Atomic Proximal Policy Optimization for Electric Robo-Taxi Dispatch and Charger Allocation

Jim Dai, Manxi Wu, Zhanhao Zhang · 2025

Read Atomic Proximal Policy Optimization for Electric Robo-Taxi Dispatch and Charger Allocation by Jim Dai, Manxi Wu, Zhanhao Zhang (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2502-13392.txt.

Description

Dai, Wu, and Zhang studied how to run fleets of electric robo-taxis in twenty twenty-five. Companies such as Waymo already run such services in some United States cities. Operators must match rides, move empty cars, and charge batteries, all with limited chargers and uncertain demand.

The team modeled fleet operations as a step-by-step decision process. Their learning algorithm, Atomic-PPO, breaks big choices into small atomic actions, cutting training cost. They tuned a simulation on real New York City trip records. The algorithm produced near-best dispatch policies and beat rule-based methods. Tests showed charger placement should follow ridership, and that charger count, charging speed, and fleet size should be planned together.

Catalog record

CREATORJim Dai, Manxi Wu, Zhanhao Zhang
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE113 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2502-13392

Browse the full library · About the Tower of Babel