Physics-Aware Safety-Assured Design of Hierarchical Neural Network based Planner
This work addresses safety assurance for neural network planners in safety-critical autonomous driving, representing an incremental improvement with a focus on verifiability.
The authors tackled the challenge of formally ensuring safety for neural network planners in autonomous driving by proposing a hierarchical planner that analyzes physical scenarios and uses scenario-specific strategies, and they developed an efficient verification method with overapproximation and partition techniques, empirically showing effectiveness in tasks like unprotected left turns and highway merging.
Neural networks have shown great promises in planning, control, and general decision making for learning-enabled cyber-physical systems (LE-CPSs), especially in improving performance under complex scenarios. However, it is very challenging to formally analyze the behavior of neural network based planners for ensuring system safety, which significantly impedes their applications in safety-critical domains such as autonomous driving. In this work, we propose a hierarchical neural network based planner that analyzes the underlying physical scenarios of the system and learns a system-level behavior planning scheme with multiple scenario-specific motion-planning strategies. We then develop an efficient verification method that incorporates overapproximation of the system state reachable set and novel partition and union techniques for formally ensuring system safety under our physics-aware planner. With theoretical analysis, we show that considering the different physical scenarios and building a hierarchical planner based on such analysis may improve system safety and verifiability. We also empirically demonstrate the effectiveness of our approach and its advantage over other baselines in practical case studies of unprotected left turn and highway merging, two common challenging safety-critical tasks in autonomous driving.