Occupancy-aware Trajectory Planning for Autonomous Valet Parking in Uncertain Dynamic Environments

2025-09-11

AI Summary

To address inaccurate parking-slot state prediction, perceptual limitations, and planning–perception decoupling in autonomous valet parking under dynamic and uncertain environments, this paper proposes a unified probabilistic modeling and information-driven planning framework. Methodologically: (1) we formulate a slot occupancy estimation model that distinguishes initial occupied/free states, integrating finite-field-of-view observations with uncertainty propagation; (2) we introduce an information-gain-based exploration–exploitation strategy to enable wait-or-proceed decisions; and (3) we tightly couple dynamic obstacle motion forecasting with adaptive trajectory optimization. Evaluated on large-scale stochastic parking simulations, our approach achieves significant improvements over state-of-the-art methods: +12.7% parking success rate, 63% reduction in collision rate (enhancing safety margin), and 38% reduction in jerk (improving trajectory smoothness).

Technology Category

• Planning, Routing, and Scheduling: Model-Based Reasoning
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Intelligent Robots: State Estimation

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Accurately reasoning about future parking spot availability and integrated planning is critical for enabling safe and efficient autonomous valet parking in dynamic, uncertain environments. Unlike existing methods that rely solely on instantaneous observations or static assumptions, we present an approach that predicts future parking spot occupancy by explicitly distinguishing between initially vacant and occupied spots, and by leveraging the predicted motion of dynamic agents. We introduce a probabilistic spot occupancy estimator that incorporates partial and noisy observations within a limited Field-of-View (FoV) model and accounts for the evolving uncertainty of unobserved regions. Coupled with this, we design a strategy planner that adaptively balances goal-directed parking maneuvers with exploratory navigation based on information gain, and intelligently incorporates wait-and-go behaviors at promising spots. Through randomized simulations emulating large parking lots, we demonstrate that our framework significantly improves parking efficiency, safety margins, and trajectory smoothness compared to existing approaches.

Problem

• Predicting future parking spot availability in dynamic uncertain environments
• Integrating occupancy estimation with adaptive trajectory planning strategies
• Balancing goal-directed parking with exploratory navigation using information gain

Innovation

• Predicts future parking spot occupancy
• Probabilistic occupancy estimator with limited FoV
• Adaptive strategy balancing goal and exploration