A Diffusion-Refined Planner with Reinforcement Learning Priors for Confined-Space Parking

2025-10-15

AI Summary

To address inaccurate action modeling and low planning success rates in automatic parking within narrow spaces, this paper proposes a diffusion optimization framework integrated with reinforcement learning (RL) priors. Specifically, the method leverages the action distribution generated by a pre-trained RL policy as a prior to guide the denoising process of a diffusion model, enabling progressive refinement of action sequences in an end-to-end co-tuning manner. This work is the first to incorporate RL-derived priors into diffusion models for action sequence refinement, significantly improving modeling accuracy and robustness in extremely confined scenarios. Experiments demonstrate that the proposed method achieves state-of-the-art planning success rates across diverse constrained parking settings, reduces inference steps substantially, and maintains strong generalization performance in standard scenarios—thereby validating its effectiveness and practical applicability.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Search and Optimization: Learning to Search
• Planning, Routing, and Scheduling: Planning with Language Models

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

The growing demand for parking has increased the need for automated parking planning methods that can operate reliably in confined spaces. In restricted and complex environments, high-precision maneuvers are required to achieve a high success rate in planning, yet existing approaches often rely on explicit action modeling, which faces challenges when accurately modeling the optimal action distribution. In this paper, we propose DRIP, a diffusion-refined planner anchored in reinforcement learning (RL) prior action distribution, in which an RL-pretrained policy provides prior action distributions to regularize the diffusion training process. During the inference phase the denoising process refines these coarse priors into more precise action distributions. By steering the denoising trajectory through the reinforcement learning prior distribution during training, the diffusion model inherits a well-informed initialization, resulting in more accurate action modeling, a higher planning success rate, and reduced inference steps. We evaluate our approach across parking scenarios with varying degrees of spatial constraints. Experimental results demonstrate that our method significantly improves planning performance in confined-space parking environments while maintaining strong generalization in common scenarios.

Problem

• Developing automated parking planning for confined spaces with high precision maneuvers
• Addressing challenges in modeling optimal action distributions using explicit methods
• Improving planning success rates in spatially constrained parking environments

Innovation

• Combines diffusion models with reinforcement learning priors
• Refines coarse priors into precise action distributions
• Enhances planning accuracy in confined parking spaces