N3P: Accelerated Automated Parking via a Learning-Based Naturalistic Three-Stage Scheme

2026-05-21

AI Summary

This work addresses the challenges in autonomous parking—namely, the high computational cost of path planning, the low reliability of reinforcement learning approaches, and the difficulty of satisfying long-horizon geometric constraints—by proposing a learned, three-stage naturalistic parking framework. The method decomposes complex parking maneuvers into simpler subproblems by introducing and predicting intermediate preparatory poses, and integrates seamlessly with the Hybrid A* algorithm to ensure high-quality trajectories while significantly improving planning efficiency. Experimental results demonstrate that, in both perpendicular and parallel parking scenarios, the proposed approach achieves over 80% faster planning speeds, higher success rates, shorter trajectories, and fewer gear shifts compared to reinforcement learning baselines, all while maintaining comparable or lower computation times.

Abstract

Autonomous parking requires efficient path planning that ensures kinematic feasibility and collision avoidance in constrained environments. Hybrid A* is widely used but computationally expensive, while reinforcement learning (RL) methods lack reliability and often struggle with long-horizon geometric constraints, leading to suboptimal trajectories. We present N3P, a fast learning-based three-stage framework for automated parking. By introducing an intermediate preparatory pose and using a learning module to predict it, N3P decomposes the maneuver into simpler subproblems, thereby reducing computational complexity and accelerating path generation. We validate the framework by integrating it with Hybrid A* algorithms. Experiments in perpendicular and parallel parking scenarios show that N3P-enhanced Hybrid A* speeds up planning by more than 80%. It also outperforms RL baselines in success rate and trajectory quality, producing shorter trajectories with fewer gear changes, while achieving comparable or lower planning time in most cases.

Problem

• autonomous parking
• path planning
• kinematic feasibility
• collision avoidance
• geometric constraints

Innovation

• learning-based planning
• three-stage framework
• intermediate preparatory pose
• automated parking
• Hybrid A* acceleration