MC-Risk: Multi-Component Risk Fields for Risk Identification and Motion Planning

2026-05-20

AI Summary

This work proposes an early, calibrated, and class-aware risk identification and motion planning framework for autonomous driving. It constructs a multi-component risk field over a bird’s-eye-view occupancy grid by linearly fusing three interpretable modules: a maneuvering traffic agent field, a vulnerable road user risk field, and a road penalty field, enabling precise risk modeling. The approach innovatively integrates multimodal trajectory prediction, Gaussian torus structures, forward-biased anisotropic kernels, and high-definition map topology, while supporting a plug-and-play planning interface. Evaluated on the RiskBench collision subset, the method achieves state-of-the-art risk localization accuracy and provides the earliest hazard warnings, and it can directly generate risk-aware trajectories without requiring additional training.

Abstract

We present MC-Risk, a planner-aligned, multi-component risk field on a bird's-eye-view grid that yields early, calibrated, and class-aware risk localization. MC-Risk linearly composes three interpretable modules: (i) a motorized-agent field that fuses a black-box multimodal trajectory predictor with an analytic Gaussian-torus construction whose lateral width grows with speed/curvature and whose height attenuates with look-ahead; (ii) a VRU risk field that replaces isotropic pedestrian blobs with a forward-biased anisotropic kernel aligned to heading and speed; and (iii) a road penalty field that exploits full HD-map topology, imposing an off-road penalty and lane-aware risk exposure for same/opposite directions. We conduct, to our knowledge, the first standardized quantitative evaluation of a risk-field formulation on RiskBench's collision subset. MC-Risk attains the best overall risk localization and the earliest hazard indication. Finally, we demonstrate a plug-and-play planning interface by using the field as an MPC cost density, enabling risk-aware trajectory generation without additional training.

Problem

• risk identification
• motion planning
• multi-component risk field
• autonomous driving
• risk-aware planning

Innovation

• multi-component risk field
• risk-aware motion planning
• anisotropic VRU modeling
• planner-aligned risk representation
• MPC cost density