English

RAP: Risk-Aware Prediction for Robust Planning

Machine Learning 2023-01-13 v2 Robotics

Abstract

Robust planning in interactive scenarios requires predicting the uncertain future to make risk-aware decisions. Unfortunately, due to long-tail safety-critical events, the risk is often under-estimated by finite-sampling approximations of probabilistic motion forecasts. This can lead to overconfident and unsafe robot behavior, even with robust planners. Instead of assuming full prediction coverage that robust planners require, we propose to make prediction itself risk-aware. We introduce a new prediction objective to learn a risk-biased distribution over trajectories, so that risk evaluation simplifies to an expected cost estimation under this biased distribution. This reduces the sample complexity of the risk estimation during online planning, which is needed for safe real-time performance. Evaluation results in a didactic simulation environment and on a real-world dataset demonstrate the effectiveness of our approach. The code and a demo are available.

Keywords

Cite

@article{arxiv.2210.01368,
  title  = {RAP: Risk-Aware Prediction for Robust Planning},
  author = {Haruki Nishimura and Jean Mercat and Blake Wulfe and Rowan McAllister and Adrien Gaidon},
  journal= {arXiv preprint arXiv:2210.01368},
  year   = {2023}
}

Comments

22 pages, 14 figures, 3 tables. First two authors contributed equally. Conference on Robot Learning (CoRL) 2022 (oral)

R2 v1 2026-06-28T02:44:43.172Z