Paper accepted to IEEE Intelligent Vehicles Symposium (IV) 2023
A paper titled “Physics Constrained Motion Prediction with Uncertainty Quantification” was accepted to the IEEE Intelligent Vehicles Symposium (IV) 2023. The paper develops a novel method for predicting the motion of a vehicle by integrating a machine learning model (learned from past data of the vehicle) and a physics-based model of the vehicle dynamics. The method is shown to achieve higher accuracy and physical consistency than other pure data-driven methods. Furthermore, an uncertainty quantification method is proposed to assess the uncertainty of the prediction provided by the model. This is a collaboration with Prof. Rahul Mangharam’s team at the University of Pennsylvania. Congratulations to Prof. Truong Nghiem and all co-authors.