Occupancy-aware Trajectory Planning for Autonomous Valet Parking in Uncertain Dynamic Environments
Farhad Nawaz Savvas Sadiq Ali, Faizan M. Tariq, Sangjae Bae, David Isele, Avinash Singh, Nadia Figueroa, Nikolai Matni, Jovin D'sa
AI summary
Problem
Existing autonomous valet parking planners rely on static spot availability or ignore dynamic agent impacts, limiting foresight and adaptability in real-world lots where spot status constantly changes.
Approach
A framework that uses a Bayes filter to predict future spot occupancy by modeling dynamic agent motion and distinguishing initially vacant from occupied spots, coupled with a strategy planner that balances direct parking, waiting, and exploration based on occupancy probabilities.
Key results
- Distance-aware partial field-of-view observation model degrading confidence with range
- Bayes filter estimator predicting spot occupancy using asymmetric arrival and departure dynamics
- Strategy planner integrating occupancy forecasts with wait-and-go and exploration behaviors
- Simulation results showing reduced parking time, shorter travel distance, and smoother trajectories compared to baselines
Why it matters
Enables safer and more efficient autonomous parking in real-world, infrastructure-free environments without relying on perfect sensor data or static assumptions.
Abstract
Autonomous Valet Parking (AVP) requires plan- ning under partial observability, where parking spot availabil- ity evolves as dynamic agents enter and exit spots. Existing approaches either rely only on instantaneous spot availability or make static assumptions, thereby limiting foresight and adaptability. We propose an approach that estimates probability of future spot occupancy by distinguishing initially vacant and occupied spots while leveraging nearby dynamic agent motion. We propose a probabilistic estimator that integrates partial, noisy observations from a limited Field-of-View, with the evolving uncertainty of unobserved spots. Coupled with the estimator, we design a strategy planner that balances goal- directed parking maneuvers with exploratory navigation based on information gain, and incorporates wait-and-go behaviors at promising spots. Through randomized simulations emulating large parking lots, we demonstrate that our framework signif- icantly improves parking efficiency and trajectory smoothness over existing approaches, while maintaining safety margins. Simulation videos: https://sites.google.com/view/avp-hri/home.