Perceive, Predict, and Plan: Safe Motion Planning Through Interpretable Semantic Representations
Sadat, Abbas, Casas, Sergio, Ren, Mengye, Wu, Xinyu, Dhawan, Pranaab, Urtasun, Raquel
–arXiv.org Artificial Intelligence
In this paper we propose a novel end-to-end learnable network that performs joint perception, prediction and motion planning for self-driving vehicles and produces interpretable intermediate representations. Unlike existing neural motion planners, our motion planning costs are consistent with our perception and prediction estimates. This is achieved by a novel differentiable semantic occupancy representation that is explicitly used as cost by the motion planning process. Our network is learned end-to-end from human demonstrations. The experiments in a large-scale manual-driving dataset and closed-loop simulation show that the proposed model significantly outperforms state-of-the-art planners in imitating the human behaviors while producing much safer trajectories.
arXiv.org Artificial Intelligence
Aug-13-2020
- Country:
- North America > Canada > Ontario > Toronto (0.14)
- Genre:
- Research Report (0.50)
- Industry:
- Transportation > Ground > Road (0.94)
- Technology: