RL-OGM-Parking: Lidar OGM-Based Hybrid Reinforcement Learning Planner for Autonomous Parking

2025-02-26

AI Summary

To address the core challenges in automated parking within confined and complex environments—namely, poor generalizability of rule-based methods, insufficient stability of learning-based approaches, and difficulty in sim-to-real transfer—this paper proposes a LiDAR-based Occupancy Grid Map (OGM)-driven dual-modality cooperative planning framework integrating rule-based and learning-based paradigms. The framework synergistically combines a Reeds-Shepp analytical path planner with a DDPG-based reinforcement learning planner, coordinated via a dynamic hybrid scheduling mechanism. It introduces, for the first time, joint modeling of real-time OGM perception and dual-modality decision-making, effectively mitigating representation and dynamics gaps in sim-to-real transfer. Experimental results demonstrate a 98.7% parking success rate on real vehicles—improving upon pure rule-based and pure learning-based baselines by 12.4% and 8.9%, respectively—while reducing average parking time by 23% and achieving positioning accuracy of ±0.05 m.

Technology Category

• Planning, Routing, and Scheduling: Learning for Planning and Scheduling
• Multiagent Systems: Multiagent Planning
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Autonomous parking has become a critical application in automatic driving research and development. Parking operations often suffer from limited space and complex environments, requiring accurate perception and precise maneuvering. Traditional rule-based parking algorithms struggle to adapt to diverse and unpredictable conditions, while learning-based algorithms lack consistent and stable performance in various scenarios. Therefore, a hybrid approach is necessary that combines the stability of rule-based methods and the generalizability of learning-based methods. Recently, reinforcement learning (RL) based policy has shown robust capability in planning tasks. However, the simulation-to-reality (sim-to-real) transfer gap seriously blocks the real-world deployment. To address these problems, we employ a hybrid policy, consisting of a rule-based Reeds-Shepp (RS) planner and a learning-based reinforcement learning (RL) planner. A real-time LiDAR-based Occupancy Grid Map (OGM) representation is adopted to bridge the sim-to-real gap, leading the hybrid policy can be applied to real-world systems seamlessly. We conducted extensive experiments both in the simulation environment and real-world scenarios, and the result demonstrates that the proposed method outperforms pure rule-based and learning-based methods. The real-world experiment further validates the feasibility and efficiency of the proposed method.

Problem

• Autonomous parking challenges
• Hybrid reinforcement learning solution
• LiDAR-based sim-to-real transfer

Innovation

• Hybrid RL and rule-based planner
• LiDAR OGM for sim-to-real transition
• Enhanced autonomous parking performance