안희진 연구실

Plan safe motion avoiding surrounding agents' predicted future motions under uncertainty. Design safe motion planning systems based on predictions, formulate chance-constrained optimization problems, and improve learning-based prediction models.

Develop techniques for uncertainty quantification in control systems and anomaly detection in DNN models for semiconductor manufacturing. Characterize uncertainties in mathematical models and detect incompleteness in DNN models.

Develop safe and efficient methods of autonomous vehicle management at unsignalized intersections. Coordinate multiple autonomous vehicles using model-based optimization and data-driven generative models.

Utilize sensor data from high-positioned infrastructure to achieve precise detection and classification across broader areas. Address challenges of occlusion and sensor measurement distance limitations in urban intersections.

Implement autonomous driving technology beyond Level 4 using a scaled-down 1:15 ratio testbed with realistic sensors and V2X communication. Support multi-agent technology development with complex traffic flow scenarios.