Multiagent Reinforcement Learning for Autonomous Routing and Pickup Problem with Adaptation to Variable Demand

TL;DR

This paper presents a reinforcement learning framework for autonomous vehicle routing that adapts to variable urban demand, improving coordination, non-myopic planning, and responsiveness to demand fluctuations, demonstrated on real taxi data.

Contribution

It introduces an adaptive multiagent RL approach combining online and offline methods to handle demand variability using Wasserstein ambiguity sets.

Findings

Outperforms classical routing methods in simulations.

Successfully adapts to demand fluctuations in real-world data.

Reduces wait times through coordinated, non-myopic policies.

Abstract

We derive a learning framework to generate routing/pickup policies for a fleet of autonomous vehicles tasked with servicing stochastically appearing requests on a city map. We focus on policies that 1) give rise to coordination amongst the vehicles, thereby reducing wait times for servicing requests, 2) are non-myopic, and consider a-priori potential future requests, 3) can adapt to changes in the underlying demand distribution. Specifically, we are interested in policies that are adaptive to fluctuations of actual demand conditions in urban environments, such as on-peak vs. off-peak hours. We achieve this through a combination of (i) an online play algorithm that improves the performance of an offline-trained policy, and (ii) an offline approximation scheme that allows for adapting to changes in the underlying demand model. In particular, we achieve adaptivity of our learned policy to different demand distributions by quantifying a region of validity using the q-valid radius of a Wasserstein Ambiguity Set. We propose a mechanism for switching the originally trained offline approximation when the current demand is outside the original validity region. In this case, we propose to use an offline architecture, trained on a historical demand model that is closer to the current demand in terms of Wasserstein distance. We learn routing and pickup policies over real taxicab requests in San Francisco with high variability between on-peak and off-peak hours, demonstrating the ability of our method to adapt to real fluctuation in demand distributions. Our numerical results demonstrate that our method outperforms alternative rollout-based reinforcement learning schemes, as well as other classical methods from operations research.