SAFE-Q: Safety-Aware End-to-End Driving Using CrossQ Deep Reinforcement Learning

초록

This work presents SAFE-Q, a safety-aware endto- end (E2E) autonomous driving network using CrossQ deep reinforcement learning (DRL) designed for reliable performance under long-tail distribution (LTD) conditions. The architecture integrates a multitask learning (MTL) perception module, a safety reasoning-based local path waypoint (LPW) generator, and a DRL-based action planning and control (APC) module. The perception network extracts bird's eye view (BEV) and perspective view (PV) features from multicamera inputs, while the safety reasoning module refines global path waypoints (GPWs) to ensure rule compliance and collision avoidance. The DRL module fuses perception and waypoint information within an affordance state vector to produce optimal throttle and brake commands. Experiments in the CARLA simulator reproduced LTD scenarios to evaluate SAFE-Q's driving performance. SAFE-Q outperformed baseline methods, achieving gains of 12.6% in driving score (DS), 10.1% in route completion, 2.4% in infraction penalty, 27.9% in Euclidean distance (ED), 55.3% in average speed (AS), and 10% in Jerk. These results confirm SAFE-Q's safe, efficient, and robust control capability in complex urban LTD environments.

키워드