Using reinforcement learning to minimize taxi idle times

2019/10/25 by Kevin O’Keeffe, Kevin O'Keeffe, Sam Anklesaria +6
Decision Sciences · Engineering · Physics and Astronomy · Social Sciences · #Auction Theory and Applications #Transportation Planning and Optimization #Transportation and Mobility Innovations #physics.soc-ph

paper · pdf · doi:10.48550/arxiv.1910.11918

arxiv created 2019/10/25 · arxiv updated 2019/10/29

Abstract

Taxis spend a significant amount of time idle, searching for passengers. The routes vacant taxis should follow in order to minimize their idle times are hard to calculate; they depend on complex quantities like passenger demand, traffic conditions, and inter-taxi competition. Here we explore if reinforcement learning (RL) can be used for this purpose. Using real-world data to characterize passenger demand, we show RL-taxis indeed learn to how to reduce their idle time in many environments. In particular, a single RL-taxi operating in a population of regular taxis learns to out-perform its rivals by a significant margin.

Related

• Optimizing Taxi Carpool Policies via Reinforcement Learning and Spatio-Temporal Mining
• Where to go: Agent Guidance with Deep Reinforcement Learning in A City-Scale Online Ride-Hailing Service
• Efficient Ridesharing Dispatch Using Multi-Agent Reinforcement Learning
• Equilibrium Inverse Reinforcement Learning for Ride-hailing Vehicle Network
• Deep Reinforcement Learning for Shared Autonomous Vehicles (SAV) Fleet Management