Physics-Informed Neural Optimal Control for Precision Immobilization Technique in Emergency Scenarios

TL;DR

This paper introduces a physics-informed neural control framework for automating the Precision Immobilization Technique in emergency vehicle scenarios, improving success rates and computational efficiency.

Contribution

It develops a novel PicoPINN surrogate model and hierarchical neural optimal control architecture tailored for real-time, safety-critical vehicle immobilization maneuvers.

Findings

Adding the planning layer increases PIT success rate from 63.8% to 76.7%.

PicoPINN reduces neural network parameters from 8965 to 812.

Scaled vehicle tests demonstrate control feasibility with 3 out of 4 successful low-speed PIT trials.

Abstract

Precision Immobilization Technique (PIT) is a potentially effective intervention maneuver for emergency out-of-control vehicle, but its automation is challenged by highly nonlinear collision dynamics, strict safety constraints, and real-time computation requirements. This work presents a PIT-oriented neural optimal-control framework built around PicoPINN (Planning-Informed Compact Physics-Informed Neural Network), a compact physics-informed surrogate obtained through knowledge distillation, hierarchical parameter clustering, and relation-matrix-based parameter reconstruction. A hierarchical neural-OCP (Optimal Control Problem) architecture is then developed, in which an upper virtual decision layer generates PIT decision packages under scenario constraints and a lower coupled-MPC (Model Predictive Control) layer executes interaction-aware control. To evaluate the framework, we construct a PIT Scenario Dataset and conduct surrogate-model comparison, planning-structure ablation, and multi-fidelity assessment from simulation to scaled by-wire vehicle tests. In simulation, adding the upper planning layer improves PIT success rate from 63.8% to 76.7%, and PicoPINN reduces the original PINN parameter count from 8965 to 812 and achieves the smallest average heading error among the learned surrogates (0.112 rad). Scaled vehicle experiments are further used as evidence of control feasibility, with 3 of 4 low-speed controllable-contact PIT trials achieving successful yaw reversal.