Road Geometry Predictions for Autonomous Driving
Francesco Borrelli, Professor
Mechanical Engineering
Applications for Spring 2025 are closed for this project.
Accurate road geometry prediction is a cornerstone for building reliable world models of the driving environment, which is crucial for autonomous vehicle navigation. This project aims to develop a model to accurately predict key road geometry features, such as road curvature and bank angles, using data from multiple sensors, including cameras, radars, and IMUs.
Role: In the first phase of the project, the students will design and implement a road geometry forecast algorithm in Python using the provided camera and radar measurement data under the guidance of graduate students in the lab.
In the next phase, the students will integrate the developed algorithm in a test vehicle (IONIQ-5) equipped with a sensor suite (camera, radar, IMU) for autonomous driving in the lab.
Qualifications: Python knowledge is mandatory. Experience with computer vision and machine learning is preferred.
Day-to-day supervisor for this project: Hansung Kim, Graduate Student
Hours: 6-8 hrs
Mathematical and Physical Sciences Engineering, Design & Technologies