Lupien, Jean-Luc (UC Berkeley), Alhadlaq, Abdullah (University of California, Berkeley), Tang, Yuhan (University of California, Berkeley), Wu, Yan (University of California, Berkeley), Chen, Jiayu (UC Berkeley), Long, Yutan (University of California, Berkeley), Jiang, Xuan (University Of California, Berkeley)

Entropy-Based Dynamic Programming for Efficient Vehicle Parking

Scheduled for presentation during the Poster Session "Modeling and Optimization of Mobility and Transport Systems " (ThBT14), Thursday, September 26, 2024, 14:30−16:30, Foyer

2024 IEEE 27th International Conference on Intelligent Transportation Systems (ITSC), September 24- 27, 2024, Edmonton, Canada

This information is tentative and subject to change. Compiled on October 14, 2024

Abstract

In urban environments, parking has proven to be a significant source of congestion and inefficiency. In this study, we propose a methodology that offers a systematic solution to minimize the time spent by drivers in finding parking spaces. Drawing inspiration from statistical mechanics, we utilize an entropy model to predict the distribution of available parking spots across different levels of a multi-story parking garage, encoded by a single parameter: temperature. Building on this model, we develop a dynamic programming framework that guides vehicles to the optimal floor based on the predicted occupancy distribution. This approach culminates in our Temperature-Informed Parking Policy (TIPP), which not only predicts parking spot availability but also dynamically adjusts parking assignments in real-time to optimize vehicle placement and reduce search times. We compare TIPP with simpler policies and the theoretical optimal solution to demonstrate its effectiveness and gauge how closely it approaches the ideal parking strategy. The results highlight the potential of integrating TIPP in real-world applications, paving the way for smarter, more efficient urban landscapes.