RAP: Risk-Aware Prediction for Robust Planning
This work addresses safety-critical planning for robots in interactive environments, representing an incremental improvement by focusing on risk-aware prediction to enhance existing robust planners.
The paper tackles the problem of robust planning in interactive scenarios by addressing the underestimation of risk due to finite-sampling approximations in motion forecasts, which can lead to unsafe robot behavior. It proposes a risk-aware prediction method that learns a risk-biased distribution over trajectories, reducing sample complexity for risk estimation during online planning, with evaluation showing effectiveness in simulation and on a real-world dataset.
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.