfai; Assistant Professor: Balázs Nagy; 18 Ph.D. students

Projects

Bosch projects: Automated Unpacking Machine; Manufacturing Process Optimization; Anomaly Detection; Driver Drowsiness Detection; AI National Laboratory [National Grant, 2019-2025] – Human behaviour analysis; APOLLO 2028 [EU Grant, 4 years] – Medicine; AI EDIH [EU&National Grant, 4 years] – Knowledge Transfer to SMEs; EMOTIONAL AI For EU [EU Grant, 3 years] – Education (Digital Skills); TINLAB – National Laboratory for Social Innovation (together with Faculty of Special Needs Education) – Digital Psychodiagnostics; "Development of a complex meteorological support system for unmanned aerial vehicles." TÁMOP-4.2.1.B-11/2/KMR-2011-0001 UAV_MET; Intelligent Field Robotic Systems (IFRoS) Erasmus Mundus Joint Master Degree Co-funded by the European Union

Some important publications in the field

• Gulyás, L. et al. (2024) ‘On the Power of Graph Neural Networks and Feature Augmentation Strategies to Classify Social Networks‘, In Proc. 16^th Asian Conf. on Intelligent Information and Database Systems. DOI: 10.1007/978-981-97-4985-0_23
• Mei, J., Gulyás, L., and Botzheim, J. (2023) ‘Comparing Lamarckian and Baldwinian Approaches in Memetic Optimization‘, In Proc. 15^th Int’l Conf. on Computational Collective Intelligence, pp. 521–533. DOI: 10.1007/978-3-031-41774-0_41
• Domonkos, M., Tresó, Á. and Botzheim, J. (2023) ‘Online Surveying System for Experimentally Testing the Human Perception of Visual Gestures‘, 9^th Int’l Conf. on Automation, Robotics and Applications, pp. 329–333. DOI: 10.1109/ICARA56516.2023.10125860
• Lonklang, A. and Botzheim, J. (2023) ‘A Rapidly-Exploring Random Tree Algorithm with Reduced Random Map Size‘, 9^th Int’l Conf. on Automation, Robotics and Applications, New York University Abu Dhabi, pp. 356–361. DOI: 10.1109/ICARA56516.2023.10125934
• Gyöngyössy, N. M., Eros, G. and Botzheim, J. (2022) ‘Exploring the Effects of Caputo Fractional Derivative in Spiking Neural Network Training‘, ELECTRONICS, 11(14), pp. 2114:1–2114:20. DOI: 10.3390/electronics11142114
• Horváth D, Martín J B, Erdos F G, Istenes Z, Moutarde F, (2024) “HiER: Highlight Experience Replay for Boosting Off-Policy Reinforcement Learning Agents”, IEEE ACCESS 12 pp. 100102-100119., 18 p. DOI: 10.1109/ACCESS.2024.3427012
• Horváth D, Erdős G, Istenes Z, Horváth T, Földi S, (2023) ”Object Detection Using Sim2Real Domain Randomization for Robotic Applications”, IEEE TRANSACTIONS ON ROBOTICS 39: 2 pp. 1225-1243., 19 p. DOI: 10.1109/TRO.2022.3207619