Authors: Tung Phan-Minh, Elena Corina Grigore, Freddy A. Boulton, Oscar Beijbom, Eric M. Wolff
Published: April 1, 2020
Summary: Previous work has employed a variety of methods, including multimodal regression, occupancy maps, and 1-step stochastic policies. We instead frame the trajectory prediction problem as classification over a diverse set of trajectories. By dynamically generating trajectory sets based on the agent's current state, we can further improve our method's efficiency. We demonstrate our approach on public, real-world self-driving datasets, and show that it outperforms state-of-the-art methods.