IntentNet: Learning to Predict Intention from Raw Sensor Data

Abstract

In order to plan a safe maneuver, self-driving vehicles need to understand the intent of other traffic participants. We define intent as a combination of discrete high-level behaviors as well as continuous trajectories describing future motion. In this paper, we develop a one-stage detector and forecaster that exploits both 3D point clouds produced by a LiDAR sensor as well as dynamic maps of the environment. Our multi-task model achieves better accuracy than the respective separate modules while saving computation, which is critical to reducing reaction time in self-driving applications.

Cite

Text

Casas et al. "IntentNet: Learning to Predict Intention from Raw Sensor Data." Conference on Robot Learning, 2018.

Markdown

[Casas et al. "IntentNet: Learning to Predict Intention from Raw Sensor Data." Conference on Robot Learning, 2018.](https://mlanthology.org/corl/2018/casas2018corl-intentnet/)

BibTeX

@inproceedings{casas2018corl-intentnet,
  title     = {{IntentNet: Learning to Predict Intention from Raw Sensor Data}},
  author    = {Casas, Sergio and Luo, Wenjie and Urtasun, Raquel},
  booktitle = {Conference on Robot Learning},
  year      = {2018},
  pages     = {947-956},
  url       = {https://mlanthology.org/corl/2018/casas2018corl-intentnet/}
}