Robotics research
Learning Social Navigation in Mobile Robots
Earlier COLORSLAB research on learning socially compliant mobile-robot navigation in dynamic pedestrian environments.
- Year
- 2025
- Role
- Undergraduate research assistant
- Built with
- Python · PyTorch · Unity · Graph Attention Networks · Conditional Neural Processes
As an undergraduate research assistant at Boğaziçi University COLORSLAB, I contributed to a team project on social navigation for mobile robots. The resulting paper, Mobil Robotlarda Sosyal Navigasyon Öğrenme / Learning Social Navigation in Mobile Robots, was published at IEEE SIU 2025.
The project explored navigation behavior that accounts for people moving through the environment, rather than treating them only as obstacles. The research combined graph attention for spatial interactions with conditional neural processes for trajectory prediction and evaluated the approach in simulated pedestrian scenarios.
My work was part of the collaborative model-development and evaluation effort. This is earlier robotics research rather than the main direction of my current work.