Learned spatial risk fields improve autonomous vehicle planning safety
READ: Learning Risk-Informed Fields for End-to-End Autonomous Driving
Robotics
Summary
Driving by itself is tricky because a car needs to not only see what's around but also decide what to do next safely. The authors created a new system called READ that helps cars understand risks in their environment by making a special map that shows safer and riskier spots over time. This map lets the car plan its path better by aiming for safer areas, and it can adjust based on what’s actually happening on the road. Their tests showed that this approach helps the car choose safer routes compared to older methods.
What this means in practice
- •For automotive software engineers: Integrate learned risk fields into autonomous vehicle planners to improve trajectory safety by directly aligning routes with low-risk spatial regions.
- •For robotics system developers: Use risk-informed spatial fields to make safer motion plans for mobile robots navigating dynamic environments beyond driving.
Authors
Zhiyuan Liu, Yuanxin Tian, Zehong Ke, Jinhao Li, Hao Cheng, Zhenhua Xu, Wenhao Yu, Jianqiang Wang
Abstract
Autonomous driving requires more than recognizing what is present in a scene: a planner must determine how road structure, surrounding agents, and their motion states should influence a future maneuver. Existing learning-based planners can capture these influences through latent scene features and trajectory decoders, but the relationship between environmental factors and candidate actions often remains implicit. This limits the ability to inspect, diagnose, or refine how scene context affects the safety of a predicted trajectory. Classical safety fields provide an explicit spatial representation of this relationship, but their risk shapes and relative weights are prescribed in advance and do not adapt to each scene. We introduce READ, a framework that learns an explicit, planning-aligned risk representation from complementary geometric and behavioral constraints. READ instantiates this representation as a continuous spatiotemporal field, enabling differentiable queries along candidate trajectories. The learned field connects scene understanding with action selection by encouraging predicted trajectories to align with low-risk regions, while retaining a differentiable interface for trajectory evaluation and refinement. READ integrates with both end-to-end planners and Vision-Language-Action models. Experiments on NAVSIM show consistent gains across matched end-to-end backbones and strong performance in a VLA setting; READ also achieves competitive results on NAVSIM v2. These results establish learned spatial risk as an explicit, adaptable representation for safe planning.